Cohort Analysis: Track How Customer Groups Behave Over Time

What Is Cohort Analysis?

Cohort analysis groups your customers by a shared characteristic — usually the date they first purchased — and tracks their behavior over time. Instead of looking at all customers as one blob, you examine each group separately to understand patterns.

For example, customers who bought in January are one cohort. Customers who bought in February are another. By comparing these cohorts, you can see whether your business is improving over time.

Why Cohort Analysis Matters

Traditional metrics can be misleading. Imagine your overall revenue is growing 20% month over month. That looks healthy. But cohort analysis might reveal that all the growth comes from new customers, and every cohort stops buying after their first purchase. Your business is a leaky bucket that requires ever-increasing ad spend to maintain.

Conversely, flat overall revenue might hide the fact that newer cohorts have higher repeat purchase rates than older ones, meaning your business is actually improving.

Cohort analysis separates the signal from the noise.

Types of Cohort Analysis

Acquisition Cohorts

Group customers by when they made their first purchase. This is the most common and useful type.

Example: All customers who first purchased in March 2026 are tracked together. You measure what percentage of them purchase again in April, May, June, and so on.

Behavioral Cohorts

Group customers by an action they took.

Example: Customers who used a discount code vs. those who paid full price. You might find that discount customers rarely return while full-price customers have higher lifetime value.

Channel Cohorts

Group customers by how they found your store.

Example: Facebook ad customers vs. TikTok ad customers vs. organic search customers. This reveals which acquisition channel produces the most valuable long-term customers.

How to Perform Cohort Analysis

Step 1: Export Your Data

You need two pieces of information for each customer:

1. Date of first purchase
2. Dates and amounts of all subsequent purchases

Export this from your payment processor (Stripe) or your store's order database.

Step 2: Create Your Cohort Table

Organize data in a table where:

• Rows represent cohorts (e.g., January customers, February customers)
• Columns represent time periods after first purchase (Month 0, Month 1, Month 2, etc.)
• Cells contain the metric you are tracking (repeat purchase rate, revenue, etc.)

Step 3: Calculate Retention Rates

For each cohort, calculate what percentage of customers made a purchase in each subsequent period.

Example table:

This table immediately shows that newer cohorts have improving retention rates (March cohort: 12% Month 1 vs. January cohort: 8% Month 1).

Step 4: Visualize the Data

Use color coding in your spreadsheet (green for high, red for low) to make patterns visible at a glance. Google Sheets and Excel both have conditional formatting that works well for this.

What Cohort Analysis Reveals

Improving or Declining Retention

Are newer cohorts more likely to return than older ones? If yes, your product and customer experience are improving. If no, something is deteriorating.

True Customer Lifetime Value

By tracking revenue per cohort over time, you can calculate actual customer lifetime value (LTV) rather than estimating it. This tells you how much you can afford to spend acquiring a customer.

Impact of Changes

Did you redesign your product page in March? Compare the March cohort's behavior to February's. A significant improvement in March's metrics validates the change. No difference means the redesign did not matter.

Channel Quality

Customers from different channels may have dramatically different lifetime values. If TikTok customers rarely return while Facebook customers buy three times, you should invest more in Facebook even if TikTok has a lower initial CAC.

Seasonal Patterns

Holiday cohorts often behave differently from regular cohorts. Black Friday buyers may be bargain hunters who never return, while organic summer buyers may have higher loyalty.

Building a Cohort Analysis in Google Sheets

1. Create a sheet with columns: Customer ID, First Purchase Date, Order Date, Order Amount
2. Add a column for Cohort (month and year of first purchase)
3. Add a column for Period (number of months between first purchase and this order)
4. Create a pivot table with Cohort as rows and Period as columns
5. Use COUNTIF or SUMIF to populate cells with customer counts or revenue
6. Divide each cell by the Month 0 count to get retention percentages
7. Apply conditional formatting for visual clarity

This takes 30-60 minutes to set up but provides insights you cannot get any other way.

Cohort Analysis for Dropshipping

Dropshipping stores often have low repeat purchase rates because they sell single products. This does not mean cohort analysis is useless.

Even for single-product stores, cohort analysis reveals:

• Whether refund rates differ by cohort (product quality changes?)
• Whether different ad campaigns produce different quality customers
• Seasonal patterns in purchasing behavior
• The true effectiveness of email marketing on repeat purchases

For multi-product stores or stores with consumable products, cohort analysis is essential for understanding lifetime value and optimizing acquisition spending.

Key Takeaways

• Cohort analysis groups customers by acquisition date to reveal patterns hidden in aggregate data
• Retention trends across cohorts show whether your business is truly improving
• Customer lifetime value calculations require cohort data for accuracy
• Channel cohorts reveal which traffic sources produce the most valuable customers
• A simple Google Sheets setup is sufficient for most stores
• Even single-product stores benefit from cohort analysis for refund rates and channel quality insights