CHURN in Retail and how can CUSTOMER DATA ANALYTICS help prevent it?
Download as DOCX, PDF0 likes159 views
Churn in retail refers to existing customers who stop shopping at a store and begin shopping elsewhere. While churn may not significantly impact overall market penetration metrics, retaining existing customers is more cost effective than acquiring new ones. Customer data analytics can help identify patterns in shopping behavior that indicate a customer is at risk of churning. A predictive model can be developed using variables identified by analyzing changes in customer spending patterns, perceptions, market competition, behavioral segments, pricing and promotions over time to predict which customers are most likely to churn. Those at highest risk can then be targeted with personalized retention campaigns.
CHURN in Retail and how can CUSTOMER DATA ANALYTICS help prevent it?
- 1. CHURN in Retail and how can CUSTOMER DATA ANALYTICS help prevent it? What is churn in retail? Churn can be defined in many ways. It depends on how businesses want to measure. It could be Shoppers who have missedtwo or three purchase cycles in a row or the shoppers who didn’tshopin the past 3 monthsand so on...For thisarticle sake,we keepitsimple, Shoppers who were shopping in your stores earlier are now shopping elsewhere is CHURN. Churn can be overlookedinbigschemaof thingsif youdon’t make itpart of the keyKPIsthat businessis measuredon.Forexample:Market“penetration” akeyKPIcouldremainsame thisyeartolastyear,which may not call for any action, but underneath, the business has added several “new” shoppers and lost (churn) “existing”shoppers. NewandLost shoppersbalanceditout to keepthe penetrationthe same in boththe years. In yourmarketingportfoliowhereisyourbudgetskewedtowards?AcquiringNewBuyers or Retaining and Maximizing your existing buyers? Keeping your existing shoppers is cheaper than acquiringnewones. Whatif the business couldstopthe churnbyacertainpercentage?Yourpenetration couldhave beenhigherthanearlieryear. So it is importantto look at the CHURN RATE alongwith other KPIs so you could do something about it. So, what can you do about churn? Before we go do something about CHURN – first, it is important to understandthe reasons. While customerdataanalyticsis“not”goingtoprovide youthe reasonsforchurn butit isgoingto helpyouunderstandthe changesinthe shoppingpatternsbefore theyleaveyou.One of the patterns that I have observed is that the shoppers migrate to lower tiers / become less loyal before theycompletelyabandonyou. A predictive /regressionmodel canhelpyouidentifythose shoppersthat are more likely to churn so that you can take further steps to prevent it. Easiestpart of the process isto builda predictive model,butthe difficultpartis the work that goes into identifying the variables. Step1: Identify the churners based on the definition that business has provided Step 2: Understand the shopping patterns of the churners before they have left you. What has changed and how much it has changed
- 2. Examples: Overlaysurvey datathatcouldshed some lightonshopperperception.There couldbe something aboutcleanliness,productavailabilityandotherfactorsthatmightbe turningoff these shoppers Are these shoppers primarily in those markets where competition is fierce Overlay with behavioral segments and other segments (including digital behavior) available internally to see if there certain segments who are churning more. Pricing and Promotional decisions overtime and their correlation to churn Top skewing categories / products among these and what has changed in the past All these insightswill helpindefiningthe variablesthatgo into the model forprediction. Successof this piece depends on how integrated your different data sources are or if you have “single customer view” data strategy Step 3: Develop and Run the model Step4: Take customersinthe top3 decilesof the modelanddevisearetentioncampaign.All the insights that were acquired in step 4 will help personalize the campaign