Repeat Purchase Analytics: Which Products Actually Drive Customer Loyalty?

Two products. Same store.

Product A: 14,200 orders this year. Top seller. You feature it in ads, emails, and the homepage hero. High ROAS, high visibility.

Product B: 870 orders. A mid-tier performer by sales volume. It doesn’t get much promotion time.

But here’s the number your Shopify dashboard isn’t showing:

Product A customers buy again 9% of the time. Product B customers buy again 61% of the time.

If you’re trying to build a loyal customer base (and with acquisition costs rising, you should be), you’ve been optimizing in the wrong direction. The product driving your acquisition is the same one undermining your retention.

This is what repeat purchase analytics by product actually reveals, and why it changes every downstream decision.

Why Sales Volume Is the Wrong Lens

Revenue and volume are easy to report. They’re front and center in Shopify, GA4, and every dashboard your team looks at.

Repeat purchase rate by product is harder to surface. Most analytics tools only report it at the customer or store level: “your store has a 24% repeat purchase rate.” That tells you something useful about overall retention health, but nothing about which products are driving that number and which are dragging it down.

Without the product-level breakdown, you’re promoting your best sellers. With it, you’re promoting your best customer-builders. These are often very different products.

For context on why product-level data is categorically more useful than aggregate or session-level metrics for eCommerce decisions, what is eCommerce product analytics covers the distinction in full.

Three Categories Every Catalog Falls Into

When you break down repeat purchase rate by product, most stores find their catalog naturally sorts into three groups:

High-volume, low-repeat products. These drive top-line sales but not customer loyalty. They’re often discounted, highly searchable, or easy to impulse buy. If these are your main acquisition products, you’re spending to attract customers who mostly don’t come back.

High-repeat, lower-volume products. These are the loyalty builders. They might be consumables, premium items, or simply products customers love and reorder. First-time buyers of these products are disproportionately likely to make a second, third, and fourth purchase. These deserve much more prominence in acquisition and email strategy than they typically get.

Low-volume, low-repeat products. These neither sell well nor retain customers. They exist in the catalog consuming margin, inventory space, and sometimes ad spend. Regular repeat purchase analysis surfaces these so you can make deliberate decisions about what to keep, cut, or reformulate.

The key strategic shift: stop asking “what’s selling?” and start asking “what’s keeping customers?”

A Concrete Scenario

A home goods store with 95 products has an overall repeat purchase rate of 27%. Looks decent. But pulling the product-level breakdown:

• All-purpose cleaning spray: 22,000 orders, 14% repeat rate
• Reusable kitchen sponge set: 3,100 orders, 71% repeat rate
• Bamboo cutting board: 8,800 orders, 18% repeat rate

The cleaning spray is the top seller. It runs in acquisition ads. It’s the first product in the welcome email sequence. But 86% of customers who buy it never come back.

The sponge set, which the team nearly delisted last year due to low sales volume, has a 71% repeat rate. Customers who buy it are highly likely to return, and when they return, they tend to increase average order size.

Moving the sponge set into a more prominent acquisition position changes the unit economics significantly. The first-order ROAS might be slightly lower. But the cost per loyal customer drops substantially, and that’s the metric that predicts 12-month revenue.

[Run the repeat purchase rate report on your own catalog in Stormly. Start your free trial and see which products are actually building loyalty.]

How to Read a Repeat Purchase Rate Report

A few things to watch for when you first pull this data:

Control for product age. A product launched six weeks ago will have a naturally lower repeat rate than one in the catalog for two years. Filter for products with at least 150 to 200 lifetime orders and six or more months of history before drawing conclusions.

Compare within category, not across. A 20% repeat rate for a luxury item might be exceptional. A 20% repeat rate for a daily-use consumable might be poor. Use category benchmarks, not store-wide averages, as your reference point.

Look at time-to-repeat, not just rate. A product with a 40% repeat rate and a 15-day median time-to-repeat is a completely different animal than one with 40% repeat at a 150-day median. The first is a habit product. The second might just have a long natural purchase cycle. Stormly’s repeat purchase report includes median days-to-repeat alongside the rate, so you can distinguish between these patterns.

Flag outliers in both directions. Products with unusually high repeat rates within their category deserve immediate strategic attention. Products with rates well below their category average might have a pricing problem, a quality issue, or simply a product-market fit gap that revenue numbers are masking.

This connects directly to cohort analysis. Cohort analysis for eCommerce goes deeper on how the first product a customer buys predicts their 90-day retention trajectory, which is one of the most actionable insights you can pull from order data.

Three Decisions This Changes

Acquisition targeting. Most paid acquisition campaigns optimize for ROAS on the first order. If you know which products convert first-time buyers into loyal customers, you can run an additional optimization layer: first-order-to-loyal-customer rate. Two products might have identical first-order ROAS, but if one creates repeat buyers at 5x the rate, that difference compounds across every future purchase. Stormly connects acquisition source to product-level repeat purchase behavior so you can see this relationship directly.

Email and post-purchase flow design. If a product has a 65% repeat rate, the first post-purchase email should reinforce that relationship, not immediately pivot to an unrelated upsell. If a product has a 12% repeat rate, the post-purchase email might be better spent surfacing a high-loyalty product as a complement. Neither decision is possible without product-level repeat data. Pairing this with eCommerce customer retention analytics gives you the full picture: which customers are at risk, not just which products they bought.

Catalog and merchandising strategy. Products with consistently low repeat rates and low sales volume are not neutral. They consume inventory, margin, and merchandising attention. Repeat purchase analytics gives you data for catalog cleanup decisions alongside revenue and margin metrics.

How This Connects to LTV

Here’s the insight most LTV calculations skip: a customer’s predicted lifetime value doesn’t start with their demographic profile or acquisition channel. It starts with the first product they buy.

If your serum kit has a 69% repeat rate and the average customer who buys it makes 4.1 orders over 18 months, that SKU isn’t just a product in your catalog. It’s your highest-LTV acquisition vehicle, if you’re measuring for long-term revenue rather than first-order returns.

This changes how much you should be willing to spend to acquire a customer through it. Even if the immediate ROAS looks unflattering compared to your high-volume low-repeat product, the downstream math is very different.

Customer lifetime value analytics at the product level covers how to calculate first-purchase LTV by product category and how to set smarter CAC targets by channel once you have this data.

The Analytics Gap Most Tools Leave Open

Amplitude, Mixpanel, GA4 all track repeat behavior, but at the wrong granularity for eCommerce product decisions.

GA4 tracks sessions and events. It can tell you someone visited three times and made two purchases. It cannot tell you: “Of the 4,200 customers who bought SKU 1047, how many made a second purchase within 90 days, and how quickly?”

Mixpanel handles event-based cohorts well for SaaS products. It doesn’t have a native concept of an eCommerce product catalog.

Answering the repeat purchase rate question at the product level requires matching order history to your product catalog and calculating rates and time-windows by SKU. That’s typically an analyst task or a custom SQL job. In Stormly, it’s a standard report because the tool was built specifically for eCommerce product decisions, not web app analytics.

For teams tracking at-risk customers alongside repeat purchase behavior, how to predict eCommerce customer churn before it happens covers how leading behavioral signals can identify customers likely to leave 30 days before they actually stop buying.

Getting Started

If you’re running this analysis for the first time:

1. Pull your repeat purchase rate by product. Filter to items with at least 200 lifetime orders and six months in catalog.
2. Sort by repeat rate, not by sales volume. The list you get will look different from your usual sales report.
3. Identify your top five loyalty products. Check whether they have acquisition-level visibility in your ads, welcome email sequence, and homepage.
4. Identify your bottom five (excluding very new products). Check whether they’re receiving disproportionate marketing budget relative to the loyalty they generate.
5. Build one test: put your highest-repeat-rate product into a top-of-funnel campaign and measure the 90-day repeat purchase rate of customers acquired that way versus your current top seller.

One session with Stormly’s repeat purchase breakdown typically surfaces two or three catalog decisions that were invisible in the aggregate revenue report. What used to take a half-day spreadsheet exercise takes about 10 minutes.

Find the products that build loyal customers. [Start your free trial in Stormly.]