Decision Automation in Marketing Systems using Reinforcement Learning: Dynamics talks SF July 17th 2019
1 like403 views
- ML-based decision automation systems can make billions of micro-decisions in real-time to target customers, time promotions, and determine messaging and budgets. - These systems use techniques like propensity scoring, recommendation algorithms, and multi-armed bandits to optimize for business objectives within complex environments. - An example case study describes how a promotion targeting system for retailers and manufacturers can drive traffic, improve loyalty, and increase market share by automating decisions around targeting, timing, outreach, and promotion properties.
Decision Automation in Marketing Systems using Reinforcement Learning: Dynamics talks SF July 17th 2019
- 1. Decision Automation in Marketing Systems Ilya Katsov Head of Practice, Industrial AI Grid Dynamics SAN FRANCISCO JULY 17 2019
- 2. ML-based Decision Automation in Marketing Operations ● Billions of micro-decisions in real-time: who, when, how, what, ... ● Complex environment: human behavior, complex business models, hidden factors ● Many building blocks: propensity scoring, recommendation algorithms, multi-armed bandits, etc. ● How to design a system that can make micro-decisions based on business objectives?
- 3. Case Study: Environment Retailer 1 Manufacturer 1 Manufacturer NRetailer M purchases, clicks, loyalty IDs ... ... Promotion targeting system ● Drive traffic ● Improve loyalty ● Increase market share ● Acquire/grow/retain clients ● Improve loyalty
- 4. Case Study: Decisions to be Automated ● Targeting – who ○ Exploits variability in tastes, price sensitivity, propensity to buy ○ Optimize short-term or long-term outcomes ● Timing – when ○ Exploits variability in price sensitivity ○ Exploits individual purchasing cycles ● Outreach/budgeting – how many ○ Exploits variability in propensity ● Promotion properties – what ○ Aggregated view on a promotion calendar
- 5. Approach Retailers Brands Product • Willingness to pay • Stages of journey • Affinities to brands • Affinities to channels Predictive Models (Digital Twins) • Propensity • Life-time value • Demand Economic Models • What-if analysis • Optimization • Opportunity finding • Business objectives • Constraints Controls • Offers • Channels • Messages • Prices Signals Decisions
- 7. 7 Incremental revenue Acquisition Maximization Retention time New Cardholder $/brand current non-buyers + high propensity to buy new product current buyers + high propensity to buy more current buyers + high propensity to buy less Product Trial Replenishment Category Stretch Retention Alarm Com petitive Defence Look Alike Modeling and Survival Analysis
- 8. 8 Look Alike Modeling and Survival Analysis time no purchase Model training Model scoring purchase no purchase behavioral history outcome Customer profiles for training Customer profile to be scored score
- 9. 9 Look Alike Modeling and Survival Analysis: Target Metric Design behavioral history outcome Unconditional propensity: Expected LTV: click/purchase/CTR 3-month spend Response/value uplift:
- 10. Challenges with Basic Propensity Scoring 10 Checking Account Credit Card Brokerage Account Banking / Telecom Customer maturity Product maturity level Retail ● Does not take into account product sequences ● Does not optimize offer sequences (i.e. not strategic) ● Requires separate models for different products/offers/objectives time
- 11. profile value (LTV / ROI)M Offer 3 Offer 2 Offer 1 profile value (LTV / ROI)M Offer 3 Offer 2 Offer 1 Next Best Action Model - Naive Approach 11 profile value (LTV / ROI)M Time Offer 1 Offer 2 Offer 3 Offer 3 Offer 2 Offer 1
- 12. Refresher - Reinforcement Learning 12 ● Most basic scenario - Markov decision process (MDP) ○ State ○ Action ○ Reward ○ Value ● Most basic solution - Dynamic programming (DP) ● Two major challenges: ○ The number of states and actions can be large or infinite ○ States and rewards are not known in advance action s1 s2 s3 reward Time
- 13. Next Best Action with Reinforcement Learning 13 Customer state, t action1 action2 action3 reward32 reward33 reward34 Customer state, t+1 Customer state, t+2 Customer state, t+3 Expected LTV / ROI Q(s, a) One timer Churner Repeater Loyal customer Multi product ● Need to estimate an action-value function given a certain offer policy: State (customer feature vector up to moment t) Action (offer feature vector) ● Use Q-function to optimize the offer policy s1 s2 s3 s4 s5
- 14. Next Best Action with Fitted Q Iteration (FQI) 14 Purchase Visit No action Offer 1 Offer 2 Offer 3 2. Initialize approximate repeat 1. Generate a batch of transitions (each trajectory corresponds to 4 transitions): { (state, action, reward, new state) } A simplified test dataset is shown for illustration 3. Initialize training set 4. For each 5. Learn new from training data
- 15. Next Best Action with FQI 15 Offer 3 Offer 2 Offer 1 (default) Low state V High state V Customers who got Offer 3 in early Customers who got Offer 2 early Customers who got Offer 2 -> Offer 3 Customers who did not get offers or got Offer 1 ● Max value for each state: ● Next best action for each state (policy): A simplified test dataset is shown for illustration
- 16. Next Best Action with FQI 16 ● A generalization of the look alike modeling for multi-step and/or multi-choice strategies ● More control over LTV/ROI metrics ● Can evaluate performance of a new policy based on historical trajectories ● Batch-online learning trade-off: multi armed bandits
- 17. Budgeting Models and Decision Automation
- 18. Privileged and Confidential 18 Targeting Thresholds: Static Optimization High propensity Low propensity
- 19. Privileged and Confidential 19 Targeting Thresholds: Dynamic Optimization time $$ campaign duration target budget Decrease propensity threshold Increase propensity threshold
- 20. 20 Campaign Parameters Optimization Purchase trigger buy <X buy X+ buy 0 buy 1+ Announcement Buy X or more units and save on your next shopping trip! Promotion Y% off 1. Estimate demand elasticity 2. Estimate how many consumers will buy more, how many will redeem offers 3. Do break-even analysis for costs and benefits
- 21. 21 Solution Design: Technical Perspective Marketing Manager Campaign Template ● Steps ● Offer types ● Forecasting logic Targeting Score (Look Alike or Next Best Action) Timing Score (Replenishment) LTV Score (Monetary) Offer Database Profile Database Campaign Planner Targeting Server Forecasting Optimization Targeting decisions Budgeting decisions request response Marketing Manager (merchant) Decision automation Customer models
- 22. Objective Selection Plan and Forecast Review User Experience Execution and Measurement Privileged and Confidential 22 Solution Design: Marketer’s Perspective
- 23. Thank you!