With customer acquisition cost (CAC) rising so rapidly, marketers are increasingly under pressure to extract more value per customer. They’ve long known that the key to driving customer lifetime value (LTV) was through upselling and cross-selling. But maximizing LTV isn’t simply trying to sell an existing customer any one of your other products. Instead, you need to understand what items your customers frequently purchase together. That’s where market basket analysis (MBA) comes in.

We discussed market basket analysis for ecommerce in a previous post, but its value cannot be overstated. The saturated DTC marketplace is forcing many firms to devote more marketing dollars first to awareness; then to nurturing them down through the funnel to conversion; and finally, to keeping them onboard. And as we saw in our discussion on churn analysis last week, customer loyalty is short and often fickle. Extracting customer value is more challenging than ever before.

Fortunately, MBA can be used not only for upselling and cross-selling, but also for reducing churn and promoting longer customer loyalty. And the best part about MBA is that customers already tell you what they want together, based on their purchases.

What Market Basket Analysis Reveals

Recall from our previous post that MBA finds products frequently purchased together and uncovers meaningful relationships between them.

At its simplest, MBA produces rules like: “If a customer buys A, they are likely to buy B.”

For example:

• If customers buy peanut butter, they often buy jelly.
• If cable subscribers purchase the mid-tier package, they often add a premium movie channel.

 

What was once considered “nice to know” is now one of the most actionable – and overlooked – sources of revenue growth.

Four High-Impact Ways to Use MBA

Market basket analysis can be applied in several practical ways to increase both average order value (AOV) and LTV.

Strategically Curated Bundles

When you know items that customers frequently purchase together, you can bundle those items into pre-packaged combinations, simplifying the buying decision and increasing perceived value.

If you run a DTC beauty brand and discover that customers frequently purchase cleanser, serum, and moisturizer together, you can create a “Complete Routine” bundle with added recommendations on your website’s product pages. This can help increase AOV and accelerate customer onboarding into a full regimen.

Bundling is straightforward: you can start by identifying the top co-purchase pairs and test bundle pricing vs. individual purchases.

Recommender Systems

MBA often powers “frequently bought together” recommendations, but it can also support broader recommendation strategies.

We’re all familiar with Amazon’s recommendations for items and books like what you’ve previously viewed and purchased. Or Netflix’s recommendations for your next binge watch. These companies have mastered MBA for recommendation strategies.

In like manner, a specialty food brand might notice that customers often co-purchase crackers, cheese spreads, and wine accessories. This insight can be used to:

• Recommend complementary items on product pages
• Suggest add-ons in the cart
• Trigger follow-up emails

 

Recommender systems are helpful for increasing both conversion rates and basket sizes.

Checkout Upsells (Low-Friction Revenue)

Suggesting add-ons right before purchase is another effective way to leverage market basket analysis. The customer is already committed, and little friction is involved in the decision.

Recently, I was browsing flatware sets on Williams-Sonoma’s website. I added a set to my cart and at checkout, I was prompted with some impulse purchase opportunities: serving tongs, a flatware in-drawer organizer, and a set of cutting and carving boards. I added the boards.

These add-ons often convert at high rates because the customer has already committed to the purchase. These upsells are especially valuable if they move low-cost, high-margin items.

Post-Purchase Upsell Sequences (Lifecycle Growth)

The sale is not the end of the relationship – it’s the beginning of the next opportunity.

A pet care brand might observe that customers who purchase treats often return within 30 days and add supplements. This insight can power:

• Post-purchase emails
• Subscription offers
• Timed follow-ups

 

These actions deepen engagement and increase lifetime value.

Going Beyond Basics: What Drives Real Revenue

To drive the most revenue from market basket analysis, focus on three considerations: value, sequence, and segmentation.

Focus on Value, Not Just Frequency

Not all frequently purchased combinations are worth promoting.

Promoting high-frequency items with low margins won’t drive meaningful profit. Instead, prioritize combinations with strong margins and high incremental value.

Sequence Matters

MBA is not just about products customers buy together, but also the products they buy next.

Having bought flatware from Williams-Sonoma, my next purchase might be a tablecloth, and a food processor after that.

From these purchases, Williams-Sonoma will discern that I am gradually outfitting my kitchen. If they notice this purchase sequence across many customers, they might send a post-purchase email or SMS to customers who just purchased the same flatware set to promote the tablecloth.

By paying attention to the sequence in which customers buy products, marketers can gain clues about the next product to recommend.

Not All Customers Behave the Same

Customers should be segmented by their behavior just as much as their demographics and lifestyles. Just because several customers purchase under-bed storage containers, over-the-toilet shelving, and closet organizers at the same time doesn’t mean that’s a rule you should apply to all customers.

If customers are buying those three items together, it can mean they’re trying to maximize space and functionality. Those customers can be elderly couples downsizing or young adults renting their first apartment.

Each group requires a different messaging and cross-sell strategy.

How to Get Started (Without a Data Science Team)

You don’t need advanced machine learning to begin. You can start small and simple.

First, obtain order-level transaction data. Use that to identify the top co-purchase pairs. Prioritize the combinations with the highest margins. Then test a single channel, like cart recommendations, and then expand to bundles and post-purchase emails.

As you scale, you can incorporate more advanced techniques.

Your Customers Are Already Telling You What to Sell Next

The concept of market basket analysis is not complex; it’s practical. Knowing what customers frequently co-purchase or their purchase sequence offers valuable clues about bundling, recommending, or following up on products your customers value.

Market basket analysis turns existing purchase data into clear, actionable opportunities for growth. Start small, test quickly, and build from there.