Cohort Analysis: Repeat Purchase Behavior (2026 Guide)

TL;DR: Cohort analysis groups customers by their first purchase date to reveal repeat purchase patterns. Learn how to calculate repeat purchase rates, segment your buyers, and use these insights to boost customer lifetime value for your Amazon business.

Key Takeaways

Table of Contents

Note on marketplaces: This guide is specifically optimized for the US market.

What Is Cohort Analysis and Why Does It Matter for Repeat Purchases?

Cohort analysis is a behavioral analytics technique that groups customers based on a shared characteristic or event within a defined time period. In ecommerce, the most common cohort is based on the date of a customer’s first purchase. You then track that group’s behavior over subsequent weeks, months, or years. This allows you to see whether customers from a specific time period are more likely to repurchase, how quickly they come back, and how much they spend.

For Amazon sellers, repeat purchases are the lifeblood of a sustainable business. Acquiring a new customer can cost five times more than retaining an existing one, and repeat customers typically spend more and buy more often. But with millions of products and limited customer data, Amazon sellers often struggle to understand retention patterns. That’s where cohort analysis comes in. It gives you a clear, data-driven view of which customer groups drive repeat revenue, so you can double down on what works and fix what doesn’t.

If you’re new to analytics, start with our comprehensive Amazon Seller Analytics Guide to understand the full picture. Then come back here to master the specific technique that reveals retention and repeat purchase behavior.

How to Calculate Repeat Purchase Rate Using Cohort Analysis

Calculating repeat purchase rate using cohort analysis is straightforward once you understand the steps. You’re essentially dividing the number of customers who made a second purchase during a defined period by the total number of customers in that cohort.

Step 1: Define Your Cohort Period

Choose a time grouping that makes sense for your business. Monthly cohorts are common for ecommerce. Weekly or even daily cohorts can be useful if you have high purchase frequency and large customer bases. For most Amazon sellers, monthly cohorts are a good starting point.

Step 2: Group Customers by First Purchase Date

For each customer, identify the month of their first purchase. Everyone whose first purchase occurred in January becomes part of the 'January cohort.' This is your baseline group. Each cohort is then tracked independently.

Step 3: Track Repeat Purchases Over Time

For each follow-up period (e.g., the next month, two months later, etc.), count how many customers from that cohort made another purchase. A customer can only be counted once per period, even if they bought multiple times in that period. This gives you a distinct repeat buyer count.

Step 4: Calculate the Repeat Purchase Rate

Divide the number of repeat buyers in a given period by the total number of customers in the cohort, then multiply by 100 to get a percentage. For example, if 100 customers first bought from you in January and 25 of them bought again in February, the February repeat purchase rate for the January cohort is 25%.

You can also calculate cumulative repeat purchase rate by adding up all repeat buyers over multiple periods. This gives you a broader view of long-term retention.

Here’s a quick example to illustrate:

• January cohort: 100 first-time buyers.
• February: 25 of those 100 make a second purchase → Repeat rate = 25%.
• March: 40 of the original 100 make a second purchase (including some who also bought in February) → Repeat rate for March = 40%.
• Cumulative repeat purchase rate by end of March = (25 + 40) / 100 = 65%.

This simple math reveals critical insights. If your January cohort’s repeat rate is higher than your June cohort’s, something changed—perhaps a new product launch, a pricing update, or a shift in ad targeting. Cohort analysis helps you pinpoint exactly when retention improved or declined.

Best Practices for Cohort Analysis to Improve Repeat Purchases

Cohort analysis is only as powerful as your implementation. Here are proven best practices to get the most value from your analysis.

1. Segment Beyond Acquisition Date

While first-purchase date is the baseline, you can (and should) segment cohorts by other attributes: traffic source (organic search, PPC, social), product category, geographic region, device type, and even customer tier. This tells you which marketing channels bring in the most loyal customers and which product lines have the highest repurchase rates. For instance, you may discover that customers acquired through Amazon Ads have a 30% repeat rate, while those from social media have only 10%. That’s a signal to reallocate your ad budget.

2. Choose the Right Time Horizon

Depending on your product lifecycle, you might look at 30, 60, or 90-day repeat rates. For consumables, 30-day cohorts reveal rapid repurchase cycles. For durable goods, you may track 6-month or even annual cohorts. Always align your analysis with your actual buying cycle to avoid misleading conclusions.

3. Compare Cohorts Side by Side

The real power of cohort analysis lies in comparing different cohorts. Track each month’s cohort in a row and subsequent months in columns. This matrix format reveals trends such as improving retention over time or seasonal dips. You can quickly spot whether your latest customers are acting differently from your earlier ones.

4. Act on the Insights

Cohort data is useless if it doesn’t change your decisions. If you discover that customers who buy a particular product tend to repurchase within 60 days, create a targeted email or Amazon campaign for that segment. If a specific acquisition channel delivers high repeat rates, increase spend there. If a cohort shows declining repeat purchases, investigate the cause—maybe you changed your product listing or raised prices.

5. Avoid Common Pitfalls

Don’t mix new and existing customers in the same cohort. Choose sample sizes large enough to be statistically significant (at least 50 customers per cohort as a rule of thumb). And remember: cohort analysis is descriptive, not predictive. Use it to understand the past, then combine it with predictive techniques like customer lifetime value (CLV) modeling to forecast future behavior.

Cohort Analysis and Customer Lifetime Value

Customer lifetime value (CLV) is the total revenue a customer generates over the entire relationship with your brand. Cohort analysis is the perfect tool for calculating and optimizing CLV. By tracking how much each cohort spends over time, you can estimate the lifetime value of different customer groups and decide how much to spend on acquiring them.

A simple CLV formula is:

CLV = Average Order Value × Purchase Frequency × Customer Lifespan

Cohort analysis feeds directly into this formula. You can calculate the average order value for each cohort, how often they buy (purchase frequency per time period), and how long they stay active. Combine these to get a more accurate CLV per cohort.

For example, if your January cohort has an average order value of $50, buys twice in the first 90 days, and remains active for 6 months, their CLV is $50 × (2 × 6) = $600. That’s the number you can compare to your customer acquisition cost (CAC) to ensure profitable growth. When you repeat this for every cohort, you can identify which acquisition channels are bringing in high-CLV customers and double down on those.

To dive deeper into CLV and related metrics, check out our complete Amazon seller analytics guide, where we cover CLV alongside other key performance indicators.

How SellerSprite Helps You Run Cohort Analysis

Running cohort analysis manually in spreadsheets is possible, but it’s time-consuming and error-prone, especially if you have thousands of orders. SellerSprite’s analytics platform automates cohort reporting for Amazon sellers. You can see repeat purchase rates by month, segment customers by SKU or channel, and track lifetime value—all in a few clicks.

With SellerSprite, you can:

• Automatically group customers by first purchase date and generate cohort tables.
• Filter cohorts by product, marketplace, and traffic source.
• Visualize repeat purchase trends with charts.
• Export data for deeper analysis.
• Calculate customer lifetime value for each cohort.

This takes the guesswork out of retention and lets you focus on action. If you’re ready to see how your customers really behave, sign up for SellerSprite and start your first cohort analysis today.

FAQ

Next Steps

References

• SellerSprite: Amazon Seller Analytics Guide View
• Harvard Business Review: The Value of Keeping the Right Customers View

Note: The HBR reference is provided for context; verify the exact article before quoting.