- The DTC Newsletter
- Posts
- #19 Product Affinity
#19 Product Affinity
Reveal how past shoppers behave—and how to encourage them to buy again.
On a sunny Saturday afternoon, Hannah strolls into her kitchen, pours herself a refreshing iced coffee, and checks her phone for weather updates. She's thrilled to see a string of 85-degree days in the forecast - perfect for a quick trip to the beach. As she mentally plans her trip, one thing becomes clear: "I need a new swimsuit!"
The Shopper’s Fork in the Road
At this juncture, Hannah can take one of two routes:
Shop with a brand she’s bought from before (Brand A). – Let’s say about 35% of buyers do this when making a new purchase.
Explore the sea of new brands (Brand B). – Roughly 65% venture off to see what else is out there.
Sound familiar? We’ve all been Hannah, toggling between brand loyalty and the allure of something fresh.
Business Reality Check: Winning (and Keeping) Customers
Now let’s shift to your perspective as Brand A. You’ve poured resources into winning Hannah’s business the first time around—targeted ads, stunning lookbooks, maybe even a promo code. Yet here she is, barely a few months later, poised to look elsewhere. Did you really do anything wrong?
Probably not. But you may have overlooked opportunities to keep Hannah engaged—enticing her to come back for that next purchase. That’s where a data-powered retention strategy can make all the difference, especially through product affinity analysis.
Mastering Product Affinity Analysis
Let's consider the example of a typical grocery store. There are a variety of products we assume people will often purchase together at a grocery store: ice cream and waffle cones, peanut butter and jelly, turkey and cream cheese. By performing an affinity analysis on the grocery store’s data set, we can not only confirm our suspicions about which products are frequently purchased together, but also discover new relationships between products and customers that we never would have guessed.
These relationships between products and customers are also known as association rules. After running an affinity analysis, rules are produced in the following form:
{waffle cones} ⟹ {ice cream}
{flour, sugar} ⟹ {eggs}
In other words, if a customer buys Product X, there’s a certain probability they’ll also purchase Product Y—maybe in the same transaction, maybe later on.
Now let's apply this logic to your fashion brand:
Two key metrics come up often:
Support: The percentage of orders in which a specific product or combination shows up.
Confidence: Given the “left” product(s), the likelihood that buyers will also pick up the “right” product(s).
High support means a product pairing or combo appears a lot in your orders. High confidence indicates that if customers buy the first item, they frequently buy the second.
Turning Insights into Action
Targeted Email Campaigns: Build a list of customers who bought the left item but never tried the right one, then send them compelling reasons to try the missing piece.
Personalized Website Prompts: Add “People who also bought…” pop-ups or recommendations on product pages.
Bundles & Discounts: If you spot strong complementary products (like beach towels + sunscreen), consider bundling them or offering a discount to boost average order value.
Segment-Specific Loyalty Programs: Differentiate your loyalty tiers based on what people have purchased—and reward them for completing certain “sets” of products.
Nailing the Timing of Your Offers
Knowing which items go together is only half the puzzle. Timing your next touchpoint is equally critical. Too soon, you risk annoying buyers who aren’t ready. Too late, a competitor might snatch them first.
Let’s say you have three customers:
All three purchased Product A (maybe a particular swimsuit style) but have never tried Product B (a matching cover-up). Your data suggests a strong affinity rule between those items. Do you email them all at once? Probably not.
Examine the average time between orders for customers with 1 lifetime purchase (like Lisa) versus those with 4 (like Hannah). Maybe first-time buyers typically wait five months before purchasing again, while those with multiple orders re-up every two or three months.
By sending promotions around the time your data shows they’re likely to shop again—or a touch earlier if you want to shorten the gap—you’ll greatly increase your odds of engagement.
The time between purchases is primarily driven by the product/category purchased in the previous purchase (also by quantity for CPG brands) and how many purchases a customer has made with your brand (loyalty).
From Data to Actionable Strategy
By combining affinity analysis (the “what”) with your time-between-orders data (the “when”), you create a powerful one-two punch for bringing people back. Instead of generic blasts, you’re delivering tailored promotions at moments when your customers are actually in buying mode.
Experiment: If you see a 90-day lull between purchases, try nudging them at day 70 to see if you can encourage an early reorder.
Upsell/Cross-Sell: Have a sporty swimsuit that often pairs with a rash guard? Send out that recommendation after your data-suggested window.
Segment by Frequency: Casual shoppers may need a longer time to “recover” before buying again, whereas your loyal fans might be ready in just a few weeks.
How to get started
Corporate giants like Amazon and Netflix employ entire data science units for this. With RetentionX, you can accomplish something similar—segment your audiences, recommend related products, and personalize promotions based on concrete buying patterns fully automated.
Let's chat:
That’s it for this edition!
Any questions or topics you'd like to see me cover in the future? Just shoot me a DM or an email!
Cheers,
Alex