Customer Intelligence & Analytics - Part I

5,414 views

Published on

Published in: Technology, Business
0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
5,414
On SlideShare
0
From Embeds
0
Number of Embeds
20
Actions
Shares
0
Downloads
92
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

Customer Intelligence & Analytics - Part I

  1. 1. Module 1: The World of Marketing IsChanging - Are You Being Left Behind? 1.1 Introduction 1.2 Competing On Analytics 1.3 The Data Explosion 1.4 Evolving Context of Marketing Analytics & Research 1.5 Questions
  2. 2. • Debbie Mayville – Sr. Solutions Architect, Communications & Marketing Analytics, SAS• David Kelley – Sr. Solutions Architect, Customer Intelligence, SAS• Suneel Grover – Solutions Architect, Integrated Marketing Analytics, SAS – Adjunct Professor, Integrated Marketing Analytics, New York University (NYU)
  3. 3. Module 1: The World of Marketing IsChanging - Are You Being Left Behind? 1.1 Introduction 1.2 Competing On Analytics 1.3 The Data Explosion 1.4 Evolving Context of Marketing Analytics & Research 1.5 Questions
  4. 4. „Old Spice‟ Campaign Case Study
  5. 5. Achieving Success With Business Analytics What’s the best that can happen? Optimization What will happen next? Predictive Modeling What if these trends continue? Forecasting Why is this happening? Statistical Analysis Alerts Query What actions are needed? Drilldown Ad hoc Where exactly is the problem? Reports Std. How many, how often, where? Reports What happened?
  6. 6. Business Analytics“The extensive use of data, statistical and quantitativeanalysis, explanatory and predictive models, and fact-based management to drive decisions and actions.” Davenport and Harris (2007) Competing on Analytics: The New Science of Winning
  7. 7. Data Deluge
  8. 8. Three Consequences Of The Data Deluge1. Every problem will generate data eventually.2. Every company will need analytics eventually.3. Everyone will need analytics eventually. ...
  9. 9. Three Consequences Of The Data Deluge1. Every problem will generate data eventually. Proactively defining a data collection protocol will result in more useful information, leading to more useful analytics.2. Every company will need analytics eventually.3. Everyone will need analytics eventually. ...
  10. 10. Three Consequences Of The Data Deluge1. Every problem will generate data eventually. Proactively defining a data collection protocol will result in more useful information, leading to more useful analytics.2. Every company will need analytics eventually. Proactively analytical companies will compete more effectively.3. Everyone will need analytics eventually. ...
  11. 11. Three Consequences Of The Data Deluge1. Every problem will generate data eventually. Proactively defining a data collection protocol will result in more useful information, leading to more useful analytics.2. Every company will need analytics eventually. Proactively analytical companies will compete more effectively.3. Everyone will need analytics eventually. Proactively analytical people will be more marketable and more successful in their work.
  12. 12. The Business Analytics ChallengeGetting anything useful out of tons and tons of data
  13. 13. Hope For The Data Deluge + analytical tools = actionable knowledge
  14. 14. Changes In The Analytical Landscape Historically… ModelsAnalytical Modelers Management Historically, analytics have typically been handled in the “back office,” and information was shared only by a few individuals.
  15. 15. Changes In The Analytical LandscapeHistorical Changes – Executive Dashboards • Static reports about business processes – Customer Relationship Management (CRM) • The right offer to the right person at the right time – 360-degree customer view
  16. 16. Changes In The Analytical LandscapeRelational DatabasesEnterprise Resource Planning (ERP)Point of Sale (POS) SystemsDecision Support Systems – Reporting and Ad Hoc Queries – Online Analytical Processing (OLAP)Performance Management Systems – Executive Information Systems (EIS) – Balanced ScorecardBusiness Intelligence
  17. 17. CRM Evolution• Total Quality Management (TQM) – Product-Centric • Quality: Six Sigma • Total Customer Satisfaction • Mass Marketing• One-to-One Marketing – Customer Relationship • Wallet Share of Customer • Customer Retention• Customer Relationship Management (CRM) – Customer-Centric • Strategy • Process • Technology
  18. 18. Changes In The Analytical Landscape Now… Operations Targeting Proliferation of Models Customer CustomersAnalytical Modelers Service Retail Suppliers Now analytics are being pushed out to the “front office”. There are clear, tangible benefits that managementwill track. Data mining is a critical part of business analytics. Promotions Employees
  19. 19. Idiosyncrasies Of Business Analytics1. The Data - Massive, operational, and opportunistic2. The Users and Sponsors - Business decision support3. The Methodology - Computer-intensive adhockery - MultidisciplinaryData mining can be defined asadvanced methods for exploringand modeling relationships inlarge amounts of data.
  20. 20. The Data Experimental OpportunisticPurpose Research OperationalValue Scientific CommercialGeneration Actively controlled Passively observedSize Small MassiveHygiene Clean DirtyState Static Dynamic
  21. 21. The Data: Disparate Business UnitsMarketing Invoicing RiskAcquisitions Operations Sales
  22. 22. Opportunistic Data– Operational data • Typically not collected with data analysis in mind– Multiple business units • Silo-based data environment This makes business analytics different from experimental statistics and especially challenging
  23. 23. The Methodology: What We Learned Not to Do• Prediction is more important than inference 1. Metrics are used “because they work” 2. p-values are directional guides 3. Interpretation of a model might be irrelevant 4. The preliminary value of a model is determined by its ability to predict a holdout sample 5. The long-term value of a model is determined by its ability to continue to perform well over time 6. Models are retired as behavior and trends shifts
  24. 24. Using Analytics Intelligently• Intelligent use of analytics 1. Understanding of how marketplace shifts affect business performance 2. Ability to distinguish between effective and ineffective interventions 3. Efficient use of assets, reduced waste 4. Risk reduction via measurable outcomes 5. Early detection of trends hidden in massive data 6. Continuous improvement in decision making
  25. 25. Simple ReportingExamples: OLAP, RFM, descriptive statistics, extrapolationAnswer questions such as:1. Where are my key indicators now?2. Where were my key indicators last week?3. Is the current process behaving like normal?4. What’s likely to happen tomorrow?
  26. 26. Proactive Analytical InvestigationExamples: Data mining, experimentation, empiricalvalidation, predictive modeling, optimizationAnswer questions such as:1. What does a change in the market mean for my targets?2. What do other factors tell me about my target?3. What is the best combination of factors for maximum profit?4. What is the highest price the market will tolerate?
  27. 27. Data Stalemate• Many companies have data that they do not use or sell to third parties. These third parties might even resell the data and any derived metrics back to the original company!• Story: Retail grocery POS card
  28. 28. Every Little Bit…Taking an analytical approach to only a few key businessproblems with reliable metrics  tangible benefitThe benefits and savings derived from early analyticalsuccesses  managerial support for more analytics1. Everyone has data2. Analytics can connect data to smart decisions3. Proactively analytical companies outpace competition
  29. 29. Areas Where Analytics Are Often Used• New customer acquisition Which residents in a ZIP• Customer loyalty code should receive a• Cross-sell / up-sell coupon in the mail for a new store location?• Pricing tolerance• Supply optimization• Staffing optimization• Financial forecasting• Product placement• Churn• Insurance rate setting• Fraud detection• …
  30. 30. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty What advertising strategy• Cross-sell / up-sell best elicits positive sentiment toward the• Pricing tolerance brand?• Supply optimization• Staffing optimization• Financial forecasting• Product placement• Churn• Insurance rate setting• Fraud detection• …
  31. 31. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell What is the best next• Pricing tolerance product for this customer?• Supply optimization• Staffing optimization• Financial forecasting• Product placement• Churn• Insurance rate setting• Fraud detection• …
  32. 32. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell• Pricing tolerance What is the highest price• Supply optimization that the market will bear• Staffing optimization without substantial loss of demand?• Financial forecasting• Product placement• Churn• Insurance rate setting• Fraud detection• …
  33. 33. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell• Pricing tolerance• Supply optimization How many 60-inch HDTVs• Staffing optimization should be in stock?• Financial forecasting• Product placement• Churn• Insurance rate setting• Fraud detection• …
  34. 34. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell• Pricing tolerance• Supply optimization• Staffing optimization What are the best times• Financial forecasting and best days to have technical experts on the• Product placement showroom floor?• Churn• Insurance rate setting• Fraud detection• …
  35. 35. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell• Pricing tolerance• Supply optimization• Staffing optimization• Financial forecasting What weekly revenue• Product placement increase can be expected• Churn after the Mother’s Day sale?• Insurance rate setting• Fraud detection• …
  36. 36. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell• Pricing tolerance• Supply optimization• Staffing optimization• Financial forecasting• Product placement Will oatmeal sell better• Churn near granola bars or near• Insurance rate setting baby food?• Fraud detection• …
  37. 37. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell• Pricing tolerance• Supply optimization• Staffing optimization• Financial forecasting• Product placement• Churn Which customers are most• Insurance rate setting likely to switch to a• Fraud detection different wireless provider in the next six months?• …
  38. 38. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell• Pricing tolerance• Supply optimization• Staffing optimization• Financial forecasting• Product placement• Churn• Insurance rate setting How likely is it that this• Fraud detection individual will have a claim?• …
  39. 39. Areas Where Analytics Are Often Used• New customer acquisition• Customer loyalty• Cross-sell / up-sell• Pricing tolerance• Supply optimization• Staffing optimization• Financial forecasting• Product placement• Churn• Insurance rate setting• Fraud detection How can I identify a fraudulent• … purchase?
  40. 40. When Analytics Are Not Helpful• Snap decisions required Deciding when to run• Novel approach (no previous from danger data possible)• Most salient factors are rare (making decisions to work around unlikely obstacles or miracles)• Expert analysis suggests a particular path• Metrics are inappropriate• Naïve implementation of analytics• Confirming what you already know
  41. 41. When Analytics Are Not Helpful• Snap decisions required• Novel approach (no previous Predicting the adoption of data possible) a new technology• Most salient factors are rare (making decisions to work around unlikely obstacles or miracles)• Expert analysis suggests a particular path• Metrics are inappropriate• Naïve implementation of analytics• Confirming what you already know
  42. 42. When Analytics Are Not Helpful• Snap decisions required• Novel approach (no previous data possible)• Most salient factors are rare Planning contingencies (making decisions to work for employees winning around unlikely obstacles or the lottery miracles)• Expert analysis suggests a particular path• Metrics are inappropriate• Naïve implementation of analytics• Confirming what you already know
  43. 43. When Analytics Are Not Helpful• Snap decisions required• Novel approach (no previous data possible)• Most salient factors are rare (making decisions to work around unlikely obstacles or miracles)• Expert analysis suggests a The seasoned art critic particular path can recognize a fake• Metrics are inappropriate• Naïve implementation of analytics• Confirming what you already know
  44. 44. When Analytics Are Not Helpful• Snap decisions required• Novel approach (no previous data possible)• Most salient factors are rare (making decisions to work around unlikely obstacles or miracles)• Expert analysis suggests a particular path Predicting athletes’• Metrics are inappropriate salaries or quantifying• Naïve implementation of love analytics• Confirming what you already know
  45. 45. When Analytics Are Not Helpful• Snap decisions required• Novel approach (no previous data possible)• Most salient factors are rare (making decisions to work around unlikely obstacles or miracles)• Expert analysis suggests a particular path• Metrics are inappropriate• Naïve implementation of Only looking at one analytics variable at a time• Confirming what you already know
  46. 46. When Analytics Are Not Helpful• Snap-decisions required• Novel approach (no previous data possible)• Most salient factors are rare (making decisions to work around unlikely obstacles or miracles)• Expert analysis suggests a particular path• Metrics are inappropriate• Naïve implementation of analytics• Confirming what you already Ignoring variables that know might be important
  47. 47. The Fallacy Of Univariate ThinkingWhat is the most important cause of churn? Prob(churn) International Daytime Usage Usage
  48. 48. Expectations Leading The Analysis• Sophisticated analytics are not immune to personal bias – Selectively fitting models because they place an opinion or agenda in a positive light – Ignoring information that might disprove a hypothesis• Personal bias, whether intentional or not, can diminish the usefulness of analytics
  49. 49. Trustworthy AnalyticsLet the data guide your conclusions – Are my assumptions about the causes of the data patterns warranted? – Should I be trying something different?Assign a cynic to the analytical team whose purpose isto question the assumptions
  50. 50. Idea Exchange Identify several business problems that you could address with analytics Describe the goal, whether the variables can be measured, how the data could be obtained, and what types of specific questions you would like to address with analytics
  51. 51. Case Study – US Telco• Data Deluge: Just Get Started – Low hanging fruit – Continue to build and get smarter – 360 degree view of the customer• Tools: Efficiency & Effectiveness – Data management tools – Analytic tools• Move to data driven insights versus gut reactions• Establish measurement system – Test & Learn Environment
  52. 52. Customer Lifecycle – Touch Points
  53. 53. Obtaining 360 Degree View Of The Customer Activ- Social ation Firmo- Network graphics Usage Demo- graphics Care Point of 360 Sale Degree Customer Hard- View Service, ware Repair VOD, Network Games Commu- Billing ni- Collect- cations ions
  54. 54. Large Telco With Industry-leading Churn Rate Churn Churn Reduction By Reduction Value ($) Reason Equipment 9 bps $121M Usage 16 bps $163M Network 15 bps $158M Active Issue Resolution 11 bps $110M Contract Renewal 25 bps $273MSales Channel / Credit & 6 bps $87M Collections Total 82 bps $912M
  55. 55. Case Study US TelcoBusiness Issue• Company-wide initiative to lower the churn rate among customers• Focus on “high value” or “high value potential” customers• Improve treatment strategy and relevanceSolution • Data management • Advanced analyticsResults/Benefits• Reduced churn by 40%• Increased customer loyalty and lifetime value• Increase of operational revenues by $1B over 3 years• Ability to uncover dissatisfaction drivers and tailor proactive churn treatments
  56. 56. Module 1: The World of Marketing IsChanging - Are You Being Left Behind? 1.1 Introduction 1.2 Competing On Analytics 1.3 The Data Explosion 1.4 Evolving Context of Marketing Analytics & Research 1.5 Questions
  57. 57. Key BUSINESS Trends Affecting Marketing From Product to Customer • Customer-centric business strategy • The customer experience • 360-degree customer view Finding the Next Origin of Business Growth • Consolidation/mergers/acquisitions • Market expansion • Efficiency & optimization The Regulatory Rise • Increased disclosure and transparency • Privacy and information sharing • Consumer contact rules • Regulatory reform
  58. 58. Key CONSUMER Forces Affecting Marketing Consumer in Charge • Rising expectations and more choice • From right time to “real time” • Demographic divide Channel Adoption • Mobile devices and consumer adoption • Web 2.0 and the digital age • Cross-channel usage Huge Online and Social Adoption • Social networking • Consumer-controlled content and channels • Consumer engagement
  59. 59. A Broadened Definition of “The Customer” The Consumer The Citizen The Subscriber The Plan Member The Patient The Patron...applicable across B2C & B2B
  60. 60. Customer Intelligence Is Relevant Across IndustriesFinancial Services Insurance Retail Hospitality & Telco & Cable Manufacturing Gaming Government Marketing Service Health & Providers Life Sciences Utilities
  61. 61. The Marketer Has An Evolving Mandate ExpectationExpectation Deliver a brandedIntegrated, multi-channel customer experiencein/outbound conversations in and outside ofin real-time marketing The Marketing Campaign The The Customer Brand Experience ResponsibilitiesExpectationSustain brand healthin a rapidly changingvirtual world Insights and AnalyticsExpectationUnearth and dynamicallymanage insights to drive action
  62. 62. Key Forces Affecting Marketers Huge Online and Social Consumer Adoption2B people online,100B monthly searches and600MM people on social networks globally
  63. 63. Key Forces Affecting MarketersHuge Online and Social Consumer Adoption Ever-Growing and Converging Marketing Channels Technology advances and consumer preferences driving new channels at unprecedented rates
  64. 64. Key Forces Affecting MarketersHuge Online and Social Ever-Growing and Converging Consumer Adoption Marketing Channels Information Explosion Business information doubling every 18 months with unstructured data representing 70% of it.
  65. 65. Key Forces Affecting MarketersHuge Online and Social Ever-Growing and Converging Information Explosion Consumer Adoption Marketing Channels The Speed of Business Information traveling at unprecedented rates, compounded by rising consumer expectations.
  66. 66. Key Forces Affecting MarketersHuge Online and Social Ever-Growing and Converging Information Explosion Consumer Adoption Marketing Channels The Speed of Business Accountability and Need to do More with Less Economic and competitive pressures putting focus on marketing budgets and returns.
  67. 67. Key Forces Affecting Marketers Huge Online and Social Ever-Growing and Converging Information Explosion Consumer Adoption Marketing Channels The Speed of Business Increasingly Competitive &Converging Markets Accountability and Need to doParity markets with limited differentiation . More with LessFight for share of wallet.
  68. 68. Key Forces Affecting Marketers Huge Online and Social Ever-Growing and Converging Information Explosion Consumer Adoption Marketing Channels Brand HealthLess corporate trust compounded by The Speed of Businessbrands being publicly scrutinized.Traditional mass marketing provingless impactful. Increasingly Competitive & Converging Markets Accountability and Need to do More with Less
  69. 69. The Marketing Process Mobile Online Finance Risk Call Customer Center Service In Person Merchandising Social Corporate AffairsDirect Mail Marketing Operations Optimization Marketing Marketing Marketing Strategy Processes Campaigns Analytics Data IntegrationERP CRM EDW Online Social Campaign
  70. 70. The Data Integration & Management Challenge
  71. 71. The Flood Of Data• Customer data continues to flood the organization exponentially• Progressing from functional to strategic – Namely how to capture, integrate, manage, analyze, and apply knowledge/insight about customers – Google Executive Chairman Eric Schmidt: “We create as much information in two days now as we did from the dawn of man through 2003.”
  72. 72. Structured & Unstructured Data • Company data: billing, usage, collections, set-top box, customer, web interactions, campaign, and more! • Consumer-generated data: Social media, blogs, product reviews, and more! Structured data 25% 70% 5%Unstructured data Semistructured data
  73. 73. “Big Data” Myths• Data Volumes are “Exploding” – Did Wal-Mart suddenly sell more stuff? – Did NYSE suddenly do more stock trades? – Did Netflix suddenly rent more movies? – Did Amazon suddenly sell more books?• This is existing data that previously went un-analyzed: A. Too large to manage B. Too costly to store C. Lack of “analytic chops” to capitalize
  74. 74. “Big Data” - Why Now? Three Vs? Complex, Unstructured 1. Volume 2. Velocity 3. Variety Relational The primary driver is Value…Source: IDC .
  75. 75. “Big Data” - Why Now?• Cost of storage dropping
  76. 76. Data - Prerequisite For Everything Analytical“You can’t be analytical without data, and you can’t bereally good at analytics without really good data” • Structure • Uniqueness • Integration • Quality • Access • Privacy • Governance Davenport, Harris, Morison (2010) Analytics at Work: Smarter Decision Better Results
  77. 77. Data Structure / Uniqueness / Integration Structure • Data structure affects analysis performance • Transaction systems (tables), data cubes (limitations) • Data arrays • Unstructured data Uniqueness • Data only your company has access – proprietary • Commercially available data – be the industry 1st • Create new metrics and data fields Integration • Aggregate data from inside/outside your organization • Consolidate silos across departments • Data has to be sourced, cleaned, integrated • Evolve to “one version of the truth”
  78. 78. Data Quality / Access / Privacy Quality • Flawed data causes misleading results • To fix problems - look at the data source • Continuous process – data will never be perfect • Start based on business objectives Access • Source data and load in a form for analytics • Size or complexities can cause user issues • Speed needs require data warehouse appliances • Sample populations Privacy • Guard the information collected • Well defined policies • Privacy laws within territories or industries • Don’t sell information without permission (opt-in)
  79. 79. Data GovernanceGovernance• Ensure data is useful for analysis• Consistent, defined, sufficient quality, standardized, integrated, accessible• Standard definitions and terminology• Decide on investments• Owners and stewards• Analytical data advocates• Business intelligence competency centers, analytical data advocate group, information management
  80. 80. Module 1: The World of Marketing IsChanging - Are You Being Left Behind? 1.1 Introduction 1.2 Competing On Analytics 1.3 The Data Explosion 1.4 Evolving Context of Marketing Analytics & Research 1.5 Questions
  81. 81. „The Greatest Job In The World‟
  82. 82. The Challenge of Digital Marketing• As digital marketing continues to grow more significant, new channels add complexity to the design of a successful integrated campaign. – It’s both a blessing and a curse for when an integrated campaign goes viral – Key Challenge: How do we do it again? – No repeatable formulas or clear attribution metrics
  83. 83. Reactive Business Analytics What’s the best that can happen? Optimization What will happen next? Predictive Modeling What if these trends continue? Forecasting Why is this happening? Statistical Analysis Alerts Query What actions are needed? Drilldown Ad hoc Where exactly is the problem? Reports Std. How many, how often, where?Reports What happened?
  84. 84. Proactive Business Analytics What’s the best that can happen? Optimization What will happen next? Predictive Modeling What if these trends continue? Forecasting Why is this happening? Statistical Analysis Alerts Query What actions are needed? Drilldown Ad hoc Where exactly is the problem? Reports Std. How many, how often, where?Reports What happened?
  85. 85. Afternoon Workshop Preview• What if I could? – Automate the measurement of sentiment relevant to my business goals from digital channels – Capitalize on the hidden value in vast amounts of available structured/unstructured data associated with my brand – Become strategically more proactive to shifting (dynamic) consumer trends
  86. 86. The Marketing Process Mobile Online Finance Risk Call Customer Center Service In Person Merchandising Social Corporate AffairsDirect Mail Marketing Operations Optimization Marketing Marketing Marketing Strategy Processes Campaigns Analytics Data IntegrationERP CRM EDW Online Social Campaign
  87. 87. Module 1: The World of Marketing IsChanging - Are You Being Left Behind? 1.1 Introduction 1.2 Competing On Analytics 1.3 The Data Explosion 1.4 Evolving Context of Marketing Analytics & Research 1.5 Questions

×