First Quarter 2019 A Magazine for Talent Analytics … › assets › Uploads ›...

In This Issue

Talent Analytics Innovations Bullseye

Shaping Analytics to Improve Manager Performance at Infosys

5 Technology Bets for a Big Data Future

Featured Article 4 Predictions for Talent Analytics in the Digital Age

Gig Economy FAQs for Talent Analytics Leaders

In the News3 Questions to Ask Before Implementing Learning Analytics

First Quarter 2019

A Magazine for Talent Analytics Leaders and Data-Driven HR Professionals

Contents

Letter From the Editors 3

Talent Analytics Innovations Bullseye 4

5 Trends Driving Talent Analytics’ Speed and Ease of Use 5

Shaping Analytics to Improve Manager Performance: An Interview With Infosys’ Organizational Development Team 10

5 Technology Bets for a Big Data Future 15

4 Predictions for Talent Analytics in the Digital Age 20

Gig Economy FAQs for Talent Analytics Leaders 25

Ensure Analytics Are Actionable, Credible and Accessible 29

In the News 3 Questions to Ask Before Implementing Learning Analytics 33

Legal Caveat

© 2019 Gartner, Inc. and/or its affiliates. All rights reserved. Gartner is a registered trademark of Gartner, Inc. and its affiliates. This publication may not be reproduced or distributed in any form without Gartner’s prior written permission. It consists of the opinions of Gartner’s research organization, which should not be construed as statements of fact. While the information contained in this publication has been obtained from sources believed to be reliable, Gartner disclaims all warranties as to the accuracy, completeness or adequacy of such information. Although Gartner research may address legal and financial issues, Gartner does not provide legal or investment advice and its research should not be construed or used as such. Your access and use of this publication are governed by Gartner’s Usage Policy. Gartner prides itself on its reputation for independence and objectivity. Its research is produced independently by its research organization without input or influence from any third party. For further information, see “Guiding Principles on Independence and Objectivity.”

Any third-party link herein is provided for your convenience and is not an endorsement by Gartner. We have no control over third-party content and are not responsible for these websites, their content or their availablility. By clicking on any third -party link herein, you acknowledge that you have read and understand this disclaimer.

190064

EditorsCaitlin DutkiewiczBlakeley HartfelderAndrew Kim

Authors Matthew Dong Karthik Gopalakrishnan Richard Nguyen Peter Vail Shaileja Verma Jenna Zitomer

Creative

Graphic Designer Nora Boedecker

Editor Melanie Meaders

Talent Analytics Quarterly | First Quarter 2019 2

Letter From the EditorsIn this issue of Talent Analytics Quarterly, we start the new year with a vision of where talent analytics is headed in 2019 and beyond as well as some practical advice and examples to ensure that vision becomes reality. The future of talent analytics continues to be bright, especially as teams successfully leverage new data and technology investments to inform strategic decisions and drive business outcomes. This issue features our predictions for how talent analytics can evolve to deliver value in the digital age — from capturing the employee experience in new ways to helping leaders become better at navigating risk. Talent analytics teams can expect to further cement their position as key digital leaders to business clients over the next year. We also look into how new talent analytics technologies and innovations are being adopted, what their impact is and which innovations will be increasingly important in the coming year. With the growing availability of advanced, complex technologies, talent analytics teams have more opportunities to analyze data in different ways and support business decisions at a faster rate. We also discuss how talent analytics teams can maximize returns on data investments by making prudent technology bets in machine learning, AI, IoT and predictive analytics. We then share several examples of how specific talent analytics teams have started applying new data and technology in their organizations. We hear from the organizational development team at Infosys about a successful talent analytics initiative they implemented to improve manager performance and employee engagement. Thank you for continuing with us as we journey into 2019.

Andrew Kim Caitlin Dutkiewicz Blakeley Hartfelder


Output: 05:46PM Feb 25 2019

Modified 05:45PM Feb 25 2019

Ado

ptin

gEx

peri

men

ting

Embe

dded

Talent Analytics Projects

Statistics Software

Machine Learning Platform

Data Center/Data Warehouse

Prepackaged Analytics Platform

Text Analytics/ Natural Language Processing Software

Model-Building Tools

Data Mining/ Scraping Software

Manager Self-Service

Engagement Pulse Surveys

Talent Data Benchmarking

Data Quality Tools

Mobile Business

Intelligence

Data Discovery Service

Wearable Technology

Data Lakes

Survey-Building and Analysis Tools

Data Visualization Software

HIPO Identification

AttritionPrediction

Integration With Other IT Systems (excluding HRIT)

Ability to Import Data From Outside the Organization

Contingent Labor Optimization

Real-Time Engagement/ Sentiment Tracking

Automated Résumé Screening

Mobile Business Intelligence

Pre-Built Reports and Analytics Dashboards

Predictive Analytics Module

Integration With Other

HRIT Platforms

HR Ticketing/ Operations

Effectiveness

Organization Network Analysis

Benefits Program ROI

Calculator

Organization Design

Optimization

Workforce-Planning Scenario Planning

Compensation Scenario Planning

Real-Time Reporting

Workforce Segmentation

Scheduling Optimization

Virtual Assistants/Chatbots

AI-Based Candidate Assessment

Career Mapping and Management

Labor Market Analysis

Candidate Sourcing and Engagement

Learning Program Evaluation

Diversity and Inclusion Target Planning

Data Visualization Module

Less

Experimenting

Low

Same

Adopting

Medium

More

Embedded

High

Generalist Analytics ToolsHCM Syste

m Features

© 2019 Gartner, Inc. and/or its affiliates. All rights reserved. 190064

Talent Analytics Innovations BullseyeA Map of Adoption, Impact and Future Investment for 45 Talent Analytics Technologies and Innovations

Assessment Factors

Adoption LevelDegree of adoption is measured by the extent and nature of deployment across organizations.

Current ImpactCurrent level of impact is based on the level of value delivered by the talent analytics innovation.

Future InvestmentFuture outlook is based on the projected level of investment in the next two years.

n = 31 talent analytics leadersSource: 2018 Gartner Talent Analytics Innovations SurveyNote: To determine the relative adoption level, current impact and future investment outlook of each attribute on the Talent

Analytics Innovations Bullseye, we used a weighted scoring method. Items are scored relative to one another in each section.

Generalist Analytics ToolsPlatforms, infrastructure and software designed for the purpose of analytics and business intelligence more generally, as opposed to prepackaged applications for talent analytics

Talent Analytics ProjectsTalent analytics workstreams or initiatives used to collect data about employees and candidates or to apply that data to solve business challenges

Human Capital Management (HCM) System FeaturesThe core HR administrative system for collecting data, such as simple demographics, hiring dates and payroll data

Heads of HR agree talent analytics can have an outsized impact on business priorities, especially given the many technological and methodological analytics innovations available to HR teams. Now as you try to get a quick start in 2019, one of the most important decisions you are likely facing is which talent analytics investments to consider. As you begin answering this question, our Talent Analytics Bullseye provides a roadmap for understanding which of the many potential innovations in this space global talent analytics leaders prioritize.

The Talent Analytics Innovations BullseyeTo help you navigate the changing talent analytics landscape and determine where your function should make future investments, we collected data on 45 talent analytics innovations and technologies from talent analytics leaders globally. The results from our survey provide an objective map of how organizations are adopting these innovations, what their current impact is and which ones will be increasingly important in

the future (see our infographic, “Talent Analytics Innovations Bullseye”). We assessed three main categories of talent analytics innovations, which you should consider as you define your talent analytics strategy and needs: • Human capital management (HCM) system

features — The core HR administrative system for collecting data, such as simple demographics, hiring dates and payroll data

• Generalist analytics tools — Platforms, infrastructure and software designed for analytics and business intelligence more generally as opposed to prepackaged applications for talent analytics

• Talent analytics projects — Talent analytics workstreams or initiatives used to collect data about employees and candidates or to apply that data to solve business challenges

Our Analysis: Takeaways for 2019Most talent analytics innovations are still not widely used. In fact, only 16% are well-embedded, and most talent analytics leaders are still uncertain where they should place their bets. However, across all three categories, we noticed a theme: As the tools and methodologies continue to advance, talent analytics leaders are investing first in the innovations that make it quicker and easier for HR and business leaders to use talent data to make informed decisions.

5 Trends Driving Talent Analytics’ Speed and Ease of Useby Peter Vail


The sections below illustrate some noteworthy trends that underscore this theme.

Trend 1: Growing Adoption of Real-Time Reporting in HCM SystemsReal-time reporting produces reports with real-time talent data using HCM systems. It helps talent analytics leaders provide the business with accurate workforce data more quickly. The HCM feature also increases the ease with which HR or business leaders can monitor changes in key talent metrics or quickly confirm hypotheses about their talent. Adoption of real-time reporting is on the rise:• Sixty-five percent of talent analytics

functions use real-time reporting as part of their HCM systems.

• Sixty-nine percent of talent analytics leaders indicate real-time reporting has a high current impact.

• Nineteen percent of talent analytics functions are planning a high investment in real-time reporting in the next two years.

It’s no surprise 84% of organizations either currently use real-time reporting or plan to use it in the future (see Figure 1). The high level of adoption and impact of this innovation make it clear that many talent analytics functions are moving from delivering ad hoc analysis

to becoming on-demand functions to better support business needs in a data-reliant world. Implication for Talent Analytics Leaders: Real-time reporting provides greater flexibility for business leaders to see data when it is needed, but it also makes it harder for talent analytics teams to know how to prioritize what data and analysis business leaders see. Questions to Consider: How can we decide what talent data to proactively provide as updates to business leaders? When is it more effective to allow leaders to self-serve?

Trend 2: Increasing Investment in Prepackaged Analytics PlatformsPrepackaged analytics platforms offer a software framework for distributed storage and big data processing. Talent analytics leaders are expanding their use of these off-the-shelf packages to reduce the learning curve required to provide analytics decision support. Prepackaged analytics platforms make it easier for less experienced data analysts and data scientists to analyze and produce insight based on large quantities of data. Now that many HR organizations have hired this type of talent, it makes sense that investments in prepackaged analytics platforms will continue to increase. In fact, despite a low level of current adoption (only 26% of talent analytics functions):

Figure 1: Deployment of Real-Time Reporting in HCM SystemsPercentage of Talent Analytics Functions

n = 31 talent analytics leadersSource: Gartner 2018 Talent Analytics Innovations Survey

Not Planning to Use Planning to Use Currently Using0%

50%

100%

0%

50%

100%

16% 19%

65%


• Forty-seven percent of talent analytics leaders indicate prepackaged analytics platforms have a high current impact.

• Forty-two percent of organizations expect a high level of investment in prepackaged platforms, and another 47% expect a medium level of investment (see Figure 2).

The benefits of these platforms can extend beyond improving the speed and ease with which dedicated talent analytics professional can analyze data; they also enable other colleagues to more quickly run analyses using talent data. As other functions increasingly use talent data for strategic decision making, an investment in prepackaged analytics platforms helps organizations scale collaboration without the need for talent analytics leaders to increase head count. Implications for Talent Analytics Leaders: Although prepackaged analytics platforms make it easier to run analyses, they also increase the risk that analyses are done quickly and without fully understanding the hypotheses and parameters behind them. Question to Consider: How can I encourage talent analytics professionals to collaborate with others throughout the business to ensure analysis is relevant and not just done efficiently?

Trend 3: High Impact From Data Visualization SoftwareOne of the greatest challenges talent analytics professionals face is presenting their complex data or analysis in a way that is easy for senior leaders to understand and act on. Data visualization is one of the best tools in an analytics professional’s toolbox to quickly and concisely convey the key takeaways for a senior audience. Data visualization software enables teams to create dashboards and visuals to understand and present their data. It enables talent analytics professionals to more quickly perform data visualization, which significantly improves talent analytics’ ability to serve as a key partner for leaders. In fact, 60% of talent analytics leaders indicate data visualization has a high impact on talent analytics, and another 30% indicate a medium level of impact (see Figure 3). This perception contributes to its high adoption rate:

Figure 3: Data Visualization Tools’ Current Impact on Talent AnalyticsPercentage of Talent Analytics Functions

Figure 2: Expected Investment in Prepackaged Analytics Platforms Over the Next Two YearsPercentage of Talent Analytics Functions



Low Investment

Low Impact

High Investment

High Impact

Medium Investment

Medium Impact

0% 25% 50%

11%

47%

42%

0% 50% 100%

10%

30%

60%


• Seventy-one percent of talent analytics functions use data visualization software.

• Forty percent of talent analytics functions are planning a high level of investment in data visualization software in the next two years.

Implication for Talent Analytics Leaders: Unsurprisingly, data visualization software enables leaders to more easily make connections based on the analysis of talent data, but it can also make it difficult for them to understand the context behind the takeaway.

Question to Consider: How can I ensure my team provides business leaders with analysis and key findings without losing the contextual factors that help inform decision making?

Trend 4: Impactful Results From Real-Time Engagement TrackingInnovations in talent analytics methods have made it possible to supplement annual engagement surveys — and even more targeted pulse surveys — with other forms of continuous employee listening. Real-time engagement tracking solutions measure and monitor the mood, culture or level of engagement in an organization using short, frequent employee surveys. However, while nearly all organizations report tracking engagement, few do so in real time:

• Only 13% of talent analytics functions use real-time engagement tracking.

• Only 21% of talent analytics functions are planning a high level of investment in real-time engagement tracking in the next two years.

Although organizations are only scratching the surface of ways to ultimately use new forms of engagement data, those that have been investing in real-time engagement tracking have seen a quick impact. In fact, 58% of talent analytics leaders report a high impact of these tools, and another 37% indicate a medium impact (see Figure 4). Implication for Talent Analytics Leaders: Many organizations already have talent data sources that can help build a well-rounded, more real-time view of engagement. Questions to Consider: Could we collect existing talent data to supplement our current approaches to measuring employee engagement? How could we more quickly get a pulse on our employees?

Trend 5: Positive Outlook for AI-Based Candidate AssessmentTalent analytics functions are beginning to experiment with advanced technologies, such as AI, that can make it easier for leaders to make talent decisions. This trend is particularly

Figure 4: Real-Time Engagement Tracking’s Positive Impact on Talent AnalyticsPercentage of Talent Analytics Functions


Low Impact Medium Impact High Impact0%

50%

100%

0%

50%

100%

5%

37%

58%


prevalent in the recruiting space, where 33% of organizations are planning high levels of investment in AI-based candidate assessment and another 67% are planning a medium level of investment (see Figure 5). AI-based candidate assessments use machine learning functionality to deduce candidates’ competencies, skills or personalities and improve hiring decisions. They can improve recruiting professionals’ ability to find relevant candidates, as it provides higher-quality candidates to review while allowing the organization to review a greater volume of applicants. Only 10% of talent analytics functions use AI-based candidate assessment, but of those, 33% report it has a high current impact.Implication for Talent Analytics Leaders: As with other algorithm-based processes, the way the candidate assessment “learns” profoundly impacts candidate outcomes and sometimes introduces unintended consequences. Questions to Consider: What safeguards do we have in place to ensure our organization’s use of machine learning or automation does not introduce bias into the candidate selection process? Can we apply AI to other contexts beyond talent acquisition?

Figure 5: Expected Investment in AI-Based Candidate Assessment Over the Next Two YearsPercentage of Talent Analytics Functions


Low Investment

High Investment

Medium Investment

0% 25% 50%

11%

47%

42%


Let’s start with an overview of your team. What does your team look like, and what are your overall responsibilities? The corporate organizational development (OD) team drives key talent management priorities across organization design, performance management, career development, leadership development, climate assessments and workforce analytics. Our overarching objective is to ensure our talent management practices are in line with evolving external and internal imperatives.

What do you see as the role of talent analytics in HR today at Infosys?There are three levels at which the talent analytics team contributes to the organization. The first and most basic is reporting HR metrics. This is how we keep a check on the pulse of key organizational indicators, such as attrition, engagement scores, role and diversity ratios. Secondly, we tackle high-priority business problems through data analytics. This is typically need-based, hypothesis-oriented and reactive. The third and largest contribution we make is through predictive analytics. We identify areas where we can utilize data and statistics to define models that benefit the organization and employees. The application of predictive analytics is usually proactive and aligned to organizational objectives, such as talent development, retention and reskilling.

HR teams can obviously do a lot with predictive analytics today. How do you think about where and when to use that kind of analytics?As mentioned, our initiatives are always steered by larger organizational objectives. Analytics pitches in everywhere — be it to solidify a

About the Team

Dr. Nandini S Senior Vice President and Group Head — Organization Development, Infosys

Nandini leads talent management and development across Infosys. In her current role, among other key achievements, she has led a performance management transformation for Infosys; the design of a first-of-its-kind digital career platform for job seekers and learners to find their best fit; and the development of an analytics-led manager enablement platform as well as a more contemporary, innovative and ongoing employee-engagement-sensing channel.

Deepa Prabhakaran Principal — Organization Development, InfosysDeepa leads the manager

enablement charter, engagement surveys and leadership assessments at Infosys.

Niharica Singh Principal — Organization Development, InfosysNiharica heads workforce and

skill analytics, career movements and internal job markets.

Shaping Analytics to Improve Manager PerformanceAn Interview With Infosys’ Organizational Development Team


theory or hypothesis that’s being tested or to connect the dots between different systems and processes. Decisions at Infosys are always data-driven; therefore, any new process being introduced or overhauling of an existing system is always backed by strong data analytics.For example, we used predictive analytics to look at the attrition and longevity of our engineers. At first, we tried a one-size-fits-all analysis, but we quickly realized we needed to bucket employees by their career stages to start seeing emerging patterns. We then found different reasons for leaving in the first six months versus the second year, third year and so on. We also created a two-by-two matrix that plotted the criticality of employees based on their performance, skills, etc., and their risk of attrition across various stages of their career life cycle (see Figure 1). This helped us identify employees who were critical to the organization and at risk of attrition so we could target our efforts toward this highly critical talent segment.We also used large-scale data mining to look at the emerging skills and then recommended skills to help our engineers. Not everyone learns the same skills; we have tools and techniques to understand individuals’ skill proficiency and competency to see what’s most critical for that specific person to learn at that specific time. This became important for getting our employees to stay ahead of the curve.

New types of analysis require new data. How has the way you collect data from employees changed over the past few years?Infosys has almost completely eliminated paper surveys that required physical filling Source: Adapted from Infosys

Figure 1: Infosys’ Talent Criticality MatrixPe

rson

Cri

tical

ity

Critical but Low Attrition Risk High Criticality

Low Criticality Low Criticality but High Attrition Risk

Situation Criticality

and transcription. All processes — from entry, training, allocation, utilization, performance, recognition and engagement to exit — are now digitized. Digitization of our processes has meant that our systems are interconnected. There is better standardization of definitions and metrics as well as automation of these metrics. In an organization with over 200,000 employees, the lack of standardization used to be a big pain point. Now, different teams that are consumers of the same data can interact with the same systems to retrieve the data they need for decision making.


How do you encourage teams and employees to engage with pulse surveys more frequently?In a large and delivery-focused organization, daily deliverables can take precedence over seemingly softer aspects such as talent management. That is why communication and branding of the initiatives is key. This, followed up with metrics tracking and identifying pockets of nonutilization, helps us identify why an initiative is not as effective as we would like it to be. Take, for example our introduction of quick pulse surveys in addition to the existing large, organization-level engagement surveys. The organization-level engagement surveys were a great way to gather employee feedback on a range of themes that affect the employee’s workplace experience, such as company practices, opportunities, people, rewards and recognition, and work life. These surveys were dispatched once a year and were often influenced by employees’ recent positive or negative experiences at the time of the survey (such as being selected for an aspirational project opportunity, or low variable pay in the quarter). We realized the need to connect more often at the manager and team levels rather than the organization level to keep a check on employee sentiment and allow employees to give anonymous, objective feedback. We now release our engagement survey to at least 25% of every manager’s employees every quarter. More importantly, we can customize parts of the survey to be based on more pressing issues specific to the quarter (for example, performance reviews and compensation). We’ve also introduced more frequent pulse surveys that allow us to collect data more frequently at a team level on areas that affect the team climate and morale. The pulse surveys are also highly customizable in that questions

are designed by managers to elicit feedback on areas they know their teams are currently struggling with. The survey frequency is also decided by the manager. Typically, they are sent after a long week on a Friday afternoon with questions such as, “How was your week?” or, “What can I do to help make daily tasks more engaging?” The open-ended text provides a real-time tool that gives managers candid data on how the team is feeling. After we have analyzed the data, managers can come in on Monday morning with this insight. With the data available, managers decide what to do with the feedback and how they will let their employees know the feedback was heard.

Could you give us an example of an action or change a team has been able to make because of the new process?We are inundated with responses by managers on how beneficial it has been that their teams have a channel of communication that is regular, tailored to their current needs and anonymous. This has helped managers recognize and attack their blind spots. At an organizational level, as well, these responses are aggregated for pattern identification and grouped into themes. One recent action we took as a result of this analysis was to open external hiring opportunities to internal employees first, particularly so there are more role change opportunities at the mid-management level.

Changing direction slightly, let’s talk about some of the other applications of talent analytics at Infosys. What work has your team done recently that you are most proud of?Over the past 12 months we have launched a completely new way of driving manager enablement in the company. We call it MaQ

“ Over the past 12 months we have launched a completely new way of driving manager enablement in the company. We call it MaQ (short for Manager Quotient).”


(short for Manager Quotient). MaQ was the result of our curiosity to find a contemporary and personalized way to enable today’s managers. MaQ was born out of a necessity to swiftly enable the manager community to meet today’s and future challenges in a very personalized way. Our end objective was to make MaQ the one-stop shop for all managers to find out how they were doing as managers and also chart and track their own learning and development paths to improve their managerial styles. In short, we wanted to connect managers’ execution to their goals.

What does the analytics model behind MaQ look like? What process did you go through to vet and refine it?The process involved first working on the data to get the analytics model in place, then verifying its findings with reliability tests and focus group discussions with managers. We then began designing the system and the user interface. In parallel, we also began the search, curation and classification of learning content that would be recommended by MaQ to managers. We use multiple data points and data analysis to see what variables to include and which have the biggest impact.

Once the system was ready to go, we began with a pilot approach in a subsection of managers and worked through the “teething issues” of our communication approach, FAQs and the feedback from managers. After a couple months of running this in a pilot mode, we launched MaQ across all Infosys managers (see Figure 2). The model, its metrics and the learning offerings all undergo regular review and revamp to ensure they respond to manager feedback and needs.

For others trying to determine how to help managers understand their effectiveness, what kind of data would you recommend looking at? In our experience so far, we have used data that is available:

• For at least 30% of the group

• At the individual project or manager’s level — so generalization is minimized

• In a systemized format rather than spreadsheets — so manual interventions are limited

Also, in the case of qualitative inputs, we chose to keep the anonymity of the feedback and therefore had to ensure that had a cutoff of a

Source: Infosys

Figure 2: MaQ — Infosys’ Personalized Manager PlatformOutput: 06:18PM Jul 12 2018

Modified 06:16PM Jul 12 2018MaQ: Infosys’ Personalized Manager Platform

CLC181576

Source: Infosys.


minimum of five inputs. Data sources that did not meet these requirements were not quite suitable for our tool. Some of the data we ultimately used came from our engagement surveys, manager performance reviews and team feedback, and even business data from business performance scorecards.

What was challenging about conducting the project? Anything you didn’t expect? For all new things, acceptance and adoption is always a challenge in an organization. The bigger challenge is to keep the engagement going on — because in this age of information overload, it is so easy for people to lose interest and move on to something else. So our work has a lot to do with continually upgrading the tool, the learning content and how we create the “pull factor” for managers to make MaQ a way of life at work.

The analytics behind the tool is certainly impressive, but it also seems like the interface is what really makes it useful for managers. What was the philosophy behind the user interface?Oh yes, the interface is a big part of engaging these managers in a simple yet catchy way. In the past, when we had detailed reports from other sources or vendors, our managers never used them much. So we wanted to create a simple, distilled version of all the feedback they were getting to let them know exactly what they should prioritize. And as with any product, we continue to explore even better formats and ways to engage with our users. The feedback from our managers has been positive, enthusiastic and very encouraging. Many have said they are impressed with how we are using analytics to work on their development, and that this was the first time they were seeing a holistic view of their managerial abilities in this way. Over 51% of our people managers are using MaQ today to either view their assessments or learn small nuggets on managerial effectiveness.

What other results has the project had? We are seeing some early results around increased learning effort and improved engagement in teams. Our leadership is definitely more receptive to talent analytics, as we now use analytics capabilities more and more in development, engagement, retention — and now even in skilling.

What’s next? How do you plan on building on your success? Currently, the bulk of our time is spent on refining the model, but we’re looking into how automation can help us. It won’t make it easier to read the data, since it will still take a lot of time to manually look through it and draw conclusions. This raises the question, how do you automate it? Human judgment is also necessary for projects like this, since we can sense a business problem in the first place. We can’t automate building a framework, since it comes from the journey and the experience with the business. Nevertheless, knowing where to draw data from and building the framework require that human component; after those two things, the rest can be automated. Finally, we look forward to continually innovating and optimizing. Each model we make should take less time because we’re always improving.

Any last advice for organizations that are just beginning their talent analytics journeys? Just start, and have curiosity! Analytics is extremely powerful, and it can be easy to be bogged down by problems, such as not having enough or “proper” data, not having team talent that’s analytics-ready and, especially, not knowing what to do. Any sort of analytics is iterative. Focus on analytics maturity models, and level up as you build capability and scale. Just be careful not to fall into the trap of being a data report churner rather than a partner in key data-based decision making. Good luck!


As talent analytics teams look to big data to produce insights for their businesses, they must ensure they have the correct tools and technologies to support advanced analytics.

Many have already established a strong infrastructure for running talent analytics, including investments in data warehouses, data lakes and analytics tools. Now, talent analytics leaders must determine what other investments to make to maximize the return on those data investments.

Our research shows IT infrastructure leaders are starting to invest primarily in five technologies: artificial intelligence (AI), machine learning, Internet of Things (IoT) platforms, predictive modeling tools and text analytics (see Figure 1). As talent analytics teams look to expand their analytics capabilities and use of big data, they should partner with IT to learn about and leverage these new technologies.

5 Technology Bets for a Big Data FutureBy Matthew Dong

Our Five Technology Bets

3%5%

4%4%

4%5% 4%

3%6%

12%

13%

6% 12%

6%

Wireless Sensor N

etwork

Wearable Technologies

Virtual Reality

Virtual Personal Assistants

Next-Generation Biometrics Systems

3D Printing

Blockchain

AI Augm

ente

d Re

ality

Inte

grat

ed In

tellig

ent S

enso

rs

IoT Pl

atfo

rms

Machine Learning

Natural Language Processing

Predictive Modeling Tools

Text Analytics

5%5% 10%10% 15%15%

Proof of Concept The technology is being evaluated through pilot programs for eventual deployment.

Monitoring The technology is being actively

evaluated for investment, but

there are no current deployment plans.

n = 50Source: Gartner 2017 Digital Technology Trends 2020 Survey

11%

Figure 1: Digital Technology Trends 2020Percentage of Respondents Anticipating Investment in These Technologies


Technology 1: AIAI is technology that appears to emulate human performance by appearing to understand complex content. What makes AI systems so powerful is their ability to make independent decisions without human guidance. Most talent analytics professionals (68%) believe it will be important to have AI systems in the future. And this belief is also prevalent throughout the business; 59% of IT leaders believe AI will have a transformational effect on the business.

To get started implementing and leveraging AI:

• Set a clear strategy for what you want AI to accomplish. With the multitude of AI programs available (e.g., chatbots, self-service automation), choosing where to invest can be overwhelming. Setting a clear strategy for what your team wants to accomplish can help narrow the programs and determine what kinds of results you need.

• Make sure your data is clean and available. AI relies on the data you have on hand for analysis and decision making. Therefore, talent analytics teams must ensure the data their AI programs are using is accurate.

Technology 2: Machine LearningMachine learning is a form of AI that applies computer algorithms to find patterns in data. It is a powerful tool that can improve various talent processes by detecting small, unusual trends in employee behaviors. Fifty-seven percent of technology leaders believe machine learning will transform the business. Even more talent analytics leaders (70%) believe it will be an important capability to have over the next five years.

As talent analytics teams prepare to use machine learning to glean deeper insight into their HR data, they should:

• Test first with a model. As with AI, typical machine learning programs take a subset of the available data to learn from and then create a model. However, to fully test its accuracy and functionality, the program must be tested against the rest of the available data to verify its accuracy.

• Be wary of bias. Machine learning processes are built by humans, so their algorithms might be programmed with the author’s own biases. Therefore, talent analytics teams should


periodically verify whether their machine learning programs are running correctly or are biased; they should proactively test for bias and follow up with the data.

Technology 3: IoT PlatformsAn IoT platform is the network of physical objects that contain embedded technology (e.g., sensors, devices, multidevice systems, systems of systems) to communicate and sense or interact with their internal states or the external environment. As technology improves, new IoT platforms are emerging, offering new data such as employee movement data and employee biometric data (see Figure 2). Many talent analytics teams are looking at the potential to capitalize on IoT during this exponential growth in enterprise data.

IoT is commonly used to monitor employees through various applications (e.g., cameras, sensors, social media). But before immediately pursuing such tactics, talent analytics teams should:• Show how the data will be used to support

employees. Employees are more likely to be open to monitoring techniques when they can see how it will be used to benefit them. Therefore, talent analytics teams should clearly communicate what they plan to do with the data from their monitoring initiatives before doing any work in this space.

• Adhere to data privacy laws. GDPR and other regulations have strict guidelines on how employers can use employee data. These regulations protect employees and help mitigate growing employee concerns regarding how their data will be used. Organizations that

n = 44 talent analytics leadersSource: Gartner 2019 Future of Talent Analytics SurveyNote: Percentages may not add up to 100% because of rounding.

Currently Do It Plan to Do It May Do It Will Never Do It

Employee Movement Data (e.g., electronic badges)

Employee Fitness Data (e.g., activity trackers)

Employee Biometric Data (e.g., fingerprint, facial recognition)

Work Computer Usage Data

Work Computer Location Data

Work Phone Usage Data

Work Phone Location Data

Workspace Usage Data (e.g., sensors on chairs)

0% 50% 100%

5% 43% 49%

5% 62% 27%

8% 62% 27%

11% 68% 16%

11% 65% 16%

8% 43% 49%

11% 38% 49%

14% 57% 24%5%

5%

5%

3%

3%

3%

8%

0%

Figure 2: Percentage of Talent Analytics Leaders Using New Data Sources


adhere to these laws can avoid fines, employee backlash and other consequences.

Technology 4: Predictive Modeling ToolsPredictive modeling, a commonly used statistical technique for predicting future outcomes based on available data, is one area many talent analytics leaders have already started exploring. Forty-eight percent of talent analytics teams already conduct predictive analytics in their organizations, and the rest plan to or may start in the future. Over the next two years, most talent analytics teams plan to use predictive analytics to better understand attrition and retention and improve strategic workforce planning (see Figure 3).

Other topics, such as engagement and talent attraction are on talent analytics leaders’ minds to pursue in the future.

To use predictive modeling effectively, talent analytics leaders should:

• Scope potential action plans based on hypotheses. Many times, when talent analytics teams explore data and generate insights, they only introduce recommended actions when they deliver the results to the business. But this approach has a low chance of driving implementation. Instead, talent analytics teams should hypothesize action steps for necessary stakeholders from the very beginning. This approach better aligns the project with what the business wants and increases the chance the business will act.

n = 44 talent analytics leaders Source: Gartner 2019 Future of Talent Analytics Survey

Figure 3: Percentage of Talent Analytics Leaders Planning to Use Predictive Analytics in These Areas

Attrition and Retention

Strategic Workforce Planning and Talent Forecasting

Engagement

Talent Attraction and Acquisition

Employee Experience

Employee Performance Management

Leadership and Succession Planning

New Hire Quality

Diversity and Inclusion

Labor Market Assessment and Candidate Sourcing

Compensation or Benefits

HR Functional Performance

High-Potential Employee Identification

Learning and Development

0% 45% 90%

81%

78%

46%

43%

38%

38%

35%

32%

30%

30%

22%

22%

19%

19%


Technology 5: Text Analytics ProgramsText analytics is the process of deriving usable information from text. Text analytics can be difficult because computers are designed to interpret and analyze numerical data. But if talent analytics teams can harness text analytics successfully — using internal messages, for example — they can gain valuable insights on current employee experiences. Organizations looking to advance their use of text analytics should:• Stop reporting and start doing deeper

analyses. Instead of reporting word occurrences and just creating word clouds for the business, talent analytics teams should use text analytics programs to analyze trends, sentiments and other elements. These analyses can provide deeper insight into what employees are truly expressing.

• Be wary of data quality. When performing text analytics, talent analytics teams should be careful how they analyze their data because their machines cannot understand all human intentions. Many programs still have trouble reading sarcasm, for example. Talent analytics teams should also be aware some of their data sources may be biased toward more vocal employees. For example, if you rely too much on internal collaboration platforms, you might miss employees’ opinions from elsewhere in the organization.

ConclusionThe growing availability of technologies that use big data gives talent analytics teams more opportunities to analyze data in different ways and support business decisions at a faster rate. These five technology bets specifically will better equip talent analytics teams to maximize the potential of big data at their organizations and provide long-term value to the organization.


4 Predictions for Talent Analytics in the Digital Ageby Richard Nguyen


Given these shifts in HR and business leaders’ worlds, talent analytics teams will need to evolve how they deliver value to meet their stakeholders’ changing needs. To help you prepare, we’ve made four predictions for how talent analytics will change over the next three to five years.

Prediction 1: Employee Experience Assessment Will Rise to the Top of Talent Analytics’ AgendaCapture employee voice and identify opportunities to improve the employee experience to help HR provide the most impactful support.

Employees are expecting a consumer-like experience at work; 70% expect systems and communications that better anticipate and understand their needs. With the ability to aggregate data from various sources, run advanced analyses, leverage new technologies and extract insight, talent analytics teams are uniquely positioned to help HR clearly interpret what experiences matter most to employees. Accordingly, we predict that employee experience assessment will rise to the top of talent analytics’ agenda. To prepare for this new agenda item, talent analytics teams should start:• Capturing employee voice to understand

what employees value — Talent analytics teams can leverage their skill in synthesizing disparate data to identify the experiences that matter most to employees. While surveys are a natural place to start gathering data from employees, talent analytics teams should

consider also using sources such as internal forums (e.g., town halls, focus groups), social media sites and any behavior-monitoring technologies their organizations employ.

• Identifying opportunities to improve the employee experience — By spotting patterns in employee voice data (e.g., recurring moments of employee frustration, common topics of employee discussion), talent analytics teams can identify crucial opportunities for HR intervention that might otherwise go unnoticed.

Prediction 2: Effortless Delivery of Data Will Become Just as Important as the Analysis ItselfPush the right data at the right time to increase data-driven decision making throughout the organization.

In today’s digital age, many employees are looking for an effortless experience with technologies at work — just like they have in their personal lives. This demand for easy-to-use products and systems is particularly important for talent analytics clients who, as key decision makers, often have little time and attention to spare. As a result, we predict that the effortless delivery of data will become just as important as the analysis itself.In good news, talent analytics has already made strides toward satisfying demands for an effortless experience by creating easy-to-use products, such as real-time dashboards. And 94% of talent analytics teams are currently investing or plan to invest in self-service

Digital transformation is a top priority for about 90% of corporate leaders. In fact, in the next year, organizations will spend $1.7 trillion on digital transformation. As a result, HR and business leaders — talent analytics’ key stakeholders — are making and experiencing significant changes.


platforms.1 However, while enhancements such as self-service may make talent analytics more convenient to access, they do not necessarily make it easier to consume or understand (see Figure 1). Ultimately, “self-service” and “effortless” are not synonymous. Truly effortless delivery must be both relevant and timely, so talent analytics leaders should consider how to push the right information at the right time — or “constrain and nudge”:• Constrain — While talent analytics’ typical

goal of providing comprehensive, detailed data and analyses is admirable, decision makers often struggle to find the information most relevant to them. Instead of prioritizing comprehensiveness, talent analytics teams should constrain what they provide to only the most relevant information for their stakeholders’ challenges.

• Nudge — Delivering analytics through self-service puts the burden on decision makers to find the right data at the right time, and they may not always know what information would be helpful and when to look for it. In other cases, decision makers may only self-serve

data as confirmation, with their minds already set on a particular decision. Talent analytics teams should intervene with data and analyses closer to the point of a decision to help decision makers make more informed, higher-quality decisions.

Prediction 3: Accounting for Leader Risk Aversion Will Become Central to Talent Analytics’ WorkUnderstand leaders’ decision-making context, not just the business context, to support leader risk taking.

Leaders must often make decisions with minimal information and vague, uncertain outcomes. In these moments of uncertainty, leaders can over-rely on safe alternatives when, in reality, risks are necessary to adapt to disruptive innovation. Talent analytics teams are well-positioned to help leaders mitigate their risk aversion by providing data-based recommendations and tools to clarify uncertainties.

However, many leaders today perceive a growing sense of personal risk on top of business risk when making decisions. This growing sense of personal risk may make leaders less receptive to data that doesn’t support their gut instincts. To continue to deliver value, we predict that accounting for leader risk aversion will become central to talent analytics’ work.

Understand Leaders’ Decision-Making ContextMany talent analytics teams already involve their business clients early in the scoping process, looking to understand the business context of the request, which is a good practice. But often, personal reasons, such as fear and uncertainty, can interfere with leader risk taking.

To get ahead of leaders’ risk aversion, consider adding questions like these to your scoping process:

• What information would increase your confidence in making a decision? How about your key stakeholders’ confidence?

• Are you weighing any trade-offs for this decision that we should know about?

• Whose approval or buy-in do you need to move forward with any decision?

Figure 1: Common Themes in Feedback About On-Demand Access

n = 5,873 employeesSource: Gartner 2018 Digital Employee Experience Survey

Tuning Them Out

Too Much EffortTired

Confused Frustrated

OverloadedDo My Job in Other Ways

Need Direction

Too ComplexToo Many Hurdles

Hard to Keep Up

Conflicting Information


By getting underneath leaders’ sense of personal risk, talent analytics teams can secure trust, gather the right information and generate recommendations leaders will actually use.

Engineer Learning MomentsTo engineer effective learning moments when leaders resist recommendations that belie their gut feelings, talent analytics should:

• Probe why leaders mistrust your recommendations. Identify their underlying assumptions, and open a discussion about their misconceptions.

• Contextualize your findings. Provide comparable data from other business units or projects that can help discredit any wrong assumptions.

• Quantify the impact of leaders’ reliance on their misconceptions. How much money or talent might the organization lose if leaders rely on their guts?

By engineering learning moments, talent analytics can help create insecurity among leaders about the validity of their gut feelings and then rebuild that knowledge in the right way.

Prediction 4: Experimentation With New Data and Projects Will Create a More Sustainable Path to Driving Value Conduct frequent, small experiments to quickly determine talent analytics projects’ ROI.

To adapt to the pace of today’s business environment, we predict that rapid experimentation with new data and projects will create a more sustainable path to driving value. Experimentation will allow talent analytics teams to determine early in the design process which analytics projects will yield value. Then, they can commit resources to projects they know will have impact.To get the most out of experimentation, talent analytics teams should consider these three keys to success: 1. Prioritize seeing results quickly. Design

experiments that will yield results faster to see value earlier — or move on to a different experiment more quickly if the current one does not yield good results.

2. Make failure an option. Set clear thresholds for an experiment’s success, calculating

Source: Gartner (February 2019)

Figure 2: Experiment Proposal Template

Problem Hypothesis Targets

Our business strategy requires many employees to leverage new technical skills to succeed.

Creating an organizational map of critical skill needs will help leaders put the necessary skills in the right places.

1. Increase individual development plan submission by 10% over four weeks.

2. Reduce unfilled positions by 15% over three months. Metrics

• Number of individual development plans

• Unfilled positions


aggressive but attainable short-term targets that represent a meaningful business impact. Holding the team accountable to these targets ensures that when experiments do fail, they fail fast and early in the process (see Figure 2).

3. Capture scaling opportunities. Remember that an experiment’s purpose is to determine whether an initiative will create value, not whether it will solve a problem for the entire organization. Identify opportunities to scale targeted, experimental solutions to other areas of the business and other problems where they can create value.

ConclusionThe digital age promises many exciting opportunities for talent analytics teams, which range from capturing the employee experience in new ways to helping leaders become better at navigating risk. With our predictions in mind, talent analytics teams can take one step further in cementing their position as key partners to business clients moving forward.

1 Gartner 2019 Talent Analytics Innovations Survey


Gig or contingent workers represent 15% to 25% of the global workforce today; by 2025, they are projected to comprise 35% to 40%.1 The growing popularity of this employment model is raising questions for many: Where do you find this type of talent? What do you have them do? What’s unique about their needs and work? With the ability to analyze external labor market trends, talent analytics will likely be asked to answer many of HR and business leaders’ questions about the gig economy and how to take advantage of it.Below, we’ve featured some answers to business and HR leaders’ most common questions about

the gig economy and some project ideas you might consider if these topics are a priority for your business.

Where Can We Find Gig Workers?India, the Philippines and the U.S. are the top three hubs for contingent workforces.2 The Contingent Workforce Index is a comparison of local labor markets around the globe that measures the likelihood of identifying and accessing workers to supplement an organization’s full-time workforce. The numerical

Gig Economy FAQs for Talent Analytics Leaders

Gig EconomyA labor market characterized by temporary, flexible and short-term jobs for which organizations hire independent workers, including freelancers, contractors and temporary workers

by Shaileja Verma


values assigned to each country rank them in descending order based on the availability, cost, regulation and productivity of contingent workers (see Figure 1).

Within the U.S., the locations with the highest number of 2018 job postings for independent workers were New York; Houston; Chicago; Washington, D.C.; Atlanta; Los Angeles; Charlotte, North Carolina; Austin, Texas; Phoenix; and San Francisco.

In terms of industry, the healthcare, retail and consumer products, oil and gas, and professional services industries have witnessed the highest growth of gig workers in the past five years.

Talent Analytics Project Idea: Using external labor market data like the above, create a presentation for your recruiting team on where to find gig workers. Think about geographic location, sourcing channels and more.

Are There Different Kinds of Gig Workers?Gig workers can be divided into four segments (see Figure 2):

• Free Agents — Workers who derive their primary income from independent work and actively choose this working style

• Casual Earners — Workers who use independent work to supplement their income and do so by choice (Some have traditional primary jobs, while others are students, retirees or caregivers.)

• Reluctants — Workers who derive their primary income from independent work but would prefer a traditional job

• Financially Strapped — Workers who work independently to supplement their income but would prefer not to have to do side jobs to make ends meet

Talent Analytics Project Idea: Survey your existing gig workers to determine which segments they typically fall into and understand how preferences, needs and experiences differ in each segment. Encourage HR leaders in charge of employment branding and EVP to use this data for more targeted candidate outreach.

Primary Income

Supplemental Income

Preferred Choice Free Agents

30% (49 Million) Casual Earners40% (64 Million)

Out of Necessity Reluctants

14% (23 Million)Financially Strapped

16% (26 Million)

n = about 8,000 U.S. and European respondents Source: McKinsey Global Institute. “Independent Work: Choice,

Necessity and the Gig Economy.” October 2016.

Figure 2: 4 Contingent Worker Segments

Figure 1: Contingent Workforce Index

Source: Manpower Group. “Contingent Workforce Index.” 2016.

India

Philippines

U.S.

Israel

Ireland

U.K.

China

New Zealand

Australia

Norway

0 0.3 0.6

0.56

0.55

0.53

0.51

0.5

0.5

0.49

0.48

0.47

0.47


What Type of Work Are Gig Workers Well-Suited For?The dynamics of gig workers are now slowly shifting from traditional field and season-dependent jobs to highly skilled and project or assignment-based jobs. Although traditional gig jobs in areas such as transportation and material movement, office and administrative support and sales-related occupations account for a higher share in the world job market, skilled job functions such as content writing and marketing, PR and branding, graphic design, accounting and finance, data science and analytics, and software development comprise an average of 15% to 25% of the total job postings today (including permanent jobs).In the U.S., sales and marketing and office and administrative support jobs are the most in-demand for gig workers (Figure 3).Talent Analytics Project Idea: Examine the productivity and performance of gig workers in your organization, and identify trends based on the types of work and roles in which they

are placed. Share this information to inform recruiting and workforce planning strategies.

What Are the Main Benefits of Using Gig Workers?Benefits of using gig workers include:• Cost savings — Organizations can save a lot of

money by adjusting workforce size based on business requirements.

• Speed and agility — The gig economy allows organizations to quickly fill talent gaps to meet specific skill needs or competitive challenges as they arise.

• A boost to innovation — Involving workers from outside the organization creates an exchange of new knowledge and best practices across organizational boundaries.

Talent Analytics Project Idea: Understand your organization’s goal for using gig workers. Is it to save money or increase the speed of work? Then, develop a model to measure the ROI of investments in gig workers to understand

Source: Gartner TalentNeuron analysis

Sales and Marketing

Office and Administrative Support

Healthcare Practitioners and Technical

Transportation and Material Moving

Management Occupations

Food Preparation and Serving Related

Business and Financial Operations

Computer and Mathematical

Education, Training and Library

Healthcare Support

Others

Figure 3: Gig Economy Job Demand by Occupations in the U.S.

0% 15% 30%

16%

12%

9%

8%

7%

6%

5%

5%

4%

3%

25%


whether these goals are being met. If they aren’t, generate hypotheses to inform workforce planning as to what is happening and what could be improved.

What Are Some of the Risks of Leveraging Gig Workers? Gig workers are particularly difficult to engage, often because: • They feel less connected to the organization

and its goals than full-time employees.• They don’t get the same benefits as full-

time employees.• They face steeper learning curves than full-time

employees and receive little career support.

Any disengagement can affect productivity, work quality and likelihood of working for your organization again. Gig workers also complicate typical recruiting and workforce planning processes. Organizations must be prepared to develop new ways of managing the ebb and flow of gig talent if they want to tap into this part of the workforce.Given the stated risks of the gig economy, some of the key attractions for gig workers are:• Medical and health insurance• Retirement savings plans• Tax assistance• Occupational accident insurance• Remote working options• Monthly public transportation subsidies in case

of travel requirements• Career assistance counseling

Talent Analytics Project Idea: Measure your gig workers’ engagement levels and disengagement drivers. What is likely to make them want to work for the organization again when their assignment is done? Share this information with business leaders to help them improve the gig worker experience.

How Can We Incorporate Gig Workers Into Our Workforce Plans?Many companies are adopting freelancer management systems (FMS) that help them efficiently manage their new load of nontraditional employees. According to a 2017 study by the University of Oxford, projects sourced by Fortune 500 companies through FMS grew 26% between 2016 and 2017.3 Cisco’s Talent Cloud, PwC’s Talent Exchange, EY’s GigNow, Deloitte’s Open Talent Community, Accenture’s Digital Talent Broker and Amazon’s Mechanical Turk (MTurk) are great examples of these platforms.

To help connect recruiters with gig workers, TaskRabbit, Upwork, Fiverr, Toptal, Kaggle, PowertoFly, FlexJobs and Guru are some of the key available platforms.

Talent Analytics Project Idea: Analyze whether and how hiring gig workers affects your core workforce’s engagement levels. How do employees and managers working with gig workers feel about the coordination? These details can help business leaders not only manage gig workers but also determine whether the strategy is sustainable in the long term.

ConclusionAs organizations continue to invest in the gig economy, consider how you as talent analytics leaders can help ensure return on that investment. Working on some of the project ideas above and preparing to answer your stakeholders’ key questions can help you continue to provide value to your organization on a critically important topic in today’s business environment.

1 “Independent Work: Choice, Necessity and the Gig Economy,” McKinsey & Company

2 “2016 Global Analysis,” Contingent Workforce Index3 “Platform Sourcing: How Fortune 500 Firms Are Adopting Online

Freelancing Platforms,” University of Oxford


https://www.mckinsey.com/~/media/McKinsey/Featured%20Insights/Employment%20and%20Growth/Independent%20work%20Choice%20necessity%20and%20the%20gig%20economy/Independent-Work-Choice-necessity-and-the-gig-economy-Executive-Summary.ashx

https://www.manpowergroup.com/cwi/the+2016+contingent+workforce+index+(cwi).+from+manpowergroup.

https://www.oii.ox.ac.uk/publications/platform-sourcing.pdf

https://www.oii.ox.ac.uk/publications/platform-sourcing.pdf

Even in a world where reliance on data is deepening, only 40% of senior leaders seek out HR data when making business decisions.1 In addition, only 23% of heads of talent analytics believe leaders are even effective at using talent data to inform business decisions.1 In short, talent analytics is taking a backseat to other business analytics, leaving an untapped pool of valuable information leaders could leverage to improve business decisions.

Given this reality, heads of talent analytics

are looking to increase the demand for talent analytics and discover new ways to embed talent data in HR and business decision making. To help, we gathered a panel of HR leaders at our October 2018 ReimagineHR conference in Orlando to speak about their experiences and share key insights on pushing talent analytics to the forefront of the business. Panelists included Steve Hall, senior director of talent management analytics and solutions for Marriott, and Jocelyn Caldwell, vice president of workforce analytics and planning within HR for TIAA.

With the goal of increasing demand for and use of talent analytics, the panelists recommended focusing on three key actions:

• Increase the actionability of talent analytics.

• Ensure the credibility of talent data.

• Make talent analytics more accessible to decision makers.

Ensure Analytics Are Actionable, Credible and Accessible


by Jenna Zitomer


Actionability: Start With the End Goal in MindHeads of talent analytics know their data and analyses must be actionable for leaders throughout the organization to want (or be able) to use them. Our panelists shared that understanding leaders’ key objectives and the problems they need data to solve is a critical factor in ensuring data is actionable. This means talent analytics leaders must take the time when starting a project to clarify the problem at hand and the end goal. As one panelist noted, it’s difficult to get people to adopt something you’ve done when they didn’t ask for it and don’t see the relevance of it. To ensure final data and analyses are actionable, the end goal of the project must also be feasible. Our panelists recommended a set of questions for talent analytics leaders to ask to clarify the goal of the project and line up the fundamental factors before obtaining and analyzing any data (see Figure 1).

Credibility: Create Partnerships, Not Just SponsorshipsMost talent analytics leaders look to build sponsorships, hoping to gain credibility and buy-in for talent analytics through endorsements from influential senior leaders. However, our

panelists realized sponsorships aren’t enough. Sponsors may be beneficial for a short period of time, but their influence tends to plateau because they do not challenge and improve projects; they merely advocate for them. Instead, our panelists focus on building partnerships to help overcome the necessary gaps left by sponsorships. To do this, one panelist suggested going beyond HR to get cross-functional partners involved in talent analytics work. When a project expands beyond the confines of HR and aims to solve a business wide problem, it transforms from an HR analytics project to a business project. Two functions the panelists recommended reaching out to are marketing and IT, which both tend to have rich client data, extensive information on employees and expertise in data and analytics. Creating partnerships allows different functions to combine their unique skill sets, teach one another and build analytics capability throughout the organization, ultimately yielding a higher return on investment for project outcomes. Collaborating functions can also cross-check data to ensure its accuracy and suitability to the project, heightening credibility for talent analytics teams. These partnerships can be high or low investment, depending on the time and


How will this project drive value for the

business?

How will we know if this project is

successful?

Do other functions have the capacity to partner with us

on this project?

Is the data needed available for collection?

What is the core problem we are trying to solve?

What parts of the organization will this project affect?

Figure 1: Questions to Ensure Project Clarity and Feasibility


effort a talent analytics team is willing to expend (see Figure 2).

Accessibility: Combine Technology and Training to Increase Data UsageLeaders are more likely to use data if the process of finding and analyzing it is as streamlined as possible. Both panelists highlighted the role self-service tools can play in ensuring HRBPs, managers and senior leaders have easy, consistent access to talent analytics information.

One panelist noted, however, that the technology itself will not make talent analytics accessible; it must be combined with training. For instance, the panelists suggested putting significant effort into training HRBPs to know, use and become comfortable with the talent analytics data from self-service tools. The panelists agreed that HRBPs are often an overlooked potential partner for talent analytics teams. When trained effectively and with access to the right technology, they can act as a one-stop shop for business leaders in gathering data and information about talent — a sort of intermediary between the data scientist and the business.

Our previous research has found that analytics training in the HR community should be tailored to professionals’ skill levels and tenure, especially given that data judgment and communication skills have not always been common in HR.

Figure 2: Options for Building Cross-Functional Analytics Partnerships


• Establish permanent cross-functional teams to address recurring, shared challenges (e.g., annual forecasting and planning).

• Conduct cross-functional rotations for analytics staff.

• Update staff sourcing strategies to broaden applicable experience.

• Develop analytics networks and communities of practice.

• Conduct joint training for analytics staff from different functions.

Low-Risk Investment

High-Risk Investment


For instance, one best-practice organization employed a two-pronged approach:• For recent-graduate hires and others with some

experience in analytics, development focuses on completing a formal analytics project to help them apply their analytics capabilities within the organization and grow their organizational acumen.

• For more experienced HR professionals with little analytics background, a reverse-mentoring opportunity provides an informal method of partnering senior HRBPs with new hires who have completed their analytics projects.

One of the key benefits of this kind of training is that it ensures HRBPs (and other users) are confident enough to face pushback from senior leaders and stakeholders on recommendations made using data and analytics. While technologies such as self-service tools are certainly helpful for placing available data in leaders’ hands, heads of talent analytics must also be skeptical of any technology that makes advanced analytics too accessible (or easy). “Be wary of anybody with a shiny toy that says they’re going to use machine learning to figure out everything,” one panelist noted. “You’ll end up with wonky things.” Before relying exclusively on AI, heads of talent analytics should develop better-researched hypotheses that can be complemented by, rather than substituted with, data analysis from AI.

ConclusionTalent analytics teams that ensure their data and analyses are actionable, credible and accessible can move into 2019 knowing their efforts have a higher likelihood of making an impact. Taking extra time before a new initiative to dissect the project, form partnerships and streamline processes where possible will help the talent analytics team develop more clout. In this case, going slow to go fast is key.

1 Gartner 2019 Future of Talent Analytics Survey


3 Questions to Ask Before Implementing Learning Analytics

In the News

The digital transformation of learning and development (L&D) offers HR leaders new opportunities to embed learning within its talent strategies and make the business case for L&D investments crystal clear. Part of digital learning’s promise is the result of data and analytics, which enable organizations to measure and communicate learning programs’ impact more precisely than ever.

Unfortunately, as with all new technologies, the rapid emergence of new options can be overwhelming. Not every solution is right for every business, and adopting a technology without clearly understanding how it will generate value can be an expensive mistake.

To survey this new landscape of learning analytics, Justin Taylor, our director of talent solutions, moderated a panel discussion at our October 2018 ReimagineHR conference in Orlando, bringing together Patti Phillips, Ph.D, president and CEO of the ROI Institute; Dave Vance, Ph.D, executive director of the Center for Talent Reporting; and Kimo Kippen, former chief learning officer at Hilton. The conversation covered the range of new technologies emerging in this space, the opportunities they provide and the challenge of figuring out how to take advantage of those opportunities.

When considering an investment in learning analytics, our panelists shared three key questions leaders should ask.

Question 1: What Is Our Objective?

A number of technologies currently on the market apply analytics to L&D in different ways and to different ends. For example:• In adaptive testing, training modules and skill

assessments automatically adapt to each individual’s level of ability.

• Learning record stores and xAPI record and track learning experience data, allowing organizations to track employees’ learning more closely and draw more insights from that data.

• Learning experience platforms offer new ways to deliver learning to employees on an individualized, self-directed basis.

• Natural language processing, machine learning and augmented and virtual reality are also finding applications in learning.

With all these options out there, the panelists agreed, it’s important for organizations to identify exactly what they hope to get out of learning analytics before buying a new piece of enterprise technology. Don’t chase a shiny toy, Kippen advised; instead, ask what the business objective is and whether the investment is worth it. You might find that the extra dollar is better spent on fundamentals, Vance added, as new technology won’t fix more fundamental problems in your L&D program. “Without algebra,” he analogized, “you’re not ready for the calculus.”



With any learning solution, it’s essential to have a plan for demonstrating its impact, Phillips emphasized. These high-tech tools can do many things, but what they don’t do is prove the importance of learning in general. Thoughtful, research-based design and implementation make a difference in winning support for learning analytics initiatives and ensuring their success. This can be challenging, Vance noted, because personalized learning technologies are still new and businesses have not yet developed the measures to realize their full potential. Individual data can help illustrate the impact, he added, by showing how employees use learning and how it helps them improve in their day-to-day jobs.

Question 2: How Will We Communicate the Business Case?

Rightly or wrongly, CEOs and CFOs are primed to be skeptical of learning’s value. Top-level executives who measure everything against the bottom line can be tough customers for investments that lack a clear, direct return. Therefore, the panelists stressed, it’s important to pitch learning analytics to these executives in a language they understand. That advice can be applied as literally and tactically as using words that business leaders like and avoiding “training speak,” which often fails to resonate, Kippen noted. Knowing the business and what it needs is essential to ensuring the learning analytics initiative can be designed and framed around those needs.The challenge of the business case goes back to learning objectives. Organizations don’t need more training or new technology, Phillips said; what they need are improvements in output, quality, cost and time. Just because you can implement a new learning technology solution doesn’t mean you should. Think about what measures matter to your organization, she recommended, then determine which of these measures can be improved by changes in behavior and what people need to learn to make those changes. The business goal should be part of the process from the beginning, Vance stressed. Designing objectives around these measures will help establish a solid foundation of credibility, and so will partnering with other parts of the business to design the initiative around their real needs and capture their support.

This commitment to credibility and business results should remain firm as your learning analytics program gets underway. Learning leaders should have good answers for what they are doing, how they are doing it and why they are doing it that way, Phillips added. This means having clear standards of success, judging outcomes rigorously against these metrics and making sure you really understand what your data is telling you. “We never want to put data out without knowing what we’re talking about,” she emphasized.

Question 3: How Will It Help Learners?

Just as L&D needs to ask how an investment serves the business, it must also consider how it will serve employees. A learning program or platform has no value unless employees actually use it and benefit from it. User experience is a key success factor, Kippen remarked. Employees still spend too much of their workday searching for information they need to do their jobs. So a good, broad objective for any learning investment is to ensure employees can get the answers they need in the right place and time. To make adoption successful, solutions should be easy to use, both for the L&D function and for learners themselves, Phillips added.Thinking about the learner experience can help in deciding which solutions to pursue first. Of the learning analytics technologies listed under Question 1, adaptive testing has the most immediate payoff, Phillips noted, as it can help ensure you have the right people in the right learning programs, providing training for people who really need it while clearing the way for others to move forward in their career paths. As our research at Gartner shows, learning can be a useful tool for employee engagement and retention when employees see opportunities to grow and advance in your organization.


First Quarter 2019 A Magazine for Talent Analytics … › assets › Uploads ›...

Documents

Transcript of First Quarter 2019 A Magazine for Talent Analytics … › assets › Uploads ›...