INCEPTION REPORT // DIGISERINCEPTION REPORT // DIGISER ESPON // espon.eu 5 List of maps, figures,...

INCEPTION REPORT //

DIGISER Digital innovation in governance and public service provision Inception Report // January 2021

This Inception REPORT is conducted within the framework of the ESPON 2020 Cooperation Programme, partly financed by the European Regional Development Fund.

The ESPON EGTC is the Single Beneficiary of the ESPON 2020 Cooperation Programme. The Single Operation within the programme is implemented by the ESPON EGTC and co-financed by the European Regional Development Fund, the EU Member States and the Partner States, Iceland, Liechtenstein, Norway and Switzerland.

This delivery does not necessarily reflect the opinions of members of the ESPON 2020 Monitoring Committee.

Coordination: ESPON EGTC: Martin Gauk, Caroline Clause

Authors OASC: Davor Meersman, Lea Hemetsberger, Hugo Kerschot Polimi: Irene Bianchi, Grazia Concilio, Francesco Fagiani, Matteo Fontana, Ilaria Mariani, Federico Rita, Michelangelo Secchi CPC: Isaac Sserwanja, Bin Guan, Reza Akhavan, Thanos Bantis Deloitte: Diogo Santos, Ana Vaz Raposo, Pedro Vicente, Jean Barroca, Gulam Sidique, Andreas Steinbach, Barnabas Sandor

Advisory group Kadri Jushkin (Ministry of Finance Estonia), Christina Lohfert Rolandsen (Danish Business Authority), Akim Oural (Lille Metropole), Paulo Calçada (Porto Digital), Markku Markkula (Helsinki-Uusimaa Region), Lodewijk Noordzij (Eurocities), Wim De Kinderen (ENoLL), Olli Voutilainen (Finnish Ministry of Economic Affairs and Employment), Tanguy Coenen (imec), Martin Brynskov (Aarhus University), Gianluca Misuraca (Krems University), Serge Novaretti and Stefanos Kotoglou (EC-DG Connect), Dana Eleftheriadou and Natalia Gkiaouri (EC -DG Grow), Anke Schuster and Bert Kuby (Committee of the Regions), Paresa Markianidou and Morten Rasmussen (Technopolis Group)

Information on ESPON and its projects can be found at www.espon.eu. The website provides the possibility to download and examine the most recent documents produced by finalised and ongoing ESPON projects.

© ESPON, 2020 Printed on paper produced environmentally friendly

Layout and graphic design by BGRAPHIC, Denmark

Printing, reproduction or quotation is authorised provided the source is acknowledged and a copy is forwarded to the ESPON EGTC in Luxembourg.

Contact: [email protected]

INCEPTION REPORT // DIGISER

ESPON // espon.eu 3

INCEPTION REPORT //

DIGISER Digital innovation in governance and public service provision Inception Report // January 2021

Disclaimer

This document is a inception report.

The information contained herein is subject to change and does not commit the ESPON EGTC and the countries participating in the ESPON 2020 Cooperation Programme.

The final version of the report will be published as soon as approved.


4 ESPON // espon.eu

Table of contents

Abbreviations ...................................................................................................................................... 6 Foreword 7 1 Introduction ........................................................................................................................ 8 1.1 Scope and aims of DIGISER ........................................................................................................ 8 1.1 Overview of the work done ........................................................................................................... 9 1.1.1 Task 1: Define and implement a participatory plan ...................................................................... 9 1.1.2 Task 2: Formulate the research framework and collect data ..................................................... 10 1.2 Structure of the report ................................................................................................................ 11 2 Data gap analysis ............................................................................................................ 13 2.1 Towards the data review ............................................................................................................ 13 2.2 Data review methodology ........................................................................................................... 13 2.3 The data review .......................................................................................................................... 15 2.3.1 Relevant project studies and indicators ..................................................................................... 15 2.3.2 Detailed studies and networks ................................................................................................... 16 2.3.3 Large-scale statistical datasets .................................................................................................. 17 2.3.4 Open data by EU bodies ............................................................................................................ 17 2.4 Summary of results .................................................................................................................... 18 3 The DIGISER Conceptual Framework ............................................................................ 19 3.1 The Digital Public Service Value Index: a transitional perspective ............................................ 19 3.2 Inside the concept ...................................................................................................................... 22 3.2.1 Digital Service Innovation Maturity ............................................................................................. 24 3.2.2 Proneness to change ................................................................................................................. 25 3.2.2.1 Data management ...................................................................................................................... 26 3.2.2.2 Procurement ............................................................................................................................... 27 3.2.2.3 Societal engagement ................................................................................................................. 28 3.2.2.4 Institutional capacity building ..................................................................................................... 28 3.2.3 Orientation to mission ................................................................................................................ 29 4 The DIGISER data model ................................................................................................ 31 4.1 Scenarios of data collection between the data gap and the conceptual framework .................. 31 4.2 The overview of the Conceptual Data Model ............................................................................. 31 4.3 Logical description of the Data Model ........................................................................................ 34 4.4 Definition of the sample of analysis ............................................................................................ 35 4.4.1 Large sample ............................................................................................................................. 37 4.4.2 Case Studies .............................................................................................................................. 39 5 Future Work ..................................................................................................................... 40 5.1 Survey development, testing and deployment ........................................................................... 40 5.2 Survey distribution and engagement strategy ............................................................................ 40 5.2.1 Large scale engagement strategy .............................................................................................. 40 5.2.2 Gamification as strategy to keep survey users engaged ........................................................... 42 5.3 Data analysis and visualisation methods: geospatial visualisation and scores visualisation ..... 43 5.4 10 Case Studies - Overview ....................................................................................................... 43 5.5 Developing policy recommendations ......................................................................................... 45 5.6 Concluding remarks ................................................................................................................... 45 6 Annexes ............................................................................................................................ 46 References ........................................................................................................................................ 47


ESPON // espon.eu 5

List of maps, figures, charts and tables

List of maps Map 4.1 Cities belonging to relevant networks 37 Map 4.2 Large sample of cities 38

List of figures Figure 1.1 Work carried out according to DIGISER work plan 9 Figure 1.2 SAG engagement process 10 Figure 2.1 Data sources and coverage 13 Figure 2.2 Details on datasets: project on study source, topic, scale, accessibility, format 15 Figure 2.3 Eurostat dataset raw analysis (keyword-based) displaying datasets relevant at different

scales and keyword occurrence 16 Figure 3.1 Triple-Loop Learning and the “what, how, and why” questions 20 Figure 3.2 Conceptual structure of the Digital Public Service Value Index 20 Figure 3.3 Detailed DIGISER conceptual triplet 23 Figure 4.1 Innovation Governance vs Digital Service Innovation Maturity 31 Figure 4.2 Circular Dendrogram, fusing the process and the service perspective 32 Figure 4.3 The flow of data: from process typology to specific data 33 Figure 5.1 The stepwise strategy 40

List of tables Table 1.1 Content of the inception report 11 Table 3.1 Mapping scaling mechanisms over the DIGISER research questions 21 Table 5.1 Potential risks 41


6 ESPON // espon.eu

Abbreviations

Abbreviation Full Name

API Application Programming Interface

DESI Digital Economy and Society Index

DPSVI Digital Public Service Value Index

EAB European Advisory Board

EDCI European Digital City Index

EIF European Interoperability Framework

ESPON European Spatial Planning Observation Network

EU European Union

EU ODP European Union Open Data Portal

FUA Functional Urban Areas

GDC Green Digital Charter

GDPR General Data Protection Regulation

ICC Intelligent Cities Challenge

ICT Information and Communications Technology

KPI Key Performance Indicator

LAU Local Administrative Units

LEA Learning Technology Accelerator

NUTS Nomenclature of Territorial Units for Statistics

OASC Open and Agile Smart Cities

OECD Organisation for Economic Co-operation and Development

OGD Open Government Data

PA Public Administration

PCP Pre-Commercial Procurement

R&D Research and Development

SAG Scientific Advisory Group

SDGs Sustainable Development Goals

SEM Structural Equation Modelling

T-LL Triple-Loop Learning

ToR Terms of Reference

UNDP United Nations Development Programme


ESPON // espon.eu 7

Foreword

The digital urban and regional landscape in Europe is undergoing significant and profoundly consequential changes under the impetus of a sweeping convergence of critical actors and initiatives, with DIGISER as an unequivocal constituent enjoying growing relevance, consideration, and potential. The ESPON vision and objectives to which DIGISER contributes, and the digital transformation of European societies to the core of their innermost governance levels, are two endeavours that have a synergistic and mutually reinforcing re-lation that is increasingly becoming apparent and recognised by stakeholders who can act at scale.

This report comes at an exciting time of acceleration of change, increase in commitment, and materialisation of diligently prepared, launched, and executed strategic capabilities in the immediate landscape. This Janu-ary, Open & Agile Smart Cities (OASC) held its annual event, the CITYxCITY Festival, with an audience of more than 3000 people - more than 5 times larger than previous editions. The CITYxCITY Festival brought together representatives of the Portuguese and Slovenian EU presidencies, city and regional representa-tives, the European Commission and Parliament, global corporations, and research powerhouses, among other participants, who unfailingly expressed commitment to improving the state, accelerating the pace, and increasing the accessibility of the digital transformation of European cities and communities.

The OASC platform allows us to put these commitments into concrete actions to the benefit of DIGISER and its linked objectives. Without intent of being exhaustive, such actions include supporting the integration of digital transformation monitoring in the rural agendas of upcoming presidencies, exploring and explicating its role in the climate agenda by forging strategic partnerships with key players in that realm, as well as taking the first steps towards sustainability of outcomes by preparing branded interfaces with the OASC and CITYxCITY platforms that can be extended beyond the project.

Considering the above, the importance of this inception report can hardly be overstated. It presents a com-prehensive overview of the initial stages of the project, the intellectual foundations, as well as the operational outlook for the activities that are planned, to a degree that should allow evaluators and advisors to properly consider, ascertain, and hopefully appreciate the quality of the work.

To both tackle the complexity of the subject and hit the ambition level of the challenge, under challenging societal conditions, on time and at an agreeable level of excellence, was made possible by the inspiring, constructive, and pleasant interactions with the ESPON team and the Strategic Advisory Group over the past months. I would like to take the opportunity to thank Martin Gauk in particular for his passionate contri-bution to the success and strategic value of the project.

I would also like to thank the project team for the effective, concerted, and targeted collaboration. I would like to especially express my utmost appreciation to the rock-star team of Politecnico Di Milano, notably Irene Bianchi, Grazia Concillio, Francesco Fagiani, Matteo Fontana, Ilaria Mariani, Federico Rita, Michelangelo Secchi, Marco Brambilla, Francesca Rizzo, and Simone Vantini, for your Herculean effort which exceeded expectations of the other partners, and hopefully also those of the DIGISER reviewers, advisors, and other project stakeholders.

Dr. Davor Meersman CEO Open & Agile Smart Cities


8 ESPON // espon.eu

1 Introduction

1.1 Scope and aims of DIGISER

The digital transformation of European industry, societies, and governments is profoundly reconfiguring the ways in which services are provided and knowledge and value are generated. Digital transformation has been a key driver of public sector innovation and service provision in recent decades. Digital technologies and capabilities create opportunities to re-organise public service innovation and delivery in ways that reduce cost and increase quality, proactiveness and citizen-centricity. Many governments on all levels, with a nota-ble increase in local level transformation in recent years, have launched digital initiatives from pre-existing operational baselines of ‘brick and mortar’ administrations. By doing that the public sector is often running into challenges of siloed infrastructures; lack of interoperability; lack of data sharing across boundaries; competence and skills gaps resulting in unfavourable risk profiles; mismatched funding and procurement programmes; and a ‘do-first, govern later’ approach that has occasionally resulted in lock-ins or unpleasant discoveries of legal impotence to act in the public interest due to legacy ‘blind spots’.

The process of getting a societal grip on digital transformation and striking a balance between oxygen for innovation and the protection of European rights and values has certainly made strides under the leadership of the European Commission, particularly in recent years, and notably with the recent Digital Markets and Digital Services Acts, both representing a culmination of sustained collaboration and broad consultations across the communities and stakeholder groups of Europe. The von der Leyen Commission has now also put forward the Digital Compass providing a vision to ensure Europe’s leadership in digital transformation1.

And whilst progress is certainly being made, many challenges remain for local public administrations. Public institutions are not fundamentally reshaping their own way of working, often leading to a multiplication of particular ‘digital’ profiles at various levels. Data sharing between governmental levels is still often limited in scope, and even in the eye of increasing public distrust in institutions, the ‘transparency switch’ is insuffi-ciently used out of a variety of considerations. Although many public administrations are increasingly finding their role as a digital ‘platform of trust’ for their citizens and legal persons, most are still form-finding the translation of that capability into the support and orchestration of data and service ecosystems as engines for economic growth. But perhaps the most pertinent area of attention, and one the current pandemic has certainly re-emphasised, is the slow and reluctant adoption of the vast potential of shared and interoperable digital tools and real-time data as capabilities for swift and decisive policy making. Intelligence for policy makers on all levels, but particularly local levels, is in need of an upgrade.

While large-scale statistics bodies (e.g. Eurostat, OECD) have started to measure the impact and govern-ance of digital transformation on a national level, data on local levels is scarce. Existing initiatives are there-fore not able to measure the complex relation between digital transformation and its impact on innovative governance and public services delivered by local and regional public administrations.

DIGISER aims to close this gap, working alongside European partners in a variety of initiatives such as “Join, Boost, Sustain” and the LORDI initiative, the 100 Intelligent Cities Challenge, the European Interoper-ability Framework, the Connecting Europe Facility, among others. DIGISER’s ambition is to better under-stand the potential and limitations of existing approaches for gathering data, monitoring and measuring the extent and the impacts of digital transformation at the local level. Thus, DIGISER will analyse the transfor-mation of the public sector and its service provision through digital innovation while taking into consideration the diversity of the European territory in terms of socio-economic, cultural, and environmental endowments in different cities, municipalities, and regions.

1 https://ec.europa.eu/info/strategy/priorities-2019-2024/europe-fit-digital-age/europes-digital-decade-digital-targets-2030_en


ESPON // espon.eu 9

1.1 Overview of the work done The DIGISER work plan refers to this inception report as deliverable 1 (D1). According to this plan, D1 is due at month 3 of the working period and its development will be in parallel with the Project Team’s early interactions with the Strategic Advisory Group (SAG; see Figure 1.1).

Figure 1.1 Work carried out according to DIGISER work plan

Within the first three months, DIGISER has launched three of the five tasks that form the work plan: Tasks 1, 2 and 5. The following sections will provide a more detailed overview of the work carried out by DIGISER.

1.1.1 Task 1: Define and implement a participatory plan From the very start of our work on DIGISER, the engagement of expert stakeholders through the SAG was of key importance to gather a range of perspectives and practical expertise on digital innovation in public service provision. Furthermore, the SAG serves to closely link DIGISER to key experts of relevant projects and networks that are complementary to DIGISER such as the 100 Intelligent Cities Challenge, the ‘Join, Boost, Sustain’ movement, and the development of Local and Regional Digital Indicators (LORDI) by ES-PON, among others. Therefore, DIGISER has established a SAG with 16 members representing European cities and regions, research institutions, network associations, national governments as well as European policymakers. (See Annex I).

The SAG will remain involved throughout the DIGISER project development. To ensure that the experts of the SAG are informed about the DIGISER aims and objectives, a first virtual briefing was organised on 5 November 2020 including an opportunity to network online and further create links between DIGISER and the SAG, but also among the SAG members. At this meeting, the roadmap and participatory plan was pre-sented to the SAG. Due to the COVID-19 pandemic, a series of three virtual workshops have been planned at critical points of DIGISER (see Figure 1.2):


10 ESPON // espon.eu

Figure 1.2 SAG engagement process

For the third and last SAG meeting, the option to host a physical meeting is still being considered. The workshop and participatory co-creation approaches are developed following an agile approach. The prepa-ration adapts to the stage of DIGISER and the particular expert feedback needed to proceed. Therefore, the meetings are planned in close coordination with the DIGISER consortium. A first meeting was organised on 9 December 2020 with a focus on informing SAG members about the research carried out and the proposed DIGISER methodology. As this meeting required very specific feedback on detailed research methodology, the feedback from the SAG was gathered as part of a guided discussion which followed the presentation (see Annex I). In parallel, the engagement of end-users follows the (1) Inform, (2) Consult, (3) Engage strat-egy. Representatives of cities and communities are continuously informed about the DIGISER aims and ambitions through the internal channels of OASC and through the channels of linked initiatives such as Join, Boost, Sustain and 100 Intelligent Cities Challenge. This serves the purposes of raising interest among the community and preparing the representatives for consultation and engagement. The details thereof will be explained further Chapter 6.

1.1.2 Task 2: Formulate the research framework and collect data Task 2 is under the responsibility of Politecnico di Milano and is organised into six sub-tasks. During the first three months of DIGISER, Task 2 focussed on the four sub-tasks aimed at defining and delivering all pre-paratory activities necessary to carry out the actual collection of data, that will start in month 3 (M3 in fig. 1.2) of the project. Sub-tasks 5 and 6 are solely dedicated to the first and second rounds of data collection that will feed into the DIGISER database.

During these first 3 months, the research team focussed on developing the conceptual and operational tools that will allow us in subsequent months to implement the methodology to assess the relationship between digital transformation and the provision of public services in the city. This assessment will find a synthesis in an experimental Digital Public Service Value Index (DPSVI) that will be tested on a large sample of European cities and validated during the project.

The development of a first data model for the DPSVI is based on two main pillars: i) the data gap analysis, aimed at analysing the state of the art of the availability of potentially relevant and reusable data on local level, and ii) an extensive literature review, aimed at developing a conceptual model that will guide the data analysis. It is important to underline that these two activities, especially in the initial phase of selecting the sources of data and scientific literature, mutually influence each other.

On the one hand, the review of the scientific literature has produced useful insights for identifying the da-tasets and studies that are relevant for testing the research hypotheses. On the other hand, as gaps in available data became evident, the conceptual framework was modified to allow for the identification of analytical categories that could be effectively researched using attainable data.

In this sense, it is possible to describe the relationship between these two lines of preliminary inquiry – on data availability and on literature – as an iterative process of mutual influence that has facilitated conver-gence towards a possible data model. This approach has allowed us to maintain scientific rigour while ad-dressing operational concerns relating to data availability.


ESPON // espon.eu 11

In detail, the four sub-tasks delivered in the first 3months are:

2 T2.1 Context and development of conceptual and methodological framework (Month 1-3), entrusted with the consolidation of the conceptual framework and identification of the main criteria for analysing the interplays between digital innovation and the public sector’s organisational transformation and public service provision. The outcome of this sub-task is detailed in Chapter 3 of the inception report, while Annex III in-cludes a targeted analysis of the policy context.

3 T2.2 Analytical Framework to assess Public Service Value (M02-M09) focussed on the detailed def-inition of the analytical framework, identifying the key indicators and metrics that will be used for analytical purposes. Chapter 4 includes a detailed description of the conceptual data model that clearly identifies the items to be observed consistently with the conceptual framework provided in Chapter 3. A detailed descrip-tion of the data model is included in Annex V. Finally, preliminary insights regarding the way in which metrics will be defined and data analysed is provided in Chapter 5.3.

4 T 2.3 Sampling (M02-M03) identified the sample to be explored in this research, that includes both a suitable Pan-European sample of cities and communities representing different typologies distributed across European countries that will be analysed in Task 3. Furthermore, this sub-task selected 10 complementary case studies to undergo further in-depth research in Task 4. The detailed description of the sample is in-cluded in Chapter 4.4 while Annex II provides additional relevant details.

5 T 2.4 Data management (M02-M07) entailed an extensive review of possible data sources, aimed at identifying the state of the art of data availability and identifying data gaps (see Chapter 2). Annex VI includes the initial list of datasets considered as sources and the description of their content. Annex IV describes the discussions and choices that shaped the data collection strategy that will mainly rely on the collection of primary data through a survey that will be organised in several thematic modules and deployed digitally. While the survey prototype is still under development, preliminary insights regarding the form of the ques-tionnaires and the engagement strategy are included, respectively in Chapter 5.1 and 5.2.

1.2 Structure of the report ● Chapter 2 includes an extensive data review. After explaining the methodology applied and the

sources of data considered, this chapter provides a review of more than 700 datasets considered during this exploratory stage. Reported results are organised according to different categories of data sources, including large statistical datasets, datasets generated in in-depth studies and re-search projects, and other open datasets released by EU bodies and international organisations. The chapter includes a summary of the main gaps detected and reflects on how these relevant sources can influence DIGISER research.

● Chapter 3 introduces the DIGISER conceptual framework. First, the notion of Digital Public Service Value Index is interpreted and declined under a transitional perspective, intended as the capacity of a public authority to translate the growing ICT potential into transitional opportunities for the public sector in general and of the related socio-technical system in particular. Then, the concept is broken down in three complementary perspectives of analysis, each one referring to a specific literature stream: i) the digital service innovation maturity – clarifying WHAT the target of analysis is, ii) the proneness to change – providing insights regarding HOW socio-technical transitions are taking place at the urban scale, and iii) the orientation to mission – exploring the purposes and the reasons WHY socio-technical transitions are oriented towards achieving political goals.

● Chapter 4 finally presents the data model, drawing upon results from previous chapters. First, it summarises the results of development and selection of data collection scenarios (extensively re-ported in Annex IV) that led to the decision to focus on the collection of primary data through a massive survey (DIGISurvey). Then the chapter provides extensive information regarding the data model for DIGISER that is described both from a conceptual point of view (highlighting the integra-tion of the research questions developed in the conceptual framework) and from a logical perspec-tive, defining the classes and categories of data that are expected to be collected. Finally, this chapter includes a description of the sample of cities that will be targeted for the analysis.

● Chapter 5 provides preliminary information regarding the upcoming activities for DIGISER. Firstly, it describes the strategy foreseen for the development, testing and deployment of the survey that will be distributed through digital tools and structured in different independent modules. Secondly, the chapter includes a brief outline of the engagement strategies that will target the cities listed in the sample to answer the survey. Finally, the chapter provides possible data visualisation methods that will be defined in detail in the upcoming interim report.



In addition, the following Annexes provide extra information, maps and charts that have not been included in the main report:

● Annex I - Strategic Advisory Group Members includes the list of the members of the SAG, their affiliations, and the meeting minutes of the first SAG Meeting.

● Annex II - City Sample describes in detail the list of the cities that comprise the initial sample of analysis in DIGISER, including additional maps and a summary table that analyses the sample according to several statistical criteria.

● Annex III - Policy Context includes an analytical overview of the main policy documents (in the domain of digital transformation and societal innovation) that shape the high-level policy framework of European cities and influence directly and indirectly their technical and organisational choices.

● Annex IV – Data Collection Scenarios reports some of the key debates involving the research team and the SAG regarding the design of a data model. The first part explains what sources of data could feed into the DIGISER dataset and justifies the design of a data collection strategy relying principally on primary data to be collected during the project through a survey. The chapter also reports that the debate occurred within the research team regarding two possible scenarios of primary data collection based on the identification of key services and processes to be observed. Finally, the chapter accounts for the work done to develop in detail the data model that will steer the collection and the analysis of data.

● Annex V - Conceptual Data Model Table contains a large table representing the DIGISER data model that includes all the details currently identified regarding the data that are expected to be collected.

● Annex VI - Datasets from existent relevant studies includes an updated list of the datasets that have been used to feed the data gap review described in Chapter 2.

The following table provides a comparison between the structure of the report and the content expected by the client as described in the DIGISER Terms of Reference and the proposal.

Table 1.1 Content of the inception report

Requirements of the inception report

Reference Chapter

Elaborated proposal including experts, roles, meetings, and objectives for the implementation of Task 1

Chapter 1 + Annex I

Overview of the context developed under Task 2. Chapter 3 + Annexes Description of the conceptual framework and methodological approach to be applied to implement the different tasks and to address the policy questions (Task 2).

Chapter 3

Overview of data and data sources to be collected and used together with a plan for overcoming potential challenges in relation to data collection, data harmonisation, data gaps and data scales and including a report on data situ-ation for the EU Candidate and potential candidate countries (Task 2).

Chapter 2, Chapter 4, Annex VI

Proposal for case studies (Task 4). Chapter 4 Any additional elements results of the discussion taken during the kick-off meeting and agreed with the service provider

Chapter 5



2 Data gap analysis

2.1 Towards the data review Considering the strategic role for the public sector as catalyst of innovation, this report identifies and inves-tigates the features of public value generated through digital innovation – where public value describes the value that public organisations can bring to society as societal benefits in the perspective of relevant global challenges. From the local to the European scale, digital transformation has been a key driver of public sector innovation. Its influence on public services goes beyond the way in which they are conceived, organ-ised and delivered, redefining the boundaries and scope of public institutions as well as their governance mechanisms. For this reason, the report considers the urban ecosystem as the granular scale of observation and data collection.

As Chapter 3 will outline, indicators are chosen for their algorithmic combination into a synthetic index de-fined as the DPSVI. The index can summarise the key results of the assessment conducted on the popula-tion of cities investigated. This set of indicators is defined and aligned to EU long-term goals and missions in order to assess and correlate innovation in the public sector, with emphasis on local contexts. By holisti-cally tackling the complex interrelation between digital transformation, organisational innovation in the public sector and its societal impacts and outcomes, this index serves as a lens of analysis on long-term challenges, being able to inspire innovation strategies.

For this purpose, the DPSVI capitalises on the knowledge generated by several studies that investigated and produced data measuring innovation outcomes in a variety of contexts and scales (see Annex VI). Alt-hough the index is tailored to the local city level, the data analysis has been implemented on a broader scale, relying on principles of data harmonisation and comparability while combining primary data with open data where available.

A fundamental premise to the discourse is the provision of a precise definition of Open Data. The Open Definition (Open Knowledge, 2015) identifies knowledge as open “if anyone is free to access, use, modify, and share it – subject, at most, to measures that preserve provenance and openness”, and establishes principles for defining “openness” in respect of data and content.2 Specifically speaking of Open Data, two different aspects of openness have to be considered: (i) data being legally open, i.e., published under an open licence stating that they can be freely used, re-used and redistributed by anyone, for any purpose – with no limitations for access, re-use, manipulation, redistribution, or conditions limited to attribution; and (ii) data being technically open, meaning that the file is machine readable and possibly non-proprietary, granting free access for everybody with file formats and contents not restricted to non-open source software.

Given these premises, the study conducted a data analysis to understand the feasibility of building an index exploiting and integrating existing knowledge and datasets into a comprehensive set of KPIs and metrics in order to assess the capacity of the public sector to advance digital innovation. The data analysis consists of a systematic review of databases and datasets.

2.2 Data review methodology We recognised the presence of a plethora of studies that produced indicators and indexes exploring, to various degrees, digital performance by EU Member States and tracking their progress in terms of digital competitiveness. Depending on the specific objectives of enquiry, these studies could rely on existing, ac-cessible, or retrievable data (open data as statistical information). Once properly processed, they could pro-vide reliable measurements. Nevertheless, the unavailability of the desired data in conjunction with the spec-ificity of the topics to inquire often paved the way for ad-hoc surveys aimed at obtaining relevant information not retrievable otherwise or vetting into well-defined aspects not covered by available statistical data. The

2 opendefinition.org/od/2.1/en



resulting first-hand data are frequently published openly, granting access, possibility to use, modify, and share them.

In the light of this reasoning, the data analysis process consisted of the following steps:

1. definition of relevant data to collect, informed by the study scope and by initial insights from the conceptual framework (see Chapter 3);

2. identification of available data sources (databases, indicators, scoreboards, indexes assessing and measuring to various extents digital performances and related aspects), and definition of selection criteria;

3. assessment of the reliability, accessibility (direct, indirect, not accessible), and completeness of the datasets, and considering their data models;

4. verification of the scale of the data, considering that the urban ecosystem is the granular scale of observation and focus of data collection of DIGISER;

5. cluster mapping of datasets obtained informed by the indicators analysed and describing the do-main of the information: digital infrastructure, digital skills, funding, public sector, security and pri-vacy, start-up/business environment, use of internet services, and use of social media, among oth-ers, and;

6. assessment of the actual/definitive affinity of the accessible data considering the scope and ambi-tions of DIGISER.

For measuring and assessing digital innovation at organisational level and its broader effect at societal level, data are triangulated (Rothbauer, 2008). Recognising the efforts required to collect primary data (survey fatigue, amount of information to gather, etc.), specific thinking regarded the development of a data collection strategically linked to existing and dynamic data sources. Examples are the Eurobarometer results from different survey waves that capture public opinion, and statistical databases such as Eurostat, in conjunction with a punctual exploitation through data mining techniques of repositories containing data relevant for this study.

Figure 2.1 Data sources and coverage

The data review analysed in parallel (i) datasets from relevant project studies and indicators (extensive coverage), (ii) datasets from in-depth analysis of EU projects or studies (limited sample), together with (iii) statistical EU datasets, and complemented by an ad-hoc research of (iv) specific datasets retrieved from portals providing access to open data published by EU institutions and bodies such as the European Union Open Data Portal (EU ODP).3 The analysis was conducted considering that the quality of a performance measurement depends on the data used to build the indicators and run the calculations. Two crucial factors need to be considered: firstly, the relevance of the datasets in relation to the scope of DIGGER, and sec-ondly, the quality and completeness of data, thereby also considering the ultimate need of comparability and complementarity. Given the presence of the same or analogous data in different databases, both compara-bility and complementarity depend on the presence of the same sample and the use of similar measurement units.

We analysed the data employing different modalities, ranging from downloadable datasets/spreadsheets to maps, and application programming interfaces (APIs). Particularly APIs can provide useful linked data, es-pecially if regularly updated and complete.

3 https://data.europa.eu/euodp/en/home



2.3 The data review In the following, the datasets collected for the data review are investigated. According to the data source typology (Figure 2.1), the results of the analysis conducted are presented. Then, Chapter 2.3.5 considers the quality, availability, operability, and data-related issues, as they emerged from the enquiry conducted.

2.3.1 Relevant project studies and indicators The analysis started from examining a total amount of 33 studies or projects covering topics related to the DIGISER objectives (full list available in Annex VI). The projects/studies selected address one or more of the following topics, identified as significant for DIGISER:

● Governance models and measures for civic engagement;

● State of the digitisation of public services;

● Ecosystemic impacts of innovation policies;

● Data management & use policies;

● Funding & procurement strategies;

● Specific policies & regulations related to digital innovation management.

In consequence, the research produced:

● A list of existing data sets related to research domain that will be used to generate KPIs;

● A list of national/regional data sources that will be adapted and downscaled;

● A list of datasets generated through primary data collection in the Pan-European sample and in case studies.

The analysis of the entire list of datasets derived from other relevant studies and indicators led to the defini-tion of sources to be included in the construction of a comprehensive set of indicators summarising the degree of digital innovation of the cities investigated. Some of the complex indicators built in other projects pointed to data sources, which were particularly significant for DIGISER. Thus, they became primary sources of investigation. An example is the European Union Open Data Portal: although the portal presents limits in its completeness especially in terms of coverage of cities, we considered it as a source of possible further datasets. Moreover, these studies and projects provided fundamental foundations for hypotheses of metrics and methods useful for downscaling data from national/regional datasets, as well as for bottom-up data collection through local datasets and APIs identified at the city level.

From the original set, 14 projects were considered of particular interest for DIGISER and became the object of a more thorough, systematic review. These studies and research projects were selected according to the following criteria:

● granularity (availability of data at city scale);

● relevance for the research questions developed in the conceptual framework of DIGISER (see Chapter 3);

● extension, variety, and representativeness of the sample of city covered;

● obsolescence and integrity of data stored.

The exploration of relevant studies and indicators identified a total amount of 659 datasets. These datasets were then clustered into 9 categories: digital skills, governance, digital infrastructures, start-up/business en-vironment, use of internet services, funding, use of social media, security and privacy, others. As a result, the dataset are so distributed: 19,6% concern digital skills (n:129), 19,1% public sector and governance (n:126), 16,5% digital infrastructures (n:109), 12% start-up/business environment (n:79), 11.1% use of digital services (n:73), 3,2% funding (n:21), 1.7% use of social media (n:11), 1.5% security and privacy (n:10), and 15,3% others (n:101).

Figure 2-2 shows weighted flows reporting on the origin of the datasets, the domain of information on which they provide data, the account of their scale (from country to city), their accessibility and indication of format.



Figure 2.2 Details on datasets: project on study source, topic, scale, accessibility, format

We analysed the datasets considering that the granular scale of observation and data collection of this re-search is the urban dimension. The results show that most of the data (n:467) is available at country level, a smaller account at region/province level (n:23), and 131 represent an urban scale, which is the main inter-est of DIGISER. The remaining 38 datasets were not accessible, and their scale is unknown. We observed their relation between the databases and the conceptual framework, also considering relevance to the re-search domain and scope of the study. Being directly relevant, several of the datasets with a national and regional scale remained of interest. In this respect, they can be considered for context-related analysis and eventually for downscaling.

Based on the analysis, we divided the data into two categories: primary data (directed collected through the DIGISER survey) and secondary data (collected or derived from available datasets). Although some data are freely available and accessible online, some are only available based on a fee (n:46). Regardless of availability upon payment, 214 open datasets were fully accessible. On the other hand, 51 datasets were completely unavailable or inaccessible and 394 were not directly accessible, but synthetically presented in other indexes or indicators, making the extended data irretrievable. It follows that out of the overall amount of data, 92.2% were not usable or not relevant for DIGISER, while 4.9% of data sets were considered as interesting but with issues of obsolescence, accessibility, and scale. Only 2.9% of the data available were considered relevant and usable for DIGISER.

2.3.2 Detailed studies and networks Considering the challenges pointed out while researching recent and comparable data for a large number of European cities, a second analysis concerned a limited sample of datasets derived from in-depth analysis of EU projects or studies. We considered all the relevant projects, studies and networks addressing digital issues at an urban scale. The EU projects analysed that informed our dataset are: NESTA-EDCI, Rugged-ised, Creative and Cultural Cities, Local and Regional Digital Indicators Framework (LORDI) and Synchro-niCity.4 Although featuring the required urban scale, some of these sources display data not homogeneously available for the entire EU, considering samples of cities rather than being comprehensive of all the cities.

4 NESTA-EDCI: digitalcityindex.eu; Ruggedised: ruggedised.eu; Creative and Cultural Cities: https://composite-indica-tors.jrc.ec.europa.eu/cultural-creative-cities-monitor; LORDI: https://ec.europa.eu/newsroom/dae/docu-ment.cfm?doc_id=73046; and Synchronicity: synchronicity-iot.eu.



The analysis also examined the main European networks of cities such as OASC, Eurocities, JBS (Living-in.eu), 100 Intelligent Cities Challenge (ICC), and Green Digital Charter (GDC), extracting the cities involved in one or more of these networks. The data per se reports on the involvement of a city in one or more networks.5

2.3.3 Large-scale statistical datasets In recent years, several large-scale surveys within Eurostat, Eurobarometer and OECD have been launched, providing sets of indicators aimed at assessing the digital maturity of Member States.

We analysed the main large-scale statistical datasets, performing keyword-based search. For the scope, the keywords identified are: ICT, digital, innovation, technology/tech, internet, e-, public, web, R&D/research, municipal. The search measured the number of occurrences of the 10 terms and their iterations over different geographical scales (see Figure 2-3). The analysis showed a significant lack of data at an urban level. In addition, the greater city level can be considered out of the urban scale, due to the absence of focussed information on small and medium cities and related specificities not noticeable at a larger scale. Nonetheless, the national (NUTS1) data can be considered relevant for downscaling operations, as they enable the ob-servation of a large sample of cities and independently reflect context-related variables.

Figure 2.3 Eurostat dataset raw analysis (keyword-based) displaying datasets relevant at different scales and keyword occurrence

2.3.4 Open data by EU bodies Ad-hoc research was then performed within open data portals that collect and provide access to open da-tasets published by EU institutions. Specifically, the inspected portals were the European Union Open Data. Portal (EU ODP) and the European Data Portal.6 They are single points of access to EU-produced open data, harvesting metadata of public sector information available on public data portals across European countries. The executed inquiries were primarily led by criteria formulated contextually to the conceptual framework development (see Chapter 3). The conceptual dimensions of the framework directly drew the aspects to be analysed within the databases and datasets mentioned above.

In particular, the search covered:

1. Datasets showcasing data to various extents related to the level of digital service innovation ma-turity;

5 OASC: https://oascities.org/, Living-eu JBS: https://www.living-in.eu/, ICC: https://www.intelligentcitieschallenge.eu/, Eu-rocities: https://eurocities.eu/, Smart city marketplace: https://eu-smartcities.eu/ 6 data.europa.eu/euodp/en/data/; and www.europeandataportal.eu



2. Datasets published by the Directorates-General of the European Commission, EU Agencies or the other EU institutions and bodies;

3. Services improving in cross–border availability of services.

Informed by the conceptual framework, diverse typologies of processes related to innovation governance and surveys, namely the services to be investigated, have been identified. As a consequence, specific fields of action have emerged, related to areas of society which provide open data about services (e.g., govern-ment APIs, information about traffic, tourism, waste, ICT employment, open data provided by municipalities, e-government, and e-procurement).

Finally, the conceptual framework brought to the formalisation of specific additional queries to verify the existence and availability of data related to: apps developed by PA, blockchain, climate, co-working, fab labs, maker spaces, data management, e-commerce, e-government, e-taxation, energy, engagement, green, health, mobility, open data portals, open software/sources, procurement, and infrastructure.

2.4 Summary of results From the analyses of the datasets retrieved from relevant project studies and indicators, the following ob-servations can be made:

● Limited availability of APIs (Application Programming Interfaces): few databases presented APIs which provide direct access to data from a catalogue without granting access to its functionality;

● Unlinked data: a significant amount of data is published unlinked. Providing linked data means structuring data as interlinked and able to be easily questioned through semantic queries. Unlinked data limits and hinders shareability, machine-readability, and the consequent possibility to have these data used by public administrations, business, and citizens;

● Open data obsolescence: most of the databases and datasets analysed presented data not up-dated through time or properly maintained;

● Limited interoperability: most of the datasets have interoperability limits, being static sets of data rather than data retrieved from repositories built for communicating, exchanging data, and using existing information; and

● Incompleteness: data is incomplete under certain parameters of coverage (temporal, spatial, sam-ple, etc.).

Out of the 659 datasets and indicators from 14 frameworks and studies gathered, only the 3% were actually relevant to the scope of DIGISER and were effectively usable.

The analysis demonstrated that it is possible to reuse the data derived from other existing open datasets in an “as is”-form, even though the operation is limited to basic socio-demographic data acquired from large statistical datasets. In the process of developing a complex indicator where a variety of data plays a role in building information, these data can indeed contribute in describing the city, providing collateral information about the context of belonging, its changes, performances, and features. As such, statistical data contribute in capturing ulterior dimensions of the observed phenomenon. However, what is more evident is the need to collect primary source data, examining the specific topics that are the object of investigation of this study. Hence, the triangulation of secondary data with the primary data (see Chapter 4.1) enables further assurance of the validity of the research, including different and complementary sources. In an analogous fashion, the triangulation of primary and secondary data also provides important knowledge for the validation of interme-diate results and the contextualisation of survey data.



3 The DIGISER Conceptual Framework

This section explains the conceptual framework at the core of the development of the Digital Public Service Value Index. First, it looks at the interrelated components (i.e., digital service innovation maturity, proneness to change, and orientation to mission) that orient the observation of the public administration’s digital transi-tional capacity, relating them to the DIGISER research questions. It then explains how these components are explored, thereby introducing the role of scaling mechanisms in relation to the conceptual framework’s specification. This concept is articulated through three interrelated components, which concur to build com-plementary perspectives of analysis, each one referring to a specific literature stream. Firstly, the digital service innovation maturity identifies WHAT the target of analysis is; secondly, the proneness to change provides insights on HOW socio-technical transitions are taking place at the urban scale; finally, the orien-tation to mission explores the purposes and the reasons WHY socio-technical transitions are oriented to-wards achieving political goals.

3.1 The Digital Public Service Value Index: a transitional perspective The “digital” dimension of public services has been long interpreted as a driver of innovation and changes since it became evident that ICT represents a vital tool for local governance (Economist Intelligence Unit, 2010). Over the last few decades, a novel vision of the public sector activated by the use of ICT in public services emerged where principles such as information sharing, transparency, openness and collaboration became key concepts with relevant organisational and policy implications (Gauk et al., 2019). This slow yet steady process has contributed considerably to making the reflection on governmental capacity more com-plex and demanding in terms of competences required, institutional/organisational arrangements and policy actions' responsiveness and appropriateness.

Within this framework, we refer to digital innovation as the creation of novel products and service, and their implementation; moreover, digital transformation describes the combined effects that several digital innova-tions can create towards novel actors, structures, practices, values, and beliefs. This comes with changes and modifications of existing rules at an organisational level. Coherently, an effective and deep observation and analysis of the role played by digital innovation of public services needs to be framed within a more complex and longer process of technology-enabled public sector reform (Ferro et al., 2013) able to capture the complexity of the service creation process and its capacity to contribute to possible responses to global societal challenges. Making (digital) public services aware of the potential benefits offered by ICT advances represents, in fact, a crucial channel to make cities protagonists of the very urgent transition that every public institution should feel and be responsible for.

The key question targeted in the work is, therefore: are the growing potential benefits of ICT effectively adopted in services conception and provision and turned into opportunities for a public sector re-form to be aligned with the urgent socio-technical transition? This interrogative approach represents both a specification of an overarching and more fundamental question about how the technological “infra-structuring” of public administrations may be turned into value for society. Such a critical interpretational driver feeds the conceptual development of the DPSVI. The DPSVI also evaluates the capacity of a public authority to translate the growing ICT potential into transitional opportunities of the public sector in general and of the related socio-technical system in particular.

The operational translation of such a “dilemma” results in three key research questions.

1. How can digital transformation generate long-term innovation in public sector organisa-tions? This question refers to the capacity of the digital transformation of governance and services to also activate organisational restructuring and innovation (Avgerou, 2000; Poole and Van de Ven, 2004; Deserti and Rizzo, 2019). The creation of new digital services or the digital innovation of existing ones is an opportunity for reflections and learning at the institutional level and challenges institutional organisations towards more flexible and open structures. Hence, being more collabo-rative, transparent, experiments prone and capable (Van der Voet et al., 2016; Plesner et al., 2018; Elliot, 2020).



2. How does public sector organisational innovation generate public value in local contexts? Within a transitional perspective, the digital innovation of governance and services are means to drive change in everyday life practices and behaviours (Shove et al., 2012; Chilvers et al., 2018). Coherently with the shared interpretation of innovation as a process of value generation (Hansen and Birkenshaw, 2007), especially public service innovation, innovation should bind value genera-tion to the situated perspectives of local contexts and the different stakeholders that operate at that level. In doing so, innovation produces public value. Citizens groups, small organisations, innova-tors, are co-producers of such public values based on the principle that beneficiaries are also the experts of their needs (Krogstie, 2006; Emaldi et al., 2017).

3. What paths and key enablers can make best innovation replicable and scalable? Public ad-ministrations neither act alone nor in a vacuum. Their action spaces are strongly affected by several factors (the most crucial being market pressure; this is especially true in the case of ICT for public service provision and creation (Albury, 2005)), by experiences of similar public administrations, by their belonging to national or international networks enabling experiences exchange and collabo-ration (Campbell, 2013). In these dynamics, cities represent sources and drivers of the replication and diffusion of innovation.

The three research questions underlying the development of the DPSVI are opening the way towards a set of indicators that contribute to making the key mechanisms of socio-digital transition – scaling-up, scaling-deep, and scaling-out (Riddell and Moore, 2015; Omann et al., 2019) – explicit.

The introduction of a transitional perspective and the related research questions suggest looking at the DIGISER analysis as a transformative learning (Bateson, 1973) or Triple-Loop Learning (T-LL)7 process to be activated through the data collection activities at the scale of local institutions. Learning is central due to its contribution to a robust strategy for accelerating and guiding socio-technical transformation processes (Loorbach, 2010; Loorbach and Rotmans, 2010). The T-LL acts at the organisation level of learning, and it reflects on the maturity of a transition. In other words, the more changes are embedded and internalised in the subject, whatever the complexity learnt, the better T-LL is achieved. The more disruptive the change to be achieved, the more Triple-Loop Learning needs to engage all involved actors, from individuals to organ-isational and institutional infrastructure, up to the societal scale (Johannessen and Wamsler, 2017).

In T-LL, each of the existing learning loops (single, double, triple) has been conceptualised by different authors (Sinek, 2009; Engelbart, 2002): they focus on the goal each loop reflects on – the “what”, the “how”, and the “why” – as shown in the graph below (see Figure 3.1).

7 Triple-Loop Learning is required when problems are wicked and unstructured and especially when the deep underlying causes and context have to be taken into account in redefining, relearning, and “unlearning” what we have already learned before (Gupta, 2016). In T-LL the constant questionings and modifications help to create a shift in perspective and ulti-mately a transformational change. Learning in this model is relevant in the perspective of a deep change especially when it is coherent with any transformation or innovation process within the perspective of transition. In Stacey’s terms, T-LL is manifested as a form of collective awareness (Stacey, 2007); the relationship between organisational structure and human behaviour changes as the organisation learns how to learn and understands more about the values and assump-tions which lie below the patterns of actions (Kahane, 2004). Triple-Loop Learning allows not only individuals but also organisations to question whether the values and assumptions are locking them into a recurring cycle in which today’s solutions become tomorrow’s problems. In this way, the values, as well as the strategies and expectations, can be modi-fied (Argyris and Schön, 1978).



Figure 3.1 Triple-Loop Learning and the “what, how, and why” questions

Looking at the T-LL in terms of the three reflection levels able to activate learning into organisations, the DIGISER concept has been developed conceiving the Digital Public Service Value Index as structured in agreement with the “what” (single loop learning), the “how” (double loop learning), and the “why” (triple loop learning) questions (see Figure 3-2). This conceptual setting will be reflected also in the survey for data collection (see Chapter 4), making it work as much as possible as a learning infrastructure, supporting re-flection especially for those public administrations that will be the focusses of the analysis.

Figure 3.2 Conceptual structure of the Digital Public Service Value Index

1. Digital service innovation maturity. Representing the “what”, the “digital service innovation ma-turity” is the concrete object of the DIGISER analysis. “Digital innovation is about the creation and putting into action of novel products and services; by digital transformation we mean the combined effects of several digital innovations bringing about novel actors (and actor constellations), struc-tures, practices, values, and beliefs that change, threaten, replace or complement existing rules of the game within organisations and fields” (Hinings et al., 2018: 1). In DIGISER we will refer to two main elements: (i) the advancement in technological “infrastructuring”, i.e., the product, and (ii) the changes it requires or activates in the structures, practices, values/beliefs where it occurs. The advancement in the technological infrastructure asks for new service models and interactions to drive changes in the processes, structures, practices, values and culture of public administrations (Schein, 1985).



2. Proneness to change. The shift from reactive to proactive service delivery mechanisms is enabled by the transition from e-government to digital government, where the use of digital technologies is assumed as an integrated part of governments’ modernisation and innovation strategies, creating public value through the engagement of a broad ecosystem of stakeholders (OECD, 2017) and is only possible with a strong orientation to change (low internal resistances, strong competences, availability to learn; (see Fernandez and Rainey, 2006). It represents the “how”, the way chosen to guarantee both an innovation process and an effective exploitation of the growing beneficial poten-tials of digital transformation. This considering that the application of simple digital improvements to existing solutions alone is insufficient. Transformation requires a holistic application which takes into account the interplay of various challenges and drivers at different scales (Gauk et al., 2019).

3. Orientation to mission. Exploring the “why”, the “orientation to mission” represents the final aim of every public action (Mazzucato, 2018), the goal to achieve in terms of systemic change as an answer to one or more societal challenges local authorities are asked to face. It represents the overarching perspective the innovation of public services should be contributing. In short, “orienta-tion to mission” represents the transitional driver of service innovation and offers the key values that should be mobilized by the services and their digital innovation.

In the following section the triplet “digital service innovation maturity”, “proneness to change”, “orientation to mission” is presented in more detail.

3.2 Inside the concept As illustrated in the previous paragraph, DIGISER conceptualises the DPSVI by focussing on three compo-nents, “digital service innovation maturity”, “proneness to change”, and “orientation to mission”. These are described in detail in the following paragraphs are specified as follows:

● Digital service innovation maturity is described through the digital maturity and the level of service embedment

● Proneness to change is described by change management and innovation governance ● Orientation to mission is described through alignment to Sustainable Development Goals

(SDGs)

These three elements of the DPSVI are strongly linked to transition and innovation in governance and public service provision, considered crucial to the analysis and well operationalised by the three research questions described in the previous paragraph. As already underlined and described in Table 3.1, the three research questions represent a productive driver to the identification of indicators composing the DPSVI by leveraging the three key mechanisms: scaling-up, scaling-out and scaling-deep. The three mechanisms “underscore the complexities and complementary nature of the strategies involved in advancing large systems change” (Riddell and Moore, 2015: 3), and together they can explain: (i) how a socio-technical transition works, and (ii) how one innovation can contribute to transition by activating changes, transformations, learning dynamics and synergies with other innovations. Table 3.1 Mapping scaling mechanisms over the DIGISER research questions

Research questions underlying the definition of a DPSVI Perspective explored

How can digital transformation generate long-term innovation in public sector organisations?

Organisational Change and Performance (Scal-ing-Up)

How does public sector organisational innovation generate public value in local contexts?

Local Context Change and Performance (Scal-ing-Deep)

What paths and key enablers can make best inno-vation practices replicable and scalable?

Relational capacity: Replication and Transfer (Scaling-Out)



Scaling-up - Organisational change. This mechanism entails modifications of the organisational struc-tures, routines and practices, and their impacts on policies, rules and laws. Scaling-up mechanisms can be grounded in different levels of any complex organisation (Argyris and Schön, 1978). In line with the T-LL scheme illustrated above (see Figure 3.1), they can refer to changes in the actions undertaken to support service development and provision (‘“what”), but also to modification in how innovations are managed and decisions about digitalisation and innovation governance processes are taken within the public administra-tion’s (PA) organisational structure (“how”). Focussing on the role of digital transformation as a driver of change, DIGISER explores in particular how scaling-up mechanisms can contribute to support long-term innovation within public sector organisations.

Scaling-deep - Local context change. Scaling-deep refers to impacts generated on cultural roots, and it is related to the notion that “durable change has been achieved only when people’s hearts and minds, their values and cultural practices, and the quality of relationships they have, are transformed” (Riddell and Moore 2015: 3). In the context of DIGISER, scaling-deep mechanisms are understood in terms of public value generation, and they look specifically at the capacity of institutions to generate innovation in their context of reference, and at the dynamics through which public sector organisational innovation effectively impacts local context dynamics. Notably, scaling-deep dynamics are affected by the capacity of PAs to support ser-vices that meet context-specific needs, which in turn depends on “context readiness”, which is shaped by socio-demographic factors, as well as by the degree of digital capacity of local actors (including service end-users) and by the type and quality of existing digital infrastructures.

Scaling-out - Relational capacity. This mechanism emphasises replication and dissemination of innova-tions in different communities (Riddell and Moore, 2015) and deals with the spread and (re)production of innovative values, ideas and tools across the multiple levels of interactions characterising socio-technical systems (see Geels and Schot, 2007). In line with the DIGISER research questions (see Table 3-1), scaling-out dynamics are related to the paths and enablers that allow for innovation practices to be replicated in different contexts, e.g., to the mechanisms through which cities reproduce, replicate, and adopt digital ser-vice procedures or innovation practices developed in different contexts8.

Scaling mechanisms play a role in the concept specification that is twofold.

On the one hand, they represent the systemic transformation that a public authority achieves to acti-vate in its context, by managing relations and interactions with other public institutions at different levels, with private actors, with citizens, being context-aware, flexible, prone to learn and adaptive, open to the changes required by their environment. This first dimension foresees a proactive role of public institutions, which drive innovation process generation and are engaged in the spread of innovation across the multiple levels of socio-technical systems (see Geels and Shot, 2007; Geels, 2020). In this case, scaling mechanisms allow exploring different transition pathways that may emerge along service innovation processes. First, scaling-up mechanisms allow investigating the capacity of public institutions to proactively generate innova-tion dynamics through the modification of their internal governing mechanisms, of organisational routines, of the way in which decisions are taken and of tools are developed and adopted. The capacity of PAs to support scaling-up dynamics and to support long-term organisational innovation primarily depends on their “proneness to change”, and particularly on their capacity to “manage change” (see Figure 3-3), e.g., through the acquisition of new competences and skills, the redefinition of legal and funding schemes, or the modifi-cation of internal governance structures. Also, scaling-deep dynamics enable to explore how institution-led

8 Achieving innovation in the public sector can be difficult and requires additional, targeted support and resources. In recent years, there has been a significant growth in the type and number of organisations and structures dedicated to supporting innovation in the public sector (OECD, 2017). These are known as teams, units, labs, networks to name a few. Among these, innovation-focussed networks and innovation labs have attracted most of the attention. Networks can sup-port and motivate public sector innovation by creating a space where innovators can share ideas, practices and challenges for implementing innovations. Dedicated innovation units/labs can help address some of the barriers to innovation: e.g. compensate for the lack of innovative leaders and champions, and help overcome rigidities in the reward and incentive systems that can often hinder innovative performance in the public sector. They can foster the creation of organisational knowledge about how to apply innovation processes and methods and support more collaborative and harmonious ap-proaches in problem solving. This can help address departmental silo thinking by adopting cross-cutting, inter-disciplinary approaches, bringing together different or new tools, methods and skills. (OECD, 2018: 198)



innovation can support public value generation, and to identify the mechanisms through which innovation is adopted and used by local actors across the public, private and civic realms. Finally, DIGISER assumes that “the role of public managers is not necessarily to accomplish all public innovations themselves, but rather to facilitate and align constellations of diverse actors to address various societal challenges" (Bugge et al., 2019: 4). Accordingly, scaling-out dynamics that look at the capacity of PAs to generate and spread innova-tion ultimately depends on their position within the network in which they are embedded, and on their level of interconnection with other agents acting across governance levels.

On the other hand, scaling mechanisms represent the dynamics that a service creation (ideation, de-sign, experimentation, implementation, supply) or innovation may require to generate an effective response to the societal needs it is created for. In this second case, innovation is related to the service itself (or to the processes developed within specific service areas, see innovation governance section in Chapter 3.2.2). In this second set of innovation pathways, the role of PAs changes. First, it consists in ‘de-tecting’ innovation signals developed in different arenas (scaling-out) and in adopting and adapting them in a way that allows responding to context-specific needs. Second, it requires PAs to embed innovation in their actions through a modification of their internal mechanisms and procedures (scaling-up).

Accordingly, considering the role they play and the relevance they have as for the coherence with the DIGISER research questions, scaling mechanisms are considered as a cross-cutting interpretative category. They are related to two elements specifying the DIGISER concept, namely the “level of service embedment” and the “change management” (see Figure 3.3).

Figure 3.3 Detailed DIGISER conceptual triplet

3.2.1 Digital Service Innovation Maturity Digital technologies are deeply affecting people's lives in general and how people interact with public infra-structures in particular (Welby, 2019). These technologies, their growing availability and performances, the wide use of data, the wide offer of services provided by a large variety of actors are re-shaping the value supply chain of public service and the associated concept of public good. “Over the past decades, countries have enacted large-scale public sector reforms to prioritise digitalisation to enable greater efficiency and effectiveness of public services. As part of these efforts, they have been investing considerable resources to adopt new practices to modernise their services and make them more responsive to citizens’ needs” (OECD, 2019: 146).



Exploring public service means to deal with one of the two key roles of the public administrations: the “man-agement and implementation of the whole set of government activities dealing with the implementation of laws, regulations and decisions of the government and the management related to the provision of public services” (UNDP, 2015: 2). In public services provision, a crucial innovation role has been and still is played by digital technologies, that place increasing and new demands and expectations on the public sector. Re-alising these technologies’ full potential of these technologies still represents a key challenge for govern-mental organisations (DESI, 2019; DESI, 2020) although the digital transformation of government and ser-vices has now achieved, even with relevant differences in the European context, an advanced state of ma-turity. The fast evolution of technologies continuously offers new opportunities towards digital government, as well as towards transparency and openness. Innovating public service is, in fact, an integral part of digital government strategy and more and more “relies on a digital ecosystem comprised of government actors, non-governmental organisations, businesses, citizens’ associations and individuals, which supports the pro-duction of and access to data, services and content through interactions with the government (for example, open data platforms common to several governmental institutions (OECD, 2019: 146).

Most recent DESI reports (2019; 2020) show that, from the user side, the demand for digital public services is growing and this represents the evidence that the societal digital literacy is growing together with the digital transformation and maturity process. In this dwelling of growing a demand for and growing and evolving offer of digital services two elements appear relevant: the level of digital infrastructure and transformation repre-sents a relevant focus (what technologies and to what extent services are digitally offered and used); and the level of adoption of the digital service and its internalisation in the public administration organisation and setting (how far the digital service is accessible and adopted by most of the citizens; to what extent the digital service potentials are fully exploited by the skilled organisation and affect the public administration renewal and innovation. DIGISER labels the first as the digital maturity and the second as the level of service embedment specifying what has been recently defined digital readiness of innovative public services (Eu-ropean Commission, 2020) explored through four dimensions: technological, societal, organisational and legal (p. 7); in DIGISER the last three dimensions interpreted as level of service embedment.

The digital maturity mainly attains to the extent to which public administrations embrace new digital tech-nologies and deliver innovative public services. This is usually referred to the 4 layers of the European Interoperability Framework (EIF) aligned with user centricity principles defined by the Tallinn Declaration in 2017, to ensure that the adoption of new technologies does not lead to creation of new silos or new com-partmentation. The digital maturity considers the distinction between mature technologies and emerging technologies (from Blockchain to AI and cloud computing), these last playing a relevant role in describing to what extent the public authority challenges itself while developing new services, eventually exploiting exist-ing solutions and/or sharing the effort with other authorities.

The level of service embedment reflects the role played by the service in driving changes in public author-ities. The three scaling mechanisms reflect here a different perspective. The scaling-up mechanism is related to the achievement of the public authority to supply the service autonomously, reflecting the completion of a process of adaptation and renewal throughout service development. The scaling-out reflects the public au-thority’s capacity to become a driver of the adoption by others of the service either in case it has been entirely and autonomously developed, or it represents the result of an improvement process. In this respect, specific and dedicated experiences of sharing and collaborative networks are crucial and allow public authorities to access a large variety of services and related solutions opportunities. The scaling-deep mechanism reflects the level of adoption by the users in service development, so achieving the level of changes in practices and behaviours that the service innovation aims to affect.

3.2.2 Proneness to change The second dimension explored by the DPSVI concerns PA’s “proneness to change”. As previously men-tioned, (see Chapter 3.2; Fernandez and Rainey, 2006), the degree of proneness depends on a variety of factors, including organisational structuring and degree of internal resistances, attention towards capacity building and presence of adequate competencies, and availability to learn, i.e., to engage in iterative learning processes that encompass the different dimensions illustrated in the T-LL model (see Figure 3-1).

“Proneness to change” is a function of the capacity of PAs to manage change in order to engage into digital innovation processes, and to shape change (Rammel et al., 2004) by supporting innovation pathways through transitional scaling dynamics. In line with a transition management perspective (see e.g., Loorbach, 2010; Kemp and Loorbach, 2006), change management as defined in this work encompasses different



levels, including the strategic level (in terms of problem structuring and the definition of long-term goals related to specific and urgent societal challenges); the tactical level (referring to agenda-setting, partnership development and networking); and the operational level (related to actual experimentation and implementa-tion of innovative policies, practices and tools).

On the one hand, change management in the DIGISER context refers to the capacity of PAs to put in play a set of actions, norms, policies and tools either to proactively support innovation in (digital) service devel-opment and provision, or to increase its capacity to detect and adopt innovative solutions developed by other PAs. Accordingly, it relies upon the capacity of an organisation to adapt its internal procedures in order to adjust to both internal and external circumstances (see scaling-up mechanisms), but also upon its ability to create spaces for other agents (both from other contexts and the local level) to engage along the different dimensions of governance innovation processes (as described in the following paragraphs). “Proneness to change”, therefore, also includes the capacity “to utilize innovative bottom-up developments in a more stra-tegic way by coordinating different levels of governance and fostering self-organisation through new types of interaction and cycles of learning and action for radical innovations offering sustainability benefits” (Kemp et al., 2007: 3). On the other hand, change management refers to the capacity of an institution to support government modernisation and to modify its internal procedures and practices to “create space for short-term innovation and develop long-term sustainability visions linked to desired societal transitions'' (Loorbach, 2010: 163). Concerning digitalisation processes, this ultimately relates to increasing its openness, transpar-ency, and effectiveness in terms of digital service development and ICT-enabled service delivery.

Change management is affected by a variety of factors, including (i) the degree of awareness PAs have about their role and transitional potential; (ii) their commitment to change, e.g., in terms of proneness to experiment and to use advanced ICT technologies, but also to activate new modes of governance; (iii) their capacity to act, e.g., with respect to the adoption of adequate tools and procedures (see “digital maturity”, Chapter 3.3.2) and (iv) their role and position in their network, e.g., in terms of capacity to develop policies and practices in the interaction of a variety of societal actors within innovation governance structures (see Loorbach, 2010).

In addition, “proneness to change” is further explored in the following in relation to four innovation govern-ance processes, namely:

1. Data management. Data (open and big) represent an un-precedent opportunity for growing avail-ability combined with the growing computational and analytical potentials. Although public authori-ties have been exposed to the request for data disclosure, more recent reflections (Concilio and Molinari, 2019; Concilio and Pucci, 2021) also show the growing relevance played by big data as a resource mainly having private owners and making cities compete for this issue at the market level.

2. (Public) Procurement. Public procurement represents one of the most important innovation chan-nels for public authorities. The way public authorities run procurement procedures (the procurement requirements they develop, the opportunities they identify, the level at which they outsource service provision, the avoidance of vendor lock-ins risks, among other factors) can reveal a lot on their innovation strategy and their proneness to learn.

3. Societal engagement. Following a quadruple and quintuple innovation helix approach, the role of citizens is gaining importance up to obliging public authorities to develop dedicated programs over-coming participatory approach and transforming societal engagement into collaborative city making towards new citizenship models able to give answers to the need for larger and deeper democracy.

4. Institutional capacity building. No innovation is possible without learning, and the organisational level of learning is the most challenging form. Many elements contribute to institutional capacity building: employee skills and competences, the internal personnel mobility, collaboration and shar-ing, the organisational involvement in experiments and tests.

The four selected processes reflect key challenging opportunities public authorities continue to face and are widely discussed in literature on innovation and public authorities.

3.2.2.1 Data management

Data is one of the most valuable resources in today’s societies, economies, and governments and effective data management strategy is more and more becoming an imperative towards better public services. “Open Government Data (OGD) can be a powerful lever for social and economic development. It can also be used to strengthen public governance by improving the design of public services with a citizen-driven approach, by enhancing public sector efficiency and by spurring public sector integrity and accountability. By ensuring OGD availability, accessibility and reuse by public, private and civic actors, governments can design more



evidence-based and inclusive policies, stimulate innovation inside and outside the public sector, and em-power citizens to take better-informed personal decisions. (...) However, at the central/federal level, the ex-tent to which countries conduct initiatives to promote data re-use outside government (such as hackathons and co-creation events) and inside governments (via training and information sessions to civil servants) varies greatly. Moreover, few countries monitor the economic and social impact of open data as well as the impact of open data on public sector performance. Secondly, data collected by the OECD suggests that there might be an implementation gap in a number of countries where policy developments have been in-troduced very recently including notably in some of the eastern European countries such as the Czech Re-public, Latvia, the Slovak republic and Slovenia” (OECD, 2018: 192). “Open government data (OGD) policies are set on ideas and principles that centre on making data from public bodies available to everyone in open, free and accessible formats” (OECD, 2019: 148).

As part of the rhetoric surrounding the Open Government and Smart City concept, cities, at their own gov-ernment scale, are increasingly challenged in relation to data (management, governance, processing, stor-age, publishing etc.) as per the growing power acquired by the data market and the great relevance assigned to data ownership rather than to data-exploitation know-how is affecting the development of a data culture and is slowing down the embedding of data-related expertise inside public administrations. Concurrently, policies call for more open data to foster service innovation and government transparency, but it is more and more crucial the policy framework they are able to develop in order to shift the data culture from ownership to exploitation (Walravens et al., 2021).

Public authorities may become drivers for data management strategies that make cities less impacted by the (big) data market and transform cities into data ecosystems where: citizens become (big) data sources and literate users; private actors are exchanging information while transforming their services into public value production systems; public services are considered strategic data collection and utilizations systems enabled by format and procedural standards that allow the data to be used by anyone; public authorities promote initiatives of data utilisations for the creation of innovative services (Concilio and Pucci, 2021).

3.2.2.2 Procurement

(Public) Procurement refers to techniques, structured methods, and means used to streamline an organi-sation's procurement process and achieve desired results while saving cost, reducing time, and building win-win supplier relationships. “Governments continue to use public procurement to pursue secondary policy objectives while delivering goods and services necessary to accomplish their missions in a timely, econom-ical and efficient manner. The high relevance of public procurement for economic outcomes and sound public governance, as implied by its large volume, makes governments use public procurement as a strategic policy lever for achieving additional policy goals, which aim to address environmental, economic and social chal-lenges according to national priorities” (OECD, 2018: 174).

This is true at all government levels, as public procurement is one of the main demand-side innovation policies to support innovative goods and services. ICT plays a twofold role in procurement processes: (i) it may represent the object of the procurement and in this respect the manner the procurement processes are conducted may reveal important insights on the digital innovation strategies of the public authorities (the innovation perspective, the level of delegation against vendor lock-ins, the ICT advancement challenge, the data management constraints and interoperability orientation, …); (ii) it may represent the supporting infra-structure of the procurement process (e-procurement) enabling “governments to increase the transparency of public procurement activities as well as collect consistent, up-to-date and reliable data on procurement processes” eventually feeding “other government information technology (IT) systems through automated data exchanges, reducing risks of errors and duplication” (OECD, 2019: 138).

Within the first perspective, it is crucial to consider the role played by “pre-commercial” procurement (PCP) and Public Procurement of Innovative solution (PPI) as a means to use public needs as a driver for innova-tion. The concept was introduced as a response to the need to reinforce the innovation capabilities of the European Union while improving the quality and efficiency of public services: it is a very challenging and fascinating concept still having a complex implementation perspective in terms of standard procedures (what is meant by PCP?) as well as in terms of interpretation (Is it a demand- or a supply-side instrument in relation to innovation? see Edquist and Zabala-Iturriagagoitia, 2015).



3.2.2.3 Societal engagement

The notion of societal engagement, while it is characterised by a shared pool of keywords, concepts (e.g., transparency, efficiency, accessibility, inclusion, or even democracy) and methods (e-democracy, co-crea-tion, co-decision making, quadruple helix framework, innovation 2.0, etc. it proved capable to absorb the cultural influences of the contexts where it is materialised into policies and practices, cross-fertilising itself with pre-existing political and cultural traditions (Sintomer et al., 2016). Narrowing down the extreme variety of approaches to societal engagement in complex governance (Fung, 2006), in DIGISER’ interpretation, it is possible to identify two streams of literature and practices. The first reflects on the political foundations of societal engagement. It focusses on the active participation of non-elected stakeholders in public decision-making processes as a means to reinforce local democratic systems in response to the generalised crisis of institutions of representative democracy (Avritzer and Santos, 2005). In this stream, we will observe and contextualise all those practices experiments of participatory democracy, where the engagement of urban actors is explicitly finalised to influence a policy instrument, as the participatory budgeting, the engagement of local stakeholders in urban planning policies, the local petitioning and referendum mechanisms. The sec-ond stream approaches the engagement of urban actors from a co-design and co-creation perspective, considering the engagement of urban actors as both consumer and producers of public services and gener-ator of public value, focussing on the “constellation of design initiatives geared toward making social inno-vation more probable, effective, long-lasting, and apt to spread” (Manzini, 2014: 65). It is important to high-light how the implementation of a participatory process on public policies necessarily entails the active en-gagement of urban public authorities, the co-creation of public services can also be grounded on bottom-up initiatives that establish themselves as capable of generating public value, eventually institutionalised at a later stage (Ibarra, 2007). Even though both approaches to societal engagement were pre-existing the recent digital transformation, both have been significantly affected by the widespread availability of digital technol-ogies that not only provided new means and opportunity to engage people, but also enhanced the potential active role of societal actors as providers of input and data, co-producers or owners of the services (Public Private People Partnership - PPPP) and unlocked new domains of engagement. On the one hand, a new generation of collaborative platforms has been introduced and implemented by urban authorities to extend e-participation opportunities (Sestini, 2012). The deployment of collaborative platforms by an urban authority entails several challenges regarding “hard” technological choices (e.g., code, licenses, data ownership) as well as interaction design choices that will be observed in DIGISER as an indicator of both the level of digital maturity and as proxies to understand the actual orientation to interoperability and openness. On the other hand, the multiplication of co-design and co-creation initiatives, such as hubs and living labs, as ecosystems of innovation following the quadruple helix innovation model, are evidences of quasi-organisations encour-aging and sustaining projects and topics addressing the digital transformation (e.g., the co-design of e-gov-ernment services, or the active involvement of urban actors in the collection and publishing of new open data series) are intended to be a relevant indicator of the vitality and integration of the urban innovation ecosystem.

3.2.2.4 Institutional capacity building

Starting from the assumption that “organisations require both existing and novel organisational capabilities to utilise digital technologies in order to respond to transformation drivers” (Faro et al., 2019: 2), the explo-ration of institutional capacity building addresses both training and educational activities put in play to enhance the digital skills of civil servants and the proneness of PAs to enhance and mobilise their organisa-tional and technological resources through the adoption of ICT technologies or the modification of internal rules and procedures (in relation with the scaling-up dimension, see Figure 3-3).

Concerning the former, institutional capacity building focuses on the role played by employees’ skills and competences, which is crucial when digital innovation is targeted, as “civil servants need the ability, motiva-tion and opportunities to contribute to innovation. Therefore, human resource management (Hm) is an im-portant lever for supporting public sector innovation by enabling managers and front-line staff to formulate ideas that result in new and improved ways to deliver public services” (OECD, 2017: 196). DIGISER, there-fore, looks at the practices developed to enhance organisational capacity for innovation, such as “incentive structures and awards; managerial and leadership approaches; organisational practices related to recruit-ment, training, mobility and compensation of employees; and job design factors such as autonomy and ways of working” (ibidem), as one of the main factors contributing to “digital service innovation maturity”. In partic-ular, attention is paid to the capacity of local government to support both flexibility (e.g., through smart work-ing) and autonomy in service development and delivery (see “level of service embedment”, in Figure 3-3).



With respect to the latter, the proneness to proactively foster innovation through capacity building is assumed to be grounded into the PAs’ capacity to engage into iterative learning processes (see the T-LL model, Figure 3-1) affected by internal and external constraints and pressures. In this direction, capacity building is affected by the relational capacity of an institution (see scaling-out mechanisms as described in Chapter 3.2), e.g., concerning collaboration schemes with other cities and networks (relational capacity), as well as the partic-ipation to projects and programmes targeting service digitalisation objectives or facilitating a vast deployment of digital technologies.

3.2.3 Orientation to mission The mission-oriented approach to research and innovation policies has been recently receiving attention in the scientific domain and in the actual policymaking field. It refers to a methodology that aims to manage complex governance aligning it toward specific and explicit missions. Differently from other approaches to innovation based on the simplistic equation that relate innovation to mere digital transformation, it “does not facilitate innovation merely by levelling the playing field with horizontal policies that prescribe no direction. On the contrary, such policies, by definition, give explicit technological and sectoral directions to achieve the ‘mission’. At the same time, to be successful, they must also enable bottom-up experimentation and learning” (Mazzucato, 2018a: 5).

The mission-oriented approach can be recognised when integrated policies and initiatives are aligned toward a clearly defined mission, (i.e., achieving a measurable goal or implementing a solution), targeting a specific and explicit societal challenge (i.e., reducing social exclusion in a given context). Targeting missions implies that research and innovation strategies and policies shall be inscribed in a larger strategic framework and associated with consistent regulatory and organisational provisions. In this perspective, public organisations must set long-term objectives towards, and commitment to, clearly identified missions deducted through a process of prioritisation of societal targets and create conditions for very effective solutions to emerge, root and survive (Mazzucato, 2018b; European Commission, 2018).

Orientation to mission is neither achieved by implementing one single initiative nor through a set of homo-geneous initiatives. A sound orientation to mission is shaped through a differentiated set of initiatives being ambitious, cross-disciplinary, exploratory, and ground-breaking in nature and mixing narrowly defined initia-tives, aimed at single, well-defined objectives with more broadly defined initiatives addressing societal chal-lenges and targeting the transformation of the system as a whole.

Despite its fluidity and context-sensitivity, the notion of mission-oriented approach has been introduced in DIGISER after a critical reflection on the technology-driven assumptions underlying much of the research on the relation between digital transformation and societal innovation in recent years (Wyatt, 2017). Indeed, the perspective of DIGISER goes beyond simplistic techno-determinism but aims to explore the features of digital transformation in public service provisions researching the ultimate goals and purposes of those choices, and if the technical and organisational policies of public authorities are actually oriented to societal goals, and how much those goals are clearly defined, and their achievement properly monitored.

In order to adopt a common definition of possible societal missions, DIGISER adopts two different perspec-tives.

First, DIGISER addresses the Sustainable Development Goals (SDGs) as a major, urgent set of societal challenges. SDGs will represent the reference framework to conceptualise the orientation to mission while developing the Digital Public Service Value Index in coherence with a recent framework to roadmap mission-oriented innovation policy for SDGs (Miedzinski, 2019). Governments and international organisations are increasingly aware of the potential of science, technology, and innovation to accelerate the transition towards meeting the Sustainable Development Goals (SDGs) and are investing large resources. Many of these dif-ferentiated efforts are characterised by being cross sectoral, multi actors, and somehow indifferent to eco-nomic conditions (Stafford-Smith et al., 2017). Recently, ESPON itself launched an initiative to explore the strategies to achieve SDG at city level. This will be taken into account in DIGISER (Markianidou et al., 2020).

Second, DIGISER takes into account several challenges and missions established in policy documents, declarations and commitments that were developed directly at the city scale and focussed specifically on topics of digital transformation and innovation. In particular, DIGISER put at the centre of its analysis the principles established by the members of the Living-in.eu network, that already formulated a vision of the relationship between technological transformation and organisational innovation in the public sector and in the urban societies:



● a citizen-centric approach; ● a city-led approach at EU level; ● the city as a citizen-driven and open innovation ecosystem; ● ethical and socially responsible access, use, sharing and management of data; ● technologies as key enablers; ● interoperable digital platforms based on open standards and technical specifications, Application

Programming Interfaces (APIs) and shared data models.

The approach of cities toward the tackling of these challenges (i.e., the “mission-oriented” approach) will be considered as a relevant parameter for the assessment of the capacity of cities´ strategies of digital trans-formation to generate public value.

In addition to the living-in.eu declaration, we analysed in detail several other high-level policy documents focussed that are designed to define long-term objectives and goals (i.e., missions) that are expected to influence more or less directly the setup and implementation of missions-oriented strategies at the urban level. Starting from the long-term strategies, programmes and initiatives on research and innovation of the European Commission, we reviewed in detail several policy documents, trying to identify key transversal goals and missions and how supra-local policy instruments can prove capable to mobilise a mission-oriented culture in the city context. Annex IV includes the result of this in-depth review.

The orientation to mission represents an umbrella portion of the DIGISER concept that will be included in the DPSVI after the first analytical work on the other two spheres will be completed. Considering the context-sensitivity of this phenomena, the research team expects to use in-depth qualitative methods (i.e., interviews, focus groups, detailed documental analysis) to explore the orientation to missions of the cities researched. For this reason, this research perspective would be only partially fulfilled through the large-scale survey and is expected to be developed accurately in the 10 case studies where the in-depth analysis will be carried out.



4 The DIGISER data model

4.1 Scenarios of data collection between the data gap and the conceptual framework As explained in Chapter 1, the DIGISER data model has been drawn upon the results of the data gap review (Chapter 2) in order to answer the questions developed within the conceptual framework (Chapter 3).

The process of definition of the data model has been developed taking into account both theoretical and practical concerns, and required the development and discussion of several possible scenarios of data col-lection that have been finally selected involving both the client and the experts of the SAG. The detailed description of this work is included in Annex V. The main outcomes of this work can be summarised in two points.

First, the research team decided to develop a data model that will be fed in first place by primary data to be collected directly through a survey. The survey will target a statistical sample of cities that will be further described in this chapter. While in the initial research design the role of the survey was considered as secondary with respect to the possibility to harvest and reuse existing open datasets, the actual limited availability of data at city scale, the risks of biasing the data entrenched in complex downscaling processes, and finally the actual novelty of many of the objects of the DIGISER research pushed the choice in the direction of a massive survey (additional details are available in Chapter 5).

Second, the team decided to consider the key processes introduced in the concept, namely the innovation governance processes (Data Management, Procurement, Societal Engagement, Institutional Capacity Building) and the scaling mechanisms as the starting drivers to the survey so implementing a “from processes to services” data collection strategy. This choice will allow a more direct observation of socio-technical transitions taking place at urban scale within the public sector while addressing digital service innovation at the local scale. Starting from innovation governance processes and reconstructing the relation to underlying scaling mechanisms (see Chapter 3.1), the data collection strategy aims at obtaining relevant data to assess the level of maturity of digital services innovation in the public sector.

4.2 The overview of the Conceptual Data Model The conceptual model is built taking into account the needs of the data collection strategy, namely an orien-tation towards processes, which is, in any case, able to provide information on the service level. A further need to account for is obtaining pieces of information that can be gathered via one of the data collection techniques described in the previous chapter.

The conceptual model directly derives from the network represented in Annex IV. Taking a “process” orien-tation it depicts the conceptual connections between the data level and the conceptual model. It is important to note that the data feeding the processes pertaining to the “innovation governance” type can be re-aggre-gated using, instead of a process perspective, a service one, according to their relation to digital service innovation maturity. This aspect is underlined in the alluvial diagram (see Figure 4.1), where the central column (the data), is disaggregated into the interpretative and cross-cutting scaling mechanism process dimension, and into the two adversarial “lenses”, the “innovation governance” processes one and the “digital service innovation maturity” one.



Figure 4.1 Innovation Governance vs Digital Service Innovation Maturity

The disaggregation according to processes and the one according to services are summarised and dis-played in the Circular Dendrogram of Figure 4.2: the process-related hierarchy is displayed using the branches of the dendrogram tree, while the connection to the service level is shown via the colours of the balls representing the entity. In this visualization, data that feed multiple points of the dendrogram are dupli-cated and identified via a black dot inside.



Figure 4.2 Circular Dendrogram, fusing the process and the service perspective

Starting from the data, i.e., from the set of terminal elements of the hierarchy, it is now straightforward to ascertain where and how this information can be gathered: namely, if this information comes from primary data sources (Survey and focus groups), or secondary data sources (NUT3-resolution secondary data or statistical data downscaling).

Focussing on those pieces of information that can only be acquired via a survey - and being the conceptual data model also a clustering dendrogram able to group the data into thematic entities - the hierarchy allows creating modules of items that can be seen as conceptually homogeneous. This specific grouping can be used to implement modularisation strategies for the survey management and distribution, as well as gamifi-cation strategies aimed at increasing the engagement and motivating the respondent by providing direct feedback about the progress and score. This strategy is implemented to increase the response rate.

Being also a map in terms of conceptual relationships (and thus, of statistical correlation), the hierarchical structure, both in terms of processes and services, will constitute a theory-driven framework to inform the statistical analysis. The technique of choice, in this case, will be Structural Equation Modelling (SEM) (Kaplan, 2008), and specifically the use of measurement models to extract latent factors (which in our case will represent the “scores” of the cities in different categories) from data. Such a measurement model will be refined after the data collection phase via exploratory factor analysis techniques, and then validated via a confirmatory factor analysis.



The creation of a validated measurement model (and connected factor calculation procedures), and so of a statistically validated scale to measure the DPSVI along its dimensions will allow the use of the designed survey as an instrument to evaluate such dimensions also on different cities and to provide them with a quick comparison with the representative sample of cities that we aim to analyse.

Figure 4-3 shows the result of this elaboration, while the same information, shaped in a tabular form, can be found in Annex V.

Figure 4.3 The flow of data: from process typology to specific data

The figure clearly shows two process families representing the starting points of the data collection strat-egy.They refer to processes related to “innovation governance” and to “scaling mechanisms”. As stated in the concept, the latter can be considered as a cross-cutting interpretative category, shared by both the pro-cess and the service perspectives.

These typologies are then articulated in a sub-hierarchy of respective fields of action. From this level, it is then possible to identify the terminal leaves of the conceptual data model, representing the single data units retrieved. Every data unit may inform more than one “position” in the hierarchy. In Figure 4-3, this fact is represented by the stroke thickness attributed to the curves connecting the second level of the fields of action to the data level.

4.3 Logical description of the Data Model The logical data model stemming from the conceptual one is quite simple: our logical data model is consti-tuted by a table, where the rows represent the cities that have responded to the survey. The columns contain the numerical and/or categorical data referred to that specific city: answers to survey items, or data from secondary sources, either at NUT3-resolution, or downscaled. Moreover, after the validation of the factor model, the specific factors’ values will be provided.



Each column will also carry a set of metadata. More specifically, we will provide metadata information about:

● Contextual information about the data field, namely:

o Its source (i.e., primary data, secondary data, downscaled secondary data and, if downscaled, information about the downscaling procedure);

o If a calculated field, information about the computation of the field (e.g., Ratio No. of down-load of data/city population).

● The data type, which is a piece of information necessary to understand the meaning of the value stored in the database, and to proceed to its analysis in a statistically sound way. A first attempt at collecting and classifying data values is provided in Annex V, while a provisional list of data types that will be used as metadata categories is:

o Binary – representing a binary outcome (Questions with Yes/No Answers);

o Categorical – representing categorical outcomes (Questions with one or more qualitative answers);

o Ordinal – a number representing a relative score (Questions with Low/Medium/High an-swers, e.g., Likert scales);

o A count (How many downloads of an app, how many employees in R&D);

o Real-valued – a real number carrying absolute information (A cardinal number);

o Compositional – Questions with one or more percentages as answers.

● Information about the column position in the hierarchy, as provided in Annex V.

4.4 Definition of the sample of analysis As requested by the ToR of DIGISER, this inception report includes a first proposal of the sample of cities that will be researched and surveyed during the next months to collect all the data necessary to assess their DPSVI.

It is important to highlight that the choice to proceed with a limited sample is justified by practical constraint and the same experimental profile of DIGISER project, where the development of an innovative assessment framework and method requires to be validated during the project.

For this reason, the definition of the sample of cities is required to target the best level of representativeness attainable with respect to the variety of urban typologies of Europe.

In summary, DIGISER will cover two different samples of cities:

● A large sample of cities that will be researched combining secondary data available with primary data collected through a questionnaire ( in Task 2);

● A small sample of 10 cities (subgroup of the first larger sample) for which an in-depth case study analysis will be delivered through interviews and focus groups involving institutional actors and local stakeholders (case study cities, researched in Task 4).

As explained in the data gap analysis (see Chapter 2), most of the data included in DIGISER´s data model will be collected through a survey, due to the lack of existing data at the city scale.

In addition to primary data directly collected during the research, DIGISER aims to use and integrate several secondary data mapped from multiple sources, including international institutions and agencies such as the EC, Eurostat, UN, OECD, FAO, WB and ESPON itself. Over the years, these institutions adopted different conceptualizations and spatial definitions of “city”, that in some cases have a relevant impact on the possible integration and use for comparative purposes. Indeed, the account of core contextual data as the overall population or the geographic extension of a given city can vary significantly from one definition to another.

The definitions that DIGISER will consider are:

● NUTS – Nomenclature of territorial units for statistics, is a geographical nomenclature subdi-viding the economic territory of the European Union (EU) into regions at three different levels (NUTS 1, 2 and 3 respectively, moving from larger to smaller territorial units). Above NUTS 1, there is the 'national' level of the Member States. (Eurostat Glossary 2021). None of these definitions could be directly applicable to a “city”, but in a few cases DIGISER will use data referred to NUTS3 and NUTS2 either for defining broader tendencies or for counterfactual analysis purposes.



● LAU – local administrative units are used to divide up the territory of the EU for the purpose of providing statistics at a local level. They are low level administrative divisions of a country below that of a province, region or state. Not all countries classify their locally governed areas in the same way and LAUs may refer to a range of different administrative units, including municipalities, com-munes, parishes or wards. administrative for reasons such as the availability of data and policy implementation capacity (Eurostat Glossary 2021).

● C – Cities (Eurostat) - City corresponds to a local administrative unit (LAU) where the majority of the population lives in an urban centre of at least 50 000 inhabitants (Eurostat, 2021)

● GC – Greater Cities (Eurostat) is an approximation of the urban centre when this stretches far beyond the administrative city boundaries (Eurostat, 2021).

● MR – Metropolitan regions are NUTS 3 regions or a combination of NUTS 3 regions which rep-resent all agglomerations of at least 250 000 inhabitants.

● FUA – Functional urban areas consist of a city and its commuting zone. Functional urban areas therefore consist of a densely inhabited city and a less densely populated commuting zone whose labour market is highly integrated with the city (OECD, 2012) (Eurostat, 2021). FUA is a broader definition that could include those of LAU, C, and GC.

● UC – Urban Centre (version 2015). “The Urban Centres are defined by specific cut-off values on resident population and built-up surface share in a 1x1 km global uniform grid. The input data it is generated by the GHSL, and the operating parameters are set in the frame of the “degree of ur-banization” (DEGURBA) methodology.” (LORDI, 2020)

● CR – Urban Centre Centroid and Radius. “These spatial units will be produced by ESPON. They will be built up based on the centroids of Urban Centres Database and the radius will be calculated using the Minimum Bounding Geometry functionalities in ArcGIS/PostGIS/OpenJump.” (LORDI, 2020)

● C – Ambiguous “City” (or sum of cities – see urban centre database). “In many of datasets, it is impossible to understand what - [insert any city name here] - means geographically. Furthermore, in some cases, the spatial definition is also not relevant, or can be relative, change quickly in time and difficult to trace” (LORDI, 2020).

● Z – Postal codes

In detail, the adoption of the principle of the ‘right scale’ distinguishes between the city definition applied to primary data collected and the city definitions used to standardize secondary data sources.

I. The primary data collected through the DIGISER survey will focus mainly on information related to the governance of innovation mechanisms at the level of the urban authority. In this sense, it is possible to establish a direct correlation between the data and the administrative boundaries of the city. In principle primary data related to respondent cities above 50.000 inhabitants will be referred to the Eurostat category of City (C), while for smaller cities the data related to the LAU will be used. Anyway, the survey will include a subset of core contextual information that will ask to validate the boundaries and the scope of the administrative power of the urban authority observed.

II. Differently, secondary data will be standardized depending by their source itself, according to the city definition to which they have been referred at the moment of their collection and elaboration, encompassing eventually any of the categories aforementioned. In particular it is possible to further distinguish secondary sources between Statistic Data and Big Data:

A. Secondary statistical data are used in DIGISER both to define the sample of cities to be surveyed (and its statistical relevance) and to describe the main variables that shape the context of the action of urban authority as for example: spatial extension, population, gen-der, average GDP, age dependency ratios. In this case, DIGISER will principally refer to the data collected and reported in EUROSTAT under the typology of Cities (C) and – for the cases below 50.000 inhabitants – of the related LAU.

B. DIGISER will also make use of other kind of secondary data series related to specific phenomena, including big data and other kind of geo-referenced data that requires to be aggregated under spatial criteria, as for example data related to the availability of WIFI hotspots in the city space, or the data related to the access to broadband in the city. In this case each data series will be standardized according to the city definition used by the source. When a choice will be possible, DIGISER will favour administrative definitions (C and LAU) over other definitions.

The successive updates of the data model presented in annex V will provide detailed information regarding the source and the actual city definition that has been used for each secondary data series mapped and integrated in DIGISER.



4.4.1 Large sample The large sample of cities has been composed mixing three criteria.

First, statistical criteria have been used, classifying cities and reconstructing a sample that could be rep-resentative of the European variety and distribution. These criteria, applied in the order here listed are:

● The sample should include cities of all sizes (according to OECD-EC classes), trying to counter-balance the bias toward large cities and metropolis that characterize most of the studies on public service innovation and digital transformation;

● The overall sum of the inhabitants of the cities of the sample should account for at least 10% of the total EU population;

● The sample should avoid cities within the same province (NUTS3 level) that are geographically close; and

● The sample should cover all European countries of the ESPON space (European Union plus the UK, Switzerland, Norway, Iceland, Liechtenstein), proportionally to each country’s population (i.e., larger countries equals more cities in the sample), with at least one city for each country of the ESPON space. For smaller countries with only one city listed the Capital is included in the sample.9

Second, practical and operational criteria have been used to select a sample that could ensure the ex-istence of a significant set of data reusable and foster the engagement strategy for the survey. Indeed, the sample prioritizes those cities that belong to city networks and organisations related to the promo-tion of digital transformation and public sector innovation, including all those networks indicated as relevant by the members of the EAG, and in particular:

● Living-in.eu / ‘Join, Boost, Sustain’ movement, https://www.living-in.eu/;

● OASC, https://oascities.org/;

● ICC, https://www.intelligentcitieschallenge.eu/;

● Eurocities, https://eurocities.eu/;

● EIP-SCC Smart city marketplace, https://eu-smartcities.eu/.

● ENoLL; https://enoll.org/

● GDC; http://www.greendigitalcharter.eu/tag/gdc

The cities belonging to these networks are 597 and are presented in the following map, where larger bubbles mean the participation of a single city to more than one network.

9 At the current stage of the research, cities from Candidate Countries (CC) are not considered in this sample. The re-search team will consider the possibility to extend the survey to cities in countries outside of the ESPON space once the first round of data collection will be completed and assessed, and according to the availability of basic secondary data from CC (that are rarely included in EUROSTAT researches as well as other relevant data sources used to complement primary data in the data model of DIGISER).



Map 4.1 Cities belonging to relevant networks

Third, flexibility criteria are adopted in order to adapt the sample to the actual degree of feedback received by the cities involved in the survey. Indeed, for several reasons (see also Chapter 5.4 regarding the engage-ment of cities) it is likely that the initial sample will require updates and integrations along the data collection process, with the possibility to substitute/increment the number of cities involved. In this sense the research team will periodically assess the actual integrity of the sample of respondents, adopting appropriate measures to ensure quantitative and qualitative consistency to the statistical criteria.

In order to define a first hypothesis of the sample in practice, the formerly listed statistical criteria have been applied to the list of cities extracted from relevant networks, complementing it with additional cities, in order to reach an approximate level of statistical representativeness.

The selection amounts to 170 cities and is presented in Map 4.2, and in detail in Annex II.



Map 4.2 Large sample of cities

This target sample can be considered as a sort of “Minimum Viable Sample” (MVS) that is required to deliver a proper comparative analysis capable of providing a picture of the main tendencies and patterns taking place in European cities.

4.4.2 Case Studies Among the sample composed by 170 cities, a smaller group of 10 cities (case studies cities) will be selected for a deeper research, using both quantitative (exploring additional data series available at local level) and qualitative methods (through interviews with public services and other key local stakeholders). This process will allow us to collect detailed and multi-faceted additional data and evidence that will enable us to create deeper insights about the impact of digital innovation.

The preliminary list of the cities composing the 10 case studies meets the following criteria:

● Size: It covers cities of all sizes (according to OECD-EC classes);

● Geographic coverage: It includes 2 cities per region (north, south, east, west and central);

● Network of contacts: It includes only cities in which the consortium has a good network of contacts to facilitate the interviews and the data collection.

Therefore, the suggested list of case studies cities is the following: Darmstadt (Germany), Helsinki (Finland), Ljubljana (Slovenia), Luxembourg (Luxembourg), Milan (Italy), Porto (Portugal), Rotterdam (Netherlands), Stockholm (Sweden), Suceava (Romania) and Thessaloniki (Greece).



5 Future Work

5.1 Survey development, testing and deployment Once a first hypothesis of the data model has been defined in this report (Chapter 4), the following step will focus on the development and testing of a consistent survey aimed at collecting the primary data that repre-sent a majority portion of the DIGISER database under construction, and then proceed to a massive distri-bution targeting the entire large sample.

The upcoming steps will cover the following points:

● Questions' development and translation: in order to collect the data described in the data model (see the labels and the detailed description of data in Annex V), it is necessary to develop clear and unambiguous questions, that will be translated into different languages for distribution of the questionnaire on a European scale. At the moment, it is planned to translate the survey in English, French, German, Spanish, Italian. Eventually, also a small glossary could be distributed along with the survey, in order to reduce risks of misunderstanding in the interpretation of the language used.

● Organisation of the survey in modules: the amount of primary data listed in the data model collides with the practical need to create a questionnaire that should be filled in in a limited and reasonable amount of time by a city official or a civil servant. For this reason, the survey will be organised in different modules that can be filled out either all together or one at a time, following an engagement strategy for successive waves. The organisation of the modules must be structured in such a way as to allow partial analyses and an assessment relating to each module, for both re-search and engagement purposes.

● Choice of the appropriate digital survey tool: digital tools for the distribution and collection of the questionnaires are being evaluated. The characteristics identified require that the data collec-tion tool must:

○ guarantee compliance with the provisions of the GDPR;

○ offer the greatest number of engagements features possible, including the possibility to track respondents, send emails and reminders, keep track of the compilation of the differ-ent modules;

○ have a user-friendly interface that can cover the variety of the type of questions that are expected to be developed (I.e., Y / N, scalar, compositional, multiple questions, matrix, etc.).

● Test with voluntary cities and data model correction / validation: once a first draft of the survey will be developed, this will be tested on a limited number of cities, to be identified together with the stakeholders of the SAG. The test will be a fundamental moment that will provide realistic feedback on the time required for the compilation, on the clarity of the questions developed, and on the actual attainability and availability of data requested. The feedback from the test will be used to correct and consolidate the questionnaire before its large-scale distribution.

5.2 Survey distribution and engagement strategy

5.2.1 Large scale engagement strategy To achieve the ambition of DIGISER to collect 170 samples from cities and communities in the EU and EFTA Countries, a measured strategy for large-scale distribution is being developed. The foundations for the dis-tribution of the DIGISER survey, more appealingly operationalised under the DIGIsurvey brand (www.digi-survey.org) going forward, are:

● an informational campaign across large-scale city networks;

● the cooperation and engagement of key stakeholders through the Strategic Advisory Group (see Chapter 1).

DIGISER has the advantage of being able to execute its data collection operations on a sound footing of a mature, curated, comprehensive, and federated contacts database of cities and communities enabled by the coordination of data assets brought in by consortium members. OASC in particular has an extensive “high quality & high response” distribution system comprised of a best-of class, AI enabled cloud platform, and a



database of key actors in European cities, which will serve as the backbone to be complemented by Con-nected Places Catapult, POLIMI, and Deloitte. Furthermore, DIGISER can rely on a large extended network of operationally linked entities. These include OASC’s many strategic partners and initiatives, including but not limited to Eurocities, the 100 Intelligent Cities Challenge, EIP-SCC and the ‘Join, Boost, Sustain’ move-ment.

Even with a survey distribution relying on the activation of a powerful extended network and federated con-tacts database, the key challenge, among others (see Table 5-1 on Potential Risks), for DIGISER lies in the successful engagement of sufficient stakeholders to complete the survey completely. Furthermore, measures are provisioned to ensure response quality where needed. The number and quality of responses constitute ‘the’ critical success (and risk) factors for the project and will be diligently managed.

To engage cities adequately and to ensure high response rates with good quality responses, DIGISER will apply a clever execution strategy for distribution, engagement, retention, and monitoring over the data col-lection period (see Figure 5.1).

Figure 5.1 The stepwise strategy

The distribution phase will start by booting up the operational engine inherent to the consortium, and by synchronously activating the extended network. The survey will be hosted on a dedicated website (digisur-vey.org) to align with branding and to facilitate a single-entry for data collection.

DIGISER will provide the amplifying networks with precise and polished communication packages, including

● A DIGIsurvey instructional package

● An email template (including invitation to the ‘DIGIsurvey Clinics’),

● Audio-visual material explaining the purpose and relevance of the survey in an original and enter-taining way

Leveraging the analytical capabilities of the selected survey tool, customised reminders will be sent in the engagement phase. In addition, 3 so-called ‘DIGIsurvey Clinics’ will be organised over the whole duration of the data collection period. The clinics are live online webinars of 60 minutes that will provide cities with the opportunity to ask questions about the survey, both regarding its relevance and scope as well as diffi-culties cities are experiencing while to the survey. The survey clinics serve four purposes:

a) they clarify any outstanding questions, issues, and doubts, leading to better responses and higher completion rates;

b) they provide a unique opportunity for cities to engage directly and relevantly with ESPON and DIGISER;



c) they provide a rare opportunity for cities to interact and compare notes on what is generally con-sidered to be a niche activity;

d) they provide direct feedback to DIGISER that can lead to immediate improvements.

Table 5.1 Potential risks

Potential Risk Likelihood Mitigation Measure

Only a small percentage of cities contacted are responding to the invitation to fill in the survey

Moderate Clear communication about the purpose of the sur-vey with appropriate follow-up and reminder mecha-nisms. Furthermore, the survey will be oriented to-wards the stakeholders (see chapter 4)

Cities start responding to the sur-vey without completing it

Moderate User-friendly and modular creation of the survey based on feedback from the test cities. In addition, so-called survey clinics, i.e. support webinars, will be arranged to provide support for cities.

Not all cities of the proposed sample of 170 can be contacted as contacts are missing

Low The sample was composed in a comprehensive manner, taking into account the strengths and weak-nesses of the outreach capabilities of DIGISER and associated partners. Should, nevertheless, a city cannot be contacted, adequate replacement can be ensured as DIGISER has a ‘back-up’ list of cities

Cities take longer than the fore-seen to respond

Moderate Analysis of received contributions can start earlier in coherence with the modular approach of the survey structure.

In addition to the overall strategy, DIGISER will ensure to apply state-of-the-art techniques such as “modu-larization” (Chapter 4) and gamification methods (see the following paragraph 5.2.2) to ensure successful data collection.

5.2.2 Gamification as strategy to keep survey users engaged The engagement strategy will benefit from the introduction of a light gamification approach (Deterding et al., 2011) aimed at sustaining the participation by providing motivation and feedback about the progress in filling the data and the outcomes of the compilation.

Gamification can be defined as the intentional use of game elements for making non-game tasks and activ-ities gameful experiences (Huotari and Hamari, 2012). Gamifying a process means enhancing it with gaming affordances that make it a gameful experience while supporting user’s value creation, increasing engage-ment and reinforcing retention. The substantial benefits that a strategic introduction of gamification can bring made the practice gain significant and growing attention in the years, in various fields. For the purpose of this study, game thinking and game mechanics are included in the measure of being a source of constant motivation (Zichermann and Cunningham, 2011), providing encouraging feedback on the state of the pro-gression. In this regard, key are the concepts of satisfaction and gratification, deriving from enjoying the results and outcomes of the actions completed. In each step of the process, the gamified system increases the perceived value coming from providing information on the relation between the municipality and digital innovation.

In consequence, it becomes fundamental to carefully design a system of levels, scores and progression metrics aimed at accompanying the compiler(s) of the questionnaire in the municipality, and a set of profiles which are a synthesis, assessing the behaviour and attitude of a municipality towards digital innovation and its processes. In order to build a gamification system as a holistic experience supporting the entire assess-ment, each of the elements introduced serves a threefold function: engaging and motivating while informing.

Moreover, the activity triggers a learning process deriving from the process of compiling the questionnaire. The structure of the questionnaire, its modules, and its questions lead the compiler to activate a series of



fundamental reasonings regarding what it means digital innovation and what means being digitally mature. In parallel, indirect information on possible trajectories and digital possibilities also emerges from an attentive analysis of the items/questions composing the survey. Adopting an approach inspired to “game procedural-ity” (Bogost, 2007), by means of which learning comes as a result of the user activities, allows to reinforce understanding through systemic expressive responses. Profiles and scores are examples of expressive re-sponses, which can create direct benefits in terms of positive competitiveness: a comparison between mu-nicipalities can suggest trajectories of future development, encouraging improvements and actions towards digital innovation.

5.3 Data analysis and visualisation methods: geospatial visualisation and scores visualisation The detailed methods for the analysis and visualisation of the collected data will be defined in the Interim Report. Nonetheless, at the current stage it is possible to envisage the family of statistical methods that will be used in the data processing and analysis, and several possible ways to display the gathered data.

In terms of data analytics, the peculiarities of the analytical process - namely the presence of a strong theo-retical model behind the analysis, and the need to perform a thorough effort of data integration- call for the use of statistical methods that are flexible, yet able to provide a very robust validation of the complex theoretical constructs that are employed in the analysis. The way to go in this case is to use Structural Equation Models (SEM), a rich family of methods, widely used in the psychological and social sciences as a way to analyse and validate questionnaire scales, which also provide a very rich toolkit for the analysis.

The use of this kind of methods allows the extraction of features that are “latent” with respect to the acquired data and will allow the formulation of quantitative “profiles” based on such latent “scores”, representing the dimensions present in the conceptual model. The hypothesis under development is to adopt a radar-plot as a visualization method capable to represent the profile of each city, based on the “score” achieved10. Con-sistently with the organisation for modules of the survey, the radar plot method would also allow partial visualizations of the score achieved in each module filled. Having these scores, it is then possible to provide radar-plots that represent the profiles in a graphical way.

In terms of further data visualisation possibilities, information about profiles, latent scores but also infor-mation at the raw data level can be visualized by abiding to ESPON standards and templates to generate georeferenced maps that enable an overview of the phenomena analysed in DIGISER on a European scale.

Moreover, we can provide charts and maps that are able to offer an easily interpretable interpretation of each single case that is part of the city sample, allowing each city to receive immediate feedback regarding its performance (either on the whole survey or on one of its modules).

5.4 10 Case Studies - Overview After the list of the 170 cities sample and the cities composting the 10 case studies are finalised, together with the selection of the scenarios of data collection presented in chapter 4.1, the following steps will aim to complete the information gaps from the data model. This will be done through the collection of new infor-mation and evidence that will enable to create deeper insights about the impact of digital innovation and will ensure a better understanding of the future consequences regarding the importance to better capitalise knowledge by producing 10 case studies.

The following represents a starting point of this task and may be adjusted according with the progress and findings of the project. The upcoming steps will cover the following points:

● Prepare data collection:

○ Identify the data gaps from the survey, this will mainly focus on qualitative data, and can focus, for instance, on the following topics: i) Roles, strategies, and approaches of the public sector in the EU; ii) Type of organisational transformation that is needed to reshape the public sector; iii)

10 A reference and an inspiration for data visualization through radar plot is the “Going Digital Toolkit” developed by OECD - https://goingdigital.oecd.org/.



Roles of digital innovation, openness, co-creation, citizens, businesses, data, interoperability, platforms, technology, etc.; iv) Vision about what constitutive elements of Public Value and how the public sector can steer digital transformation toward the creation of Public Value; v) Interpret the main territorial preconditions, driving forces, and bottlenecks; vi) Insights regarding how the scaling up of digital innovation and services can benefit both public authorities and the innovation ecosystem.

○ Identify the main city stakeholders, such as: ■ Decision Makers that will provide a macroeconomic framework of the city. Examples of

stakeholders, include: Government, Policy Makers, City Mayors;

■ Service Provider, who will share insights of the management of the solution. Examples of stakeholders include: Head of Digital Strategy, IT managers, Chief Information Officers;

■ Solution Developers, responsible to share insights that led to the development of the solu-tion. Examples of stakeholders include: Public Operators, Private Operators, Academia;

■ Users, who will share feedback as end-users of the service and solution. Examples of stake-holders include NGOs, Representatives of citizens, Citizens.

For this stage, the strategic advisory group, and the signatories of the living-in.EU declaration will support to establish the contacts and validate the information to be collected.

○ Definition of the tools to be used for data collection: One-to-one interviews, workshops/ focus groups, questionnaires. The different tools will be used according to the type of stakeholders (e.g. with the decision-makers, priority will be given to one-to-one interviews and questionnaires).

After identifying the data gaps, and the tools to apply to the stakeholders, the interview’s script, and supporting materials (interviews invitations, presentations, brochures, etc.) will be prepared.

● Data Collection:

○ Prepare the invitations and schedule the interviews/ workshop/ focus group – This task will be done in coordination with the Project Officer. The invitations will propose specific time slots and provide additional guidelines to ensure that participants are well prepared.

○ Organise and held the interviews and workshops:

■ Focus on the decision-makers: online interviews with decision-makers, as they are the key players on the governance structure of the public service. The questionnaire will be sent in advance for decision-makers to prepare and gather information beforehand. At least 2 members of the team will conduct the interview – one coordinator and one supporter. In order to capture all the meeting notes, and in compliance with the GDPR, interviews will be recorded.

■ Focus on other city stakeholders: online questionnaires and workshops/ focus groups with other city stakeholders (service providers, solutions developers, and users). The at-tendees of the workshops/focus groups will be assembled by the dimensions of the analysis in order to better capture relevant insights about the target topic to be discussed. Moreover, the tools (questionnaires or workshops) to be used under this group will also depend on the list and number of stakeholders per city (it will be analysed case by case).

This type of approach will provide an in-depth understanding of specific topics regarding the complex objectives and questions that we commit to answer by providing interviewees and participants with materials and/or ideas that are immediately useful in the interview or work-shop.

A template will be also prepared to facilitate the collection of data during the process.

● Apply analytical tools - This activity will merge, analyse and present the preliminary findings of the quantitative data (from the survey and data model) and the qualitative data collected under the “Collect data” stage. Spatial analytical tools to quantify patterns and relationships in the data and display the results as charts and maps in a visual format, as well as correlation analysis tool to compare the relationship between different sets of data (from the data model and interviews, ques-tionnaires, workshop/ focus group). This method will enable the team to compare the results across the 10 cities. Based on the outputs, a case study city profile for each study will be reported (the format will be fine-tuned according to the outcomes of Developing policy recommendations).



5.5 Developing policy recommendations The research team is committed to developing “concrete and workable mid-term policy recommendations for digital, structural and organisational transformations of public administrations and its service provision at local, national and the EU levels” under Task 5 of the DIGISER project. We proposed to achieve this objec-tive through three broad steps:

● Undertaking a review of current policies concerning digital innovation,

● Reviewing and synthesising the findings of Tasks 2-4 to identify their implications for policy

● Developing the policy recommendations.

At this stage of the research project, a considerable part of the relevant policy review has been carried out under the current deliverable (D1). Over the coming months, as the team makes progress towards D2 and subsequent deliverables, we will be:

● Scrutinising the outputs from Tasks 2-4 to determine what these results imply for policy on DIGISER,

● Developing insights on what areas would benefit from policy guidance and recommendations,

● Fine-tuning our understanding of current policies with respect to policy gaps, what policies have worked well, where policies need improving – their strengths and weaknesses, and

● Developing a draft of recommendations.

5.6 Concluding remarks With this report, the project team had the ambition to provide an appropriately detailed overview of the critical foundations, learnings, directions, and plans of DIGISER with particular emphasis on the aspects befitting the current project phase and state of play. At the same time, we have paid particular attention to achieving a degree of detail across the board that allows reviewers and advisors to ascertain and evaluate the overall quality, coherence, effectiveness, and appropriateness. We look forward to feedback and to the next stages of the project.



6 Annexes ● Annex I Members of the Strategic Advisory Group & Workshop Minutes

● Annex II City sample

● Annex III Scenarios of data collection

● Annex IV Policy context

● Annex V Conceptual Data Model Table

● Annex VI Datasets from existent relevant studies



References Albury, D. (2005). Fostering Innovation in Public Services. Public Money & Management 25 (1), 51-56.

Argyris, C., & Schön, D. A. (1978). Organizational learning: a theory of action perspective. Massachusetts: Addison-Wesley Pub. Co.

Avgerou, C. (2000). IT and organizational change: an institutionalist perspective. Information Technology & People 13 (4), 234-262.

Avritzer, L., & Boaventura de Sousa, S. (2005). Opening Up the Canon of Democracy. In B. d. Santos, Democratizing democracy: beyond the liberal democratic canon. London: Verso.

Bateson, G. (1973). Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution and Epistemology. London: Jason Aronson Inc.

Bogost, I. (2007). Persuasive Games: The Expressive Power of Videogames. The MIT Press.

Bugge, M. M., Fevolden, A. M., & Klitkou, A. (2019). Governance for system optimization and system change: The case of urban waste. Research Policy 48 (4), 1076-1090.

Campbell, T. (2012). Beyond Smart Cities. How Cities Network, Learn and Innovate. Abingdon: Earthscan.

Chilvers, J., Pallett, H., & Hargreaves, T. (2018). Ecologies of participation in socio-technical change: The case of energy system transitions. Energy Research & Social Science 42, 199-210.

Concilio, G., & Molinari, F. (2019). The unexploitable smartness of open data. Paper presented at the Annual Aesop Conference. Venice 9-13 July.

Concilio, G., & Pucci, P. (2021 furthercoming). The data shake: an opportunity for experiment driven policy making. In G. Concilio, P. Pucci, L. Raes, & G. Mareels, The Data Shake. Opportunities and obsta-cles for urban policy making. Springer International Publishing - PoliMI SpringerBriefs.

Concilio, G., Pucci, P., & Raes, L. (2021). The Data Shake. Opportunities and obstacles for urban policy making. Springer International Publishing - PoliMI SpringerBriefs.

Deserti, A., & Rizzo, F. (2019). Embedding Design in the Organisational Culture: Challenges and Perspec-tives. In M. N.-C. Guy Julier, Design Culture Objects and Approaches (pp. 39-51). London: Blooms-bury.

Deterding, S., O’Hara, K., Sicart, M., & Nacke, L. (2011). Gamification: Using Game Design. Elements in Non-Gaming Contexts. Extended Abstracts on Human Factors in Computing Systems, 2425-2428.

Economist Intelligence Unit. (2010). ICT for City Management. Using information and communications tech-nology to enable, engage and empower city stakeholders. Singapore: The Economist Intelligence Unit.

Edquist, C., & Zabala-Iturriagagoitia, J. M. (2015). Pre-commercial procurement: A demand or supply policy instrument in relation to innovation? R& D Management 45(2), 147-160.

Elliott, I. C. (2020). Organisational learning and change in a public sector context. Teaching Public Admin-istration 38 (3), 270-283.

Emaldi, M., Aguilera, U., López-de-Ipiña, D., & Pérez-Velasco, J. (2017). Towards Citizen Co-Created Public Service Apps. Sensors (Basel) 17 (6), 1265.

Engelbart, D. C. (2002). Improving our Ability to Improve: A call for investment in a new future. Singapore: World Library Summit.

European Commission. (2018, May 08). European Commission. Retrieved from Mission-oriented research and innovation. Inventory and characterisation of initiatives: final report: https://op.eu-ropa.eu/en/publication-detail/-/publication/3b46ce3f-5338-11e8-be1d-01aa75ed71a1/language-en

European Commission. (2020, September 14). European Commission. Retrieved from A multi-dimensional framework to evaluate the innovation potential of digital public services. A step towards building an innovative public services observatory in the EU: https://op.europa.eu/es/publication-detail/-/publi-cation/cd56ca1d-f6fc-11ea-991b-01aa75ed71a1



European Commission. (2021, January). DESI - Digital Economy and Society Index. Retrieved from Euro-pean Commission: https://ec.europa.eu/digital-single-market/en/digital-economy-and-society-in-dex-desi

European Commission. (2021, January). Digital Economy and Society Index (DESI) 2019. Retrieved from European Commission: https://ec.europa.eu/digital-single-market/en/news/digital-economy-and-society-index-desi-2019

Eurostat. (2021, January). Category: Regions and cities glossary. Retrieved from Eurostat statistics ex-plained: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Category:Re-gions_and_cities_glossary

Faro, B., Babak, A., & Cetindamar, D. (2019). Continuous Transformation of Public-sector Organisations in the Digital Era. American Conference on Information Systems.

Fernandez, S., & Rainey, H. G. (2006). Managing Successful Organizational Change in the Public Sector. Public administration review 66 (2), 168-176.

Ferro, E., Caroleo, B., Leo, M., Osella, M., & Pautasso, E. (2013). The Role of ICT in Smart Cities Govern-ance. In P. Parycek, & N. Edelmann, Conference Proceedings: CeDEM13 Conference for E-De-mocracy and Open Government. Edition Donau-Universität Krems.

Fung, A. (2006). Varieties of Participation in Complex Governance. Public Administration Review 66 (s1), 66-75.

Gauk, M., Emidio, A., Brynskov, M., Voutilainen, O., & Maijala, O. (2019). ESPON REPORT - Digital inno-vation in urban environments. ESPON.

Gauk, M., & Frideres, L. (Eds.). (2017). ESPON REPORT - The territorial and urban dimensions of the digital transition of public services. ESPON.

Geels, F. W. (2020). Micro-foundations of the multi-level perspective on socio-technical transitions: Devel-oping a multi-dimensional model of agency through crossovers between social constructivism, evo-lutionary economics and neo-institutional theory. Technological Forecasting and Social Change 152, 119894.

Geels, F. W., & Schot, J. (2007). Typology of sociotechnical transition pathways. Research Policy 36 (3), 399-417.

Gupta, J. (2016). Climate change governance: history, future, and triple-loop learning? Wiley Interdiscipli-nary Reviews: Climate Change, 7(2), 192-210.

Hansen, M. T., & Birkinshaw, J. (2007). The innovation value chain. Harv Bus Rev 85 (6), 121-30, 142.

Hinings, B., Gegenhuber, T., & Greenwood, R. (2018). Digital innovation and transformation: An institutional perspective. Information and Organization 28 (1), 52-61.

Huotari, K., & Hamari, J. (2012). Defining gamification: a service marketing perspective. 12: Proceeding of the 16th International Academic MindTrek Conference, (pp. 17-22).

Ibarra, P. (2007). Participación y poder : de la legitimación al conflicto. Democracia participativa y desarrollo humano, 33-53.

Johannessen, Å., & Wamsler, C. (2017). What does resilience mean for urban water services? Ecology and Society 22 (1).

Kahane, A. (2007). Solving Tough Problems: An Open Way of Talking, Listening, and Creating New Reali-ties. Berrett-Koehler Publishers.

Kemp, R., & Loorbach, D. (2006). Transition Management: A Reflexive Governance Approach. Reflexive Governance for Sustainable Development, Cheltenham, UK and Northampton, MA, USA: Edward Elgar.

Krogstie, J. (2006). Citizens, from consumers to prosumers: e-Government Services typologies revisited. Workshop at Nordichi 2006: User involvement and representation in eGovernment Projects.

Loorbach, D. (2010). Transition Management for Sustainable Development: A Prescriptive, Complexity-Based Governance Framework. Governance Int. J. Policy Adm. Inst. 23, 161-183.

Loorbach, D., & Rotmans, J. (2010). The practice of transition management: Examples and lessons from four distinct cases. Futures 42 (3), 237-246.

Manzini, E. (2014). Making Things Happen: Social Innovation and Design. Design Issues 30 (1), 57-66.



Markianidou, P., Eaton, D., Shauchuk, P., Koopmans, M., Krylova, P., Gonciarz, G., & Grzywacz, P. (2020). ESPON REPORT - SDG localising tool: Localising and measuring Sustainable Development Goals in cities and regions. ESPON.

Mazzucato, M. (2018a). Mission-Oriented Innovation Policy: Challenges and Opportunities. Industrial and Corporate Change, 27 (5), 803-815.

Mazzucato, M. (2018b). Mission-Oriented Research & Innovation in the European Union. A problem-solving approach to fuel innovation-led growth. Brussels: European Commission. Directorate-General for Research and Innovation.

Miedzinski, M., Mazzucato, M., & Ekins, P. (2019). A framework for mission-oriented innovation policy roadmapping for the SDGs: The case of plastic-free oceans. UCL Institute for Innovation and Public Purpose, Working Paper Series (IIPP WP 2019-03).

Moore, M.-L., & Riddell, D. J. (2015). Scaling out, Scaling up, Scaling deep: Advancing systemic social innovation and the learning processes to support it. Journal of Corporate Citizenship 58.

OECD. (2017). Government at a Glance 2017. Paris: OECD Publishing.

OECD. (2019). Government at a Glance 2019. Paris: OECD Publishing.

Omann, I., Kammerlander, M., Jäger, J., Bisaro, A., & Tàbara, J. D. (2019). Assessing opportunities for scaling out, up and deep of win-win solutions for a sustainable world. Climatic Change 160, 753-767.

Plesner, U., Justesen, L., & Glerup, C. (2018). The transformation of work in digitized public sector organi-zations. Journal of Organizational Change Management 31 (5), 1176-1190.

Poole, M. S., & Ven, A. H. (2004). Handbook of Organizational Change and Innovation. Oxford University Press.

Rammel, C., Hinterberger, F., & Bechthold, U. (2004). Govern-ing Sustainable Development. A Co-evolu-tionaryPerspective on Transitions and Change. GoSD workingpaper 1.

Rothbauer, P. M. (2008). Triangulation. The SAGE Encyclopedia of Qualitative Research Methods, 893-895.

Schein, E. H. (1985). Organizational Culture and Leadership. Jossey-Bass Publishers, San Francisco.

Sestini, F. (2012). Collective Awareness Platforms: Engines for Sustainability and Ethics. IEEE Technology and Society Magazine 31, 54-62.

Shove, E., Pantzar, M., & Watson, M. (2012). The Dynamics of Social Practice: Everyday Life and How It Changes. London: SAGE Publications Ltd.

Sinek, S. (2009). Start with Why: How Great Leaders Inspire Everyone to Take Action. Penguin Publishing Group.

Sintomer, Y., Röcke, A., & Herzberg, C. (2016). Participatory Budgeting in Europe: Democracy and Public Governance. Routledge.

Stacey, R. D. (2007). Strategic Management and Organisational Dynamics: The Challenge of Complexity to Ways of Thinking about Organisations. Pearson Education.

Stafford-Smith, M., Griggs, D., Gaffney, O., Ullah, F., Reyers, B., Kanie, N., . . . O’Connell, D. (2017). Inte-gration: the key to implementing the Sustainable Development Goals. Sustainability Science 12, 911-919.

UNDP. (2015). Public Administration Reform Practice Note. UNDP.

Voet, J. V., Kuipers, B. S., & Groeneveld, S. (2016). Implementing Change in Public Organizations: The relationship between leadership and affective commitment to change in a public sector context. Public Management Review 18 (6), 842-865.

Walravens, N., Ballon, P., Compernolle, M. V., & Borghys, K. (2021 forthcoming). Data ownership and open data: the potential for data-driven policy making. G Concilio, P Pucci, L Raes.

Welby, B. (2019). The impact of digital government on citizen well-being. OECD Working Papers on Public Governance, No. 32, OECD Publishing, Paris.

Wyatt, S. (2017). Making Policies for Open Data: Experiencing the Technological Imperative in the Policy World. Science, Technology, & Human Values 42 (2), 320-324.



Zichermann, G., & Cunningham, C. (2011). Gamification by Design: Implementing Game Mechanics in Web and Mobile Apps. O'Reilly Media, Inc.



ESPON 2020

ESPON EGTC 4 rue Erasme, L-1468 Luxembourg Grand Duchy of Luxembourg Phone: +352 20 600 280 Email: [email protected] www.espon.eu The ESPON EGTC is the Single Beneficiary of the ESPON 2020 Cooperation Programme. The Single Operation within the programme is implemented by the ESPON EGTC and co-financed by the European Regional Development Fund, the EU Member States, the United Kingdom and the Partner States, Iceland, Liechtenstein, Norway and Switzerland. Disclaimer This delivery does not necessarily reflect the opinion of the members of the ESPON 2020 Monitoring Committee.

INCEPTION REPORT // DIGISERINCEPTION REPORT // DIGISER ESPON // espon.eu 5 List of maps, figures,...

Documents

Transcript of INCEPTION REPORT // DIGISERINCEPTION REPORT // DIGISER ESPON // espon.eu 5 List of maps, figures,...