Implementing analytics? You need decision modeling and business rules

5,717 views

Published on

Delivering the promise of data mining and predictive analytics requires an operational platform that is agile, business-friendly and decision-centric - decision modeling with DMN and business rules.

Published in: Technology
1 Comment
10 Likes
Statistics
Notes
No Downloads
Views
Total views
5,717
On SlideShare
0
From Embeds
0
Number of Embeds
19
Actions
Shares
0
Downloads
0
Comments
1
Likes
10
Embeds 0
No embeds

No notes for slide
  • Putting business analytics to work is top of mind for organizations like yours. Business agility and operational responsiveness are more important than ever. There is a real opportunity to use analytics – especially predictive analytics – to seek out increasingly small margins and understand your customers, products, channels, partners and more. But predictive analytics is only part of the solution – you must put these analytic insights to work making better decisions every day. Business rules offer the agile, business-centric platform you need to manage decisions and effectively deploy predictive analytics. Putting them together requires a new conceptual framework – Decision Management.
  • Applying analytics to acquire, retain and grow 100M customersBusiness challenge:100M customers and 3Bn calls / day200TB of customer information1.3M Retail partnersRural and urban consumers, large and small companiesSolution:Integrated data warehouse across all channels, all productsReal-time analytics for micro-segmentation, offer targetingWeb, retail, call-center and mobile channelsBenefits:Rapid growth with 2-3M new customers/monthGrowing and accelerating Revenue Market Share
  • Models make predictions but predictions alone will not help much – you must ACT based on those predictions.When you are thinking about smarter systems, taking action means having the system take action in a way that uses the predictions you made. You need to make a decision based on those predictions and this means combining the models with rules about how and when to act.Let’s take our retention example from earlier. Knowing that a customer is a retention risk is interesting, acting appropriately and in time to prevent them leaving is usefulGrovel index story
  • Remember – decisions are where the business, analytics and IT all come together
  • Once deployed analytics cannot be a “black box”, we must understand analytic performanceObviously you need a 'hold out sample' or business as usual random group to compare to.You need to understand what's working and what's the next challenge – which segments are being retained, for instanceYou must understand operational negation.You need to track input variables, scores, decisions or actions taken (classic example is in collections where a strategy may dictate a 'do nothing' strategy, but the collections manager overrides the decision and puts the accounts into a calling queue) and operational data that fed the decisionBoth analysts and business users must think about what they can do to improve decision making, which is the foundation of adaptive controlIn our retention example I need to have some customers I don’t attempt to retain or that I don’t spend any money retaining. I have to capture what the call center representative ACTUALLY offered and what was actually accepted (if anything), not just what SHOULD have been offered and I have to be able to show the results to my business users in terms they understand.
  • Actions not predictions - Business rules add actions to analytic insightTime to impact - Externalized decisions, rapid deploymentBusiness results - Decisions impact KPIs, implement strategyEngage IT, Business - Rules for the business, Decision services for ITMonitoring , compliance - Rules and explicit models expose decision making
  • Sometimes the ROI is discussed in terms of keeping a company in business, eliminating those company killing risks.This company offered trade credit insurance and a decision service provides trade credit calculations, combining business rules and algorithms developed by using predictive analytic techniques. Business experts interact directly with trade credit rules by using rule templates to ensure that rules match the underlying object model, without business users having to understand the object model’s technicalities.They got some classic BRE ROI:A country can be added in a few weeks rather than months, so the organization went from 2 to 16 covered countries in just 3 months.Ongoing changes can be made in hours rather than weeks.Most importantly though the system allows immediate changes to rules in a crisis, preventing the possibility of liability or other legal exposure for the organization.
  • So making decisions correctly will be hard unless we can pull all these rules together.Given this is how rules often look to start with, this is clearly going to be hard.But rules also change…Because your business policies doBecause your competitors doBecause the law doesBecause stock levels doBecause your services and products doBecause your customers doBecause your customers’ needs doSo we need something that will let us collect, manage and update the business rules that drive our decisions
  • The most basic representation is a list of rules or a rule set. These are simple atomic rules grouped into a logical set for execution and storage. Rule sets are often shared between decisions.
  • A big part of the benefit companies get from managing rules comes from putting the business in charge more directly. Having business users manage business rules reduces costs by eliminating a step – that of having the business tell IT what they want so that IT can do code it – and improves accuracy by eliminating the impedance of this step. It also increases business agility by making it easier for a company to respond to changes – after all the business folks notice the changes firstIn my experience, the use of business rules and a BRMS to manage high-volume, operational decisions have a proven track record in reducing application development costs and application maintenance. It takes fewer developers and less time to specify how a system or service should behave using a BRMS thanks to the increased expressive power of business rules and the improved verification and testing offered by BRMS. Maintenance of these rules is easier, often dramatically easier, than the maintenance of the equivalent code. Not only are can the business rules be changed independently and safely; business users can participate directly in the maintenance process for the first time. Domain expertise is applied more directly and less time and money are spent making changes.
  • Faster, easier, independent changes to decision logic Coordination of decisions across channels and products Higher employee productivity and resource utilization a leading French retailer of cosmetics, faced the challenge of multiple channels and overlapping marketing and loyalty offers. A customer might be eligible for a loyalty offer, have downloaded a web coupon and heard a “discount word” on the radio. This made it hard for retail staff to ensure the price was handled correctly at the point of sale. In addition, they needed a better way to get loyalty offers to the customer. Yves Rocher replaced their POS devices with Linux-based terminals and developed a rules-based system that allowed all the pricing rules to be defined by the marketing department and then downloaded into the terminals. All relevant offers are correctly combined at the point of sale. This system also takes the customer’s loyalty card and applies loyalty offers. Using purchases and loyalty history, it prints an incentive offer designed to bring the customer back to the store on the card itself—the cards are re-printable so the customer sees the offer that will be made when they return.
  • the classic sources of rules – policies, regulations, best practices, expertise. Many, most, of your rules will come from these sources. But your data is also rich source of rulesYou can analyze the data you have to find exactly which thresholds you should use in your rules – is it really customers above 21 years old or would 20 or 19 be a more meaningful cut off?You can use data mining techniques to actually find the rules – association rules such as if someone is buying this product try and sell them that product or segmentation rules – using your historical data to find out what were the most successful rules in the pastYou can use this historical data to create new insights into customers – predicting things about them that extend the data you can work with and against which you can write rules.
  • Descriptive analytics can be used to categorize customers into different categories – to find the relationships between customers - which can be useful in setting strategies and targeting treatment. But this analysis must be delivered not just to your analysts, also to your systems. Analysis is generally done offline, but the results can be used in automated decisions – such as offering a given product to a specific customer – often by developing rules that embody the analytics.For instance a decision tree can be created where each branch, each end node, identifies the segment for a particular member.Data mining can also create rules with less effort and with a quicker time to market in certain circumstances
  • Predictive analytics often rank-order individuals. For example, rank-order members by their likelihood of renewing – the higher the score, the more “completers” for every “non-completer”. The risk or opportunity is assessed in the context of a single customer or transaction and these models are not an overall pattern, even if they are predictive. Models are called by a business rules engine to “score” an individual or transaction, often in real time, though the analysis is done offline.These models are often represented by a scorecard where each characteristic of a member adds to the score and where the total score can then be returned.
  • All these pieces contribute to ever-more sophisticated decision services that support your business processes.Decision Services externalize and manage the decisions production processes and systems needBusiness rules allow business users to collaborate in the declarative definition of decisionsAnalytics can create better more data-driven business rulesAnd ultimately additional predictive analyticsAdaptive control allows test and learn to become part of a continuous improvement loop
  • Here’s another example, this time of an insurance company with about 750,000 policies that implemented a risk-based underwriting decision service for use across its systems. In the first year an eight-point reduction in combined ratio – a big deal for an insurance companyThey got this improvement from all the areas I see when clients apply decision managementThey reduced costs by eliminating many manual reviews and by putting underwriters and actuaries in charge of the rules behind the decision – they eliminated or reduced many of their IT costs.They boosted revenue, the second major area, by improving risk management (far more tiers and more fine grained decisioning) and by focusing their staff on the book of business and helping agents improve it rather than on transactional approvalsThe third area does not show up in the specifics but when I talked to them it was clearly the most powerful aspect of the whole thing. They gained true strategic control over their underwriting decisions.
  • Actions not predictions - Business rules add actions to analytic insightTime to impact - Externalized decisions, rapid deploymentBusiness results - Decisions impact KPIs, implement strategyEngage IT, Business - Rules for the business, Decision services for ITMonitoring , compliance - Rules and explicit models expose decision making
  • Little decisions add up so focus on operational or front-line decision makingThe purpose of information is to decide so put your data and analytics to workYou cannot afford to lock up your logic so externalize it as business rulesNo answer, no matter how good, is static so experiment, challenge, simulate, learnDecision Making is a process to be managed
  • Begin!Identify your decisionsHidden decisions, transactional decisions, customer decisionsDecisions buried in complex processesDecisions that are the difference between two processesConsiderWho takes them nowWhat drives changes in themAssess Change ReadinessConsider Organizational changeAdopt decisioning technologyAdopt business rules approach and technologyInvestigate data mining and predictive analyticsThink about adaptive control
  • Decision Management Solutions can help youFind the right decisions to apply business rules, analyticsImplement a decision management blueprintDefine a strategy for business rule or analytic adoptionYou are welcome to email me directly, james at decision management solutions.com or you can go to decision management solutions.com / learn more. There you’ll find links to contact me, check out the blog and find more resources for learning about Decision Management.
  • Implementing analytics? You need decision modeling and business rules

    1. 1. Implementing Analytics? You Need Decision Modeling and Business Rules JamesTaylor, CEO
    2. 2. 1 2 3 4 5 6 AGENDA The power of analytics Challenges in analytics Introducing business rules Integrating business rules and analytics Decision Management Wrap Up
    3. 3. ©2009-2017 Decision Management Solutions 3 The one slide you need Analytics have great potential to improve day to day operations Challenges include Business value gap Engaging the business and IT Deployment A business rules management system (BRMS) is an ideal platform for analytics Decision Management ties analytics and business rules together in an effective decision model framework
    4. 4. The power of analytics ©2009-2017 Decision Management Solutions4
    5. 5. Analytics have power Customer Churn Campaign Response Acquisition Rates Online Conversion Fraud Risk ©2009-2017 Decision Management Solutions 5
    6. 6. ©2009-2017 Decision Management Solutions 6 And that power is operational How do I… prevent this customer from churning? convert this visitor? acquire this prospect? make this offer compelling to this person? identify this claim as fraudulent? correctly estimate the risk of this loan? It’s not just about “aha” moments It’s about making better operational decisions
    7. 7. ©2009-2017 Decision Management Solutions 7 Operational Decisions Scale For Big Impact Strategic Decision Tactical Decision Operational Decision
    8. 8. ©2009-2017 Decision Management Solutions 8 Case: Telco Example: Analytics enabling an individualized decision about the best plan/upgrade offer to make to each customer
    9. 9. Challenges in Analytics ©2009-2017 Decision Management Solutions9
    10. 10. Knowing is not enough ©2009-2017 Decision Management Solutions 10 Those who know first, win Those who ACT first, win Provided they act intelligently
    11. 11. Latency in decisions costs you ©2009-2017 Decision Management Solutions 11 Business event Action taken Decision latency
    12. 12. Operational decisions are at the center Business ©2009-2017 Decision Management Solutions 12
    13. 13. Monitoring and compliance matter ©2009-2017 Decision Management Solutions 13
    14. 14. Analytics improves operations but… Power of analytics is in improving operations Specifically operational decisions Which means : Focus on actions not predictions Minimize the time to impact Engage analytics, IT, and business teams Handle monitoring and compliance ©2009-2017 Decision Management Solutions 14
    15. 15. Case: Trade credit insurance New countries in weeks not months Ongoing changes in hours, not weeks Immediate changes in a crisis ©2009-2017 Decision Management Solutions 15
    16. 16. Introducing business rules
    17. 17. What are business rules? “… a directive intended to influence behavior.” “… a formal expression of knowledge or preference, a guidance system for steering behavior (a transaction) in a desired direction.” “… statements of the actions you should take when certain business conditions are true.” ©2009-2017 Decision Management Solutions 17
    18. 18. Business Rules are everywhere Experienced Personnel RegulationsPolicy Manuals Legacy Systems Historical Data ©2009-2017 Decision Management Solutions 18
    19. 19. ©2009-2017 Decision Management Solutions 19 Business rules drive decisions Decision History Experience Policy Regulations Legacy Applications
    20. 20. public class Application { private Customer customers[]; private Customer goldCustomers[]; ... public void checkOrder() { for (int i = 0; i < numCustomers; i++) { Customer aCustomer = customers[i]; if (aCustomer.checkIfGold()) { numGoldCustomers++; goldCustomers[numGoldCustomers] = aCustomer; if (aCustomer.getCurrentOrder().getAmount() > 100000) aCustomer.setSpecialDiscount (0.05); } } } Unmanageable business rules ©2009-2017 Decision Management Solutions 20
    21. 21. Decision Service A Business Rules Management System After Smart (Enough) Systems, Prentice Hall June 2007. Fig 6.6 Design Tools Rule Management Applications Rule Engine Operational Database Rule Repository Production Application Validation and Verification Testing Deployment ©2009-2017 Decision Management Solutions 21
    22. 22. Manageable business rules Smart (Enough) Systems, Prentice Hall June 2007. Fig 4.3 If customer is GoldCustomer and Home_Equity_Loan_Value is more than $100,000 then college_loan_discount = 0.5% If member has greater than 3 prescriptions and prescription’s renewal_date is less than 30 days in the future then set reminder=“email” If patient’s age is less than 18 and member’s coverage is “standard” and member’s number_of_claims does not exceed 4 then set patient’s coverage to “standard” ©2009-2017 Decision Management Solutions 22
    23. 23. ©2009-2017 Decision Management Solutions 23 Rule Sheets Set of rules that share common action(s) Sequence does not matter Typically exhaustive and exclusive Sometimes called a decision table, rule family Name Condition Attribute 1 Condition Attribute 2 Condition Attribute 3 Action Attribute 4 Rule 1 > 2 < 3 =6 Rule 2 <=2 >=3 =12 Rule 3 <=2 <3 True =13 Rule 4 <=2 <3 False =14 Rule 5 >2 >=3 True =10 Rule 6 >2 >=3 False =2
    24. 24. ©2009-2017 Decision Management Solutions 24 Decision Table Set of rules that return a single result Look up tables, comparing attribute values Sequence does not matter Typically exhaustive and exclusive Attribute 2 Single Family with no children Family with children Attribute1 < 10 1 2 3 11 – 20 2 4 6 21 – 40 3 5 7 41 – 99 4 6 8 >= 100 5 6 9
    25. 25. ©2009-2017 Decision Management Solutions 25 Decision Tree Set of rules that select a goal from a set Divides into increasingly fine grained groups Often a powerful thinking tool Typically exhaustive and exclusive Existing Customer? Long term customer? Grandfather Issue change Price quoted Retract quote Issue warning Y N Y N Y N
    26. 26. ©2009-2017 Decision Management Solutions 26 A list of rules representing a rule set AdjustIncome RuleSet (newApplicant) Name: Six months or less on the job if newApplicant’s monthsInCurrentJob is less than or equal to 6 then decrement newApplicant’s income by 5600 Name: About 1 year if newApplication’s monthsInCurrentJob is between 6 and 13 then increment newApplicant’s income by 1000 Name: Greater than 1 year if newApplicant’s monthsInCurrentJob is greater than 12 then increment newApplicant’s income by 2500 Name: Computer Total Income if newApplicant is married and newApplicant’s spousalIncome is greater than 0 then newApplicant’s totalIncome is equal to newApplicant’s income plus newApplicant’s spousalIncome else newApplicant’s totalIncome is equal to newApplicant’s income Name: Minimum Income if newApplicant’s totalIncome is less than 40000 then newApplicant is not eligible Smart (Enough) Systems, Prentice Hall June 2007. Fig 6.3
    27. 27. Why manage business rules Improve Decision Making • Clear policies and procedures • Consistently applied across channels, systems • Increased accuracy from business users participation Business Agility • More rapid response to business threats • Fewer missed opportunities • Faster time to market Reduce Costs • Fewer resources, less time to change decisions • Lower fines, legal costs from bad decisions • Reduced IT costs to implement decisions ©2009-2017 Decision Management Solutions 27
    28. 28. Cross-channel coordination ensures that all relevant offers, including loyalty programs, are combined at the point of sale, with consistent pricing rules applied. ©2009-2017 Decision Management Solutions 28 Case: International retailer
    29. 29. Integrating business rules and analytics
    30. 30. 30 Different kinds of analytics Data Mining Who are my best/worst customers? How do I turn my data into rules for better decisions? Predictive Analytics How are those customers likely to behave in the future? How do they react to the myriad ways I can “touch” them? Knowledge - Description Action - Prescription Business Intelligence How do I use data to learn about my customers?What has been happening in my business? ©2009-2017 Decision Management Solutions * * * * **** ** * * ** * * * * * * * * * * * * * * * * * * * * ** * * ** * * *** * * * ** *** **
    31. 31. * * * * **** ** * * ** * * * * * * * * * * * * * * * * * * * * ** * * ** * * *** * * * ** *** ** Your data is a source of insight ©2009-2017 Decision Management Solutions 31
    32. 32. * * * * **** ** * * ** * * * * * * * * * * * * * * * * * * * * ** * * ** * * *** * * * ** *** ** Insight as presentation is disconnected ©2009-2017 Decision Management Solutions 32 ?
    33. 33. * * * * **** ** * * ** * * * * * * * * * * * * * * * * * * * * ** * * ** * * *** * * * ** *** ** Insights as decisions drive action ©2009-2017 Decision Management Solutions 33
    34. 34. Existing Customer? Long term customer? Grandfather Issue change Price quoted Retract quote Issue warning * * * * * * * * ** * * ** * * * * * * * * * * * * * * * * * * * * ** * * * * * * * * * * * * * * * ** ** Low-moderate income, young High Income High income, low-moderate education Moderate-high education low-moderate income High Moderate education, low income, middle-aged Low education, low income Education High Descriptive analytics ©2009-2017 Decision Management Solutions 34
    35. 35. Predictive analytics Attribute 2 Single Family with no children Family with children Attribute1 < 10 1 2 3 11 – 20 2 4 6 21 – 40 3 5 7 41 – 99 4 6 8 >= 100 5 6 9 ©2009-2017 Decision Management Solutions 35
    36. 36. Scorecards are a powerful tool Years Under Contract 1 0 2 5 More than 2 10 Number of Contract Changes 0 0 1 5 More than 1 10 Value Rating of Current Plan Poor 0 Good 10 Excellent 20 Score Smart (Enough) Systems, Prentice Hall June 2007. Fig 5.4 30 ©2009-2017 Decision Management Solutions 36
    37. 37. Building Decision Services Smart (Enough) Systems, Prentice Hall June 2007. Fig 5.1 Production ProcessData Warehouse Operational Data Store Business Rules Business Analytics Decision Service Enterprise IT Infrastructure Adaptive Control Events ©2009-2017 Decision Management Solutions 37
    38. 38. Case: Underwriting Manual reviews Control Business focus Risk 8 Point Reduction in Combined Ratio ©2009-2017 Decision Management Solutions 38
    39. 39. Decision Management
    40. 40. Decision Management Ties analytics and business rules together in an effective decision model framework Models decision-making (DMN standard) Integrates with execution environment Builds on existing enterprise architecture ©2009-2017 Decision Management Solutions 40
    41. 41. ©2009-2017 Decision Management Solutions 41 Delivering Decision Management 3 stages to better operational decisions Identify the decisions (usually about customers) that are most important to your operational success Design and build independent decision services to replace decision points embedded in operational systems Create a “closed loop” between operations and analytics to measure results and drive improvement
    42. 42. Case: State dept of taxation Business challenges Solution Benefits Paper tax returns increased costs and slowed responses Information system silos Manual fraud detection and return review Single central taxpayer database Integrated system Sophisticated real- time predictive analytics Recovered millions of dollars from dubious tax returns Increased collection of unpaid taxes Decreased number of questionable returns Increased customer satisfaction ©2009-2017 Decision Management Solutions 42
    43. 43. Implementing Analytics Gets Results Actions not predictions Minimum time to impact Engage IT, Business Handle monitoring , compliance Decision Management ties analytics and business rules together in an effective decision model framework ©2009-2017 Decision Management Solutions 43
    44. 44. Wrap Up
    45. 45. ©2009-2017 Decision Management Solutions 45 The one slide you need Analytics have great potential to improve day to day operations Challenges include Business value gap Engaging the business and IT Deployment A business rules management system (BRMS) is an ideal platform for analytics Decision Management ties analytics and business rules together in an effective decision model framework
    46. 46. ©2009-2017 Decision Management Solutions 46 Action Plan Identify your decisions before analytics Adopt business rules to implement analytics Bring business, analytic and IT people together
    47. 47. Decision Management Solutions Decision Management Solutions can help you Focus on the right decisions Implement a blueprint Define a strategy Learn more: Contact Us www.decisionmanagementsolutions.com info@decisionmanagementsolutions.com ©2009-2017 Decision Management Solutions 47
    48. 48. Thank you! James Taylor, CEO james@decisionmanagementsolutions.com www.decisionmangementsolutions.com

    ×