Proforma tech sfo march 2013

29
© 2013 Proformative 1 Turning Operational outcome metrics into actionable predictive forecasting model Shyam Desigan CFO & Sr. VP of IT, American Academy of Physician Assistants

description

 

Transcript of Proforma tech sfo march 2013

  • 1. Turning Operational outcomemetrics into actionablepredictive forecasting modelShyam DesiganCFO & Sr. VP of IT, American Academyof Physician Assistants1 2013 Proformative

2. The second-fastest-growing health profession in the nextdecade86,500Number of certified PAs in 2012, according to theNational Commission on Certification of PhysicianAssistants 83,600Employment in 2010 108,300 Projected employment in 20202 2013 Proformative 3. 86,500Number of certified PAs in 2012, according to theNational Commission on Certification of PhysicianAssistants 83,600Employment in 2010 108,300 Projected employment in 20203 2013 Proformative 4. Membership 43,495Number of AAPA Members31,756PA Members 11,739 PA Student Members4 2013 Proformative 5. Top 4 Primary Work Settings for a PA Hospitals 39.4% Single-Specialty Physician Groups 19.7% Solo Physician Practice 11% Multi-Specialty Physician Group 9.5%5 2013 Proformative 6. Top 3 Primary PA SpecialtiesFamily medicine 24.8% Other Surgical sub specialties 23.2% Internal Medicine sub specialties 10.3%6 2013 Proformative 7. The Forecasting Challenge Can you gather all your data in one place, so you can analyze it? What are the revenue and expense drivers? Can you query the data and see immediate results?ERPDatabases What are the operating dependencies? Can line managers leverage BI to drive operational changes and can Finance quantify the impact of these variables? Is it fast to deploy and affordable? CRMSpreadsheetsSCMCPM7 2013 Proformative 8. Business Intelligence Business MonitoringReporting Intelligence What is happening?What happened? CorporatePerformanceManagementForecastingAnalyticsWhat will happen? Why did it happen?Budgeting& Planning What should happen? 2013 Proformative 9. Better BI & FP&A In the Cloud augments limited flexibility withexisting systemsLow entry cost and norequirements for capitalexpenditure Rapid deployment and self-service; norequirement for ITinvolvement Integrates securely andeasily with any other application, platform,environment 2013 Proformative 10. Adaptive Planning Analytics Platform 2013 Proformative 11. Analyze revenue and expense drivers 2013 Proformative 12. Drill-Down into KPIs 2013 Proformative 13. What-If Scenarios 2013 Proformative 14. Why Visualization? 2013 Proformative 15. Highly Visual Graphics to Aid in Quick UnderstandingRevenue Trend On Time Delivery New Sales PipelineIncident Type Production Trend Financial Status 2013 Proformative 16. Visualize Trends 2013 Proformative 17. Glean Insights from Dashboards Visualize KPIs Get Visual & Email Alerts Analyze & Perform What-IfScenarios Collaborate Personalize 2013 Proformative 18. Business Discovery:Business User-Driven BI RemixabilityApp Model and ReassemblySocial and Collaborative InsightMobility EverywhereIT ProductionFinance HRMarketing Sales 2013 Proformative 19. QlikView Business Discovery Presentation Layer Marketing FinanceOperationsPresentation Sales QLIKVIEW Windows IIS BUSINESS DISCOVERY APPSWEBSERVER BUSINESS USERSApplicationData / Business IT AdminsAnalysts DevelopersQLIKVIEWQVW; QVD files SERVER QLIKVIEW Windows QLIKVIEWMANAGEMENTBased QLIKVIEW DEVELOPERData Access CONSOLEFile Share PUBLISHER(Optional)Custom connectors; ODBC; SecurityThird-PartyOLEDB; QVX; XML Integration:Integration: Windows Server Informatica EXCEL SQL SAPERP Tivoli Software Dell IBM Boomi SybaseDATAORACLESALESFORCEINFORMATICA and many more WAREHOUSE OPERATIONAL DATA SOURCES 2013 Proformative 20. 2013 Proformative 21. iDiscover for AAPA 2013 Proformative 22. AAPA Member Geo Analysis 2013 Proformative 23. AAPA Financial Ratio Trends 2013 Proformative 24. SPSS Modeler for Predictive analytics 2013 Proformative 25. Leveraging SPSS Modeler We utilize SPSS Modeler to create predictive models with our structureddata, including splitting, resampling and variable creation Developing predictive models including decision trees, support vector machinesand ensemble methods Visualization: Exploratory Data Analysis (EDA), and tools that persuade Evaluating predictive models, including viewing lift curves, variable importanceand avoiding overfitting 2013 Proformative 26. Leveraging SPSS Modeler We can apply record selection, sampling (including clustered and stratifiedsampling), merging (including inner joins, full outer joins, partial outer joins,and anti-joins), sorting, aggregation and balancing Choose from options for data restructuring, partitioning and transposition Select from extensive string functions: string creation, substitution, search andmatching, whitespace removal and truncation Apply RFM scoring: aggregate customer transactions to provide Recency,Frequency, and Monetary value scores and combine these to produce acomplete RFM analysis 2013 Proformative 27. Modeling Algorithms Apriori Popular association discovery algorithm with advanced evaluationfunctions Bayesian Networks Graphical probabilistic models CHAID & QUEST Decision tree algorithms including interactive treebuilding CARMA Association algorithm which supports multiple consequents K-Means, Kohonen, Two Step, Discriminant, Support Vector Machine (SVM) Clustering and segmentation algorithms Regression, Linear, GenLin (GLM), Generalized Linear Mixed Models(GLMM) Linear equation modeling Time-series Generate and automatically select time-series forecasting models 2013 Proformative 28. &28 2013 Proformative 29. Thank You For AttendingTurning Operationaloutcome metrics intoactionable predictiveforecasting modelCPE Code For This Session:___________29 2013 Proformative