Pwc
- 1. How to turn data into actionable insights? 7 November 2016
- 2. PwC Advisory My mission is to transform data into insights and insights into actions in order to solve important problems 2 • KPN’s D&A team aims to generate more impact for adequate decision making within the organization • KPN D&A analysts have done a good job but the next step is to show even more added value towards internal stakeholders (e.g. Marketing, Sales, Products) • Our understanding of KPN’s requirements is two-fold; 1. Find a partner who can help KPN D&A moving forward on specific themes and providing meaningful insights towards its main internal stakeholders. The theme to start with is customer journey. 2. Obtain flexible support (project wise or other) in areas like specific deep analytical techniques, project management, data management, story telling, change management and reporting and structurally boost the skills of KPN’s D&A team in these areas. Astrid Wisse Director Data Analytics PwC
- 3. PwC Advisory Data is everywhere, but…. Doesn't answer my question What is this telling me? …making it accessible and understandable can be challenging. These graphs are too complex Mainly looking backwards
- 4. PwC Advisory Many companies find it difficult to capitalize on data analytics… Leading companies in R&C are investing in analytical capabilities … … however most of them struggle to capitalize on these insights • Trusting less on gut feeling, because of the complexity of their environment and wrong business decisions in the past • Gathering and storing a huge amount of data every day • Need for using and combining multiple data sources: e.g. sales and operational data • Ambition to build internal expertise and teams for analytics and BI consulting • Not knowing where to start. The number and variety of internal and external data sources is exploding • Lack of good data management makes it extremely hard to combine data from different data sources • BI activities are spread across the organization leading to many different models • A large part of the current BI work is backward looking bringing “nice to know” insights instead of forward looking bringing actionable insight to anticipate on the things to come
- 5. PwC Advisory …but the opportunities to improve performance by using data analytics remain significant in almost every part of the value chain PwC Data Analytics Supply chain • Spend analysis • Stock optimization Production • Demand forecasting • Overall equipment effectiveness • Predictive maintenance Customer • Basket analysis • Segmentation Pricing • Price promotions optimization • Price elasticity analysis Marketing & Sales • Campaign effectiveness (ROI) • Channel performance & optimization HR • Workforce planning • Workforce efficiency Brand • Brand loyalty • Brand cannibalization Finance / Management reporting • Revenue forecasting • Revenue leakage analysis
- 6. PwC Advisory Our way of working is an iterative process to learn quickly from insights found 6 1. Identify and diagnose high priority business problems 5. Take action4. Learn from new insights found 2. Hypothesis 3. Build & test with real data Iterate
- 7. PwC Advisory Increase store performance Case 1 7PwC Advisory
- 8. PwC Advisory Step 1 Identify and diagnose high priority business problems 8 • Issue: Sales behind budget • Business question: How can we increase store performance?
- 9. PwC Advisory Step 1: Analyze store performance to find root causes Rephrased business question: How can we increase the basked size for low performing stores? 9 Store A Store B Store C Store D Store E Store F Sales (€) 1,988,878 1,362,104 5,472,055 5,537,453 10,887,708 7,654,320 Sales area (SqM) 3,150 2,020 3,313 3,451 5,210 3,520 Sales/SqM 631 674 1,652 1,606 2,090 2,175 Traffic/SqM 180 160 190 200 230 210 Conversion (%) 28.1% 28.0% 31.9% 36.2% 34.4% 33.5% Basket Size (€) 112.50 120.40 167.10 170.45 183,25 185,65
- 10. PwC Advisory Step 2 Formulate hypothesis 10 • Issue: Sales behind budget • Business question: How can we increase the basked size for low performing stores? • Hypothesis: Low performing stores are less successful in cross selling
- 11. PwC Advisory 11 Step 3: Build and test with real data
- 12. PwC Advisory Step 4: Learn from new insights found: cross sell potential per store 12 28% 23% 10% 8% 1% 0% Store A Store B Store C Store D Store E Store F Compared to best in class
- 13. PwC Advisory 13 Step 5: Take action Store layout Staff utilization Staff selling skills
- 14. PwC Advisory Find new locations 14PwC Advisory Case 2
- 15. PwC Advisory Step 1 Identify and diagnose high priority business problems 15 • Issue: We need to open new fitness centers to generate growth • Business question: What are the best locations and how much additional revenue can we expect from these new locations?
- 16. PwC Advisory Step 2: Formulate hypothesis Hypothesis: There is a correlation between the revenue of a fitness center and the catchment area of that fitness center (distance < x km) 16 Heerlen x km
- 17. PwC Advisory Step 3: Build and test with real data 17 Zwolle Groningen Rotterdam Heerlen Amsterdam Eindhoven Breda Arnhem Hengelo We developed a model to estimate the revenues of current locations using different catchment areas (in km) 10 20 30 Modelfit Catchment area (km) Model fit for different catchment areas
- 18. PwC Advisory 18 Step 4: Learn from new insights found We developed a model to determine optimal new fitness center locations 1. Plot existing locations (blue) 2. Plot all zip codes without a fitness center (orange) 3. Select zip codes with a potential revenue > x Euro (catchment area of 9 km, taking into account other fitness centers in that area) 4. Iteratively placing a store in the zip code with the highest potential. Afterwards recalculate the potential of the remaining zip codes) Existing Locations All Zipcodes Potential revenue > xOptimal locations 1 2 34
- 19. PwC Advisory 19 Step 5: Make it actionable Optimal new location Location for rent
- 20. PwC Advisory How to be successful in data analytics? 20PwC Advisory Summary
- 21. PwC Advisory How to be successful in data analytics? 1. Start small 2. Involve all expertise needed 3. Define and agree on the business question 4. Work in sprints of 2 or 3 weeks 5. Do not deliver a number of graphs but make it actionable 21
- 22. PwC Advisory Thank you very much! © 2016 PwC. All rights reserved. Not for further distribution without the permission of PwC. “PwC” refers to the network of member firms of PricewaterhouseCoopers International Limited (PwCIL), or, as the context requires, individual member firms of the PwC network. Each member firm is a separate legal entity and does not act as agent of PwCIL or any other member firm. PwCIL does not provide any services to clients. PwCIL is not responsible or liable for the acts or omissions of any of its member firms nor can it control the exercise of their professional judgment or bind them in any way. No member firm is responsible or liable for the acts or omissions of any other member firm nor can it control the exercise of another member firm’s professional judgment or bind another member firm or PwCIL in any way. Astrid Wisse Director | Data Analytics PwC Netherlands M: +31 6 20 84 10 29 astrid.wisse@nl.pwc.com