Skip to main content

Retention: Cohort Analysis

How do I read the Cohort Analysis, how is it calculated, and what insights can I gain from it?

Written by Frank Birzle

tl;dr

The Cohort Report clusters your customers based on when they made their first order — one group of new customers from a specific time frame is one cohort. It lets you measure the retention and monetisation of your customers over time and highlight areas for improvement.

When you have configured your cost datasources, the Cohort Report also enables you to set Customer Acquisition Cost (CAC) targets based on how quickly you want to break even — and monitor whether you are hitting them.

The Cohort Report helps you uncover seasonal effects on Customer Lifetime Value and monitor the health of your new customers as you grow.

Date range note: The date range you define selects customers acquired in that period — but all their activity up to today is used to calculate the values. More on date range definition


How does it work?

The Cohort Report is a great tool to understand your customers' repeat purchase behaviour over time and shows you how it changes as you grow.

By default, the report groups customers on the vertical axis by the month they placed their first order. All customers who first ordered in January 2026 form the Jan 2026 cohort. The New Customers column shows how many people placed their first order in each month.

The horizontal axis shows elapsed time relative to the starting month. The example below uses Net Orders as the selected metric. The columns are:

  • First Order — the raw first order placed by each customer in the cohort month (e.g. January 2026). This is Month 0 and always equals 1 order per customer

  • 1 — follow-up orders placed in the same calendar month as the first order (e.g. still in January 2026)

  • 2 — follow-up orders placed in the next calendar month (e.g. February 2026)

  • 3 — follow-up orders in March 2026

  • 4 — follow-up orders in April 2026

  • and so on...

⚠️ Note: Cohorts are based on calendar months. If a customer places their first order on the 17th of January, Month 0 only contains their activity for the remaining 14 days of January.

Want a different granularity? Use the Timeframe toggle to switch between weeks, months, or quarters.

By using elapsed time on the horizontal axis, you can directly compare the retention and monetisation of different cohorts — regardless of when they were acquired.


What can I analyse?

Display options

Metric Display:

  • Incremental — shows only the value for activities that took place in that specific month

  • Accumulative — adds up values from the current and all prior months

In most cases, looking at accumulative values makes it easier to compare different cohorts. For some metrics like AOV, incremental values are more useful.

Metric Summary:

  • Per Customer — divides the total by the number of customers in the cohort. Use this to understand cohort quality, independent of cohort size

  • Total — shows the total sum. Use this to understand overall business impact

Metrics

  1. Net Revenue — Net Revenue generated from the cohort's customers over the months

  2. Active Customers — the number of customers in the cohort who placed at least one order in the respective month

  3. Contribution Margin 1 — CM1 generated from the cohort's customers over the months

  4. Contribution Margin 2 (CLV) — CM2 generated from the cohort's customers over the months. The "(CLV)" label highlights that accumulative CM2 per customer is the closest proxy for Customer Lifetime Value in Klar

  5. Contribution Margin 3 — CM3 generated from the cohort's customers over the months

  6. Gross Orders — total (gross) orders placed by the cohort's customers over the months

  7. Second Orders (Gross) — number of customers who placed a second gross order

  8. Average Order Value — Net AOV of all orders placed by the cohort's customers

  9. Net Orders — net orders placed by the cohort's customers (after returns)

  10. Second Net Orders — number of customers who placed a second net order

  11. Gross Average Order Value — Gross AOV of all orders placed by the cohort's customers


What can I use this for?

The Cohort Report is not a daily report. It highlights mid- to long-term trends in your retention, profitability, and overall business health. Take 30 minutes a month or an hour a quarter to work through it and identify larger trends.

Define your CAC target

Set yourself a target: you want to turn a profit from each new customer within 3 months of their first order. To find out how much you can spend per new customer, look at the accumulative CM2 after 3 months in the cohort report.

CM2 is your profit before marketing costs — so the accumulative 3-month CM2 tells you what your CAC target should be to break even at 3 months.

Monitor CLV and break-even

Once the CAC target is set, use the cohort report to monitor whether you are staying within it. Look at the accumulative CM2 per customer across months to see when each cohort becomes profitable.

Monitor new customer quality as you grow

A common pattern: as you scale acquisition, the Customer Lifetime Value (CLV) of new customers shrinks. The early customers were the easiest to convince — as you reach a broader audience, loyalty tends to be lower.

The cohort report helps you spot this early. If customers are returning less frequently or with smaller baskets, it's time to invest more in retention. The impact of those efforts can also be measured here.

Identify what drives retention

There are three main levers:

  1. Second order rate — how many customers place a second order at all?

  2. Purchase frequency — how often do repeat customers buy?

  3. Repeat AOV — what is the basket size of a repeat customer?

Most shops struggle with the first one. Understanding exactly where you stand lets you define the right actions.

Identify seasonal effects on CLV

Some shops see significant differences in CLV depending on when customers were acquired. For example, customers acquired in November via Black Friday promotions may be less loyal than customers acquired in June. The cohort report makes this visible.

Filter by campaign or product

Using the filtering options, you can view the cohort report for customers from specific marketing campaigns or who purchased specific products. For this type of comparison, the Cohort Comparison Report is better suited.

Did this answer your question?