Academy · Lesson 4 · Cohorts
Cohorts compare customers who started together,
not all customers at once.
A cohort is a group that started in the same period, such as all first-time customers in January. Instead of one average across everyone, a cohort table shows whether newer groups stay longer than older ones. This lesson builds such a table in three steps.
Where do you want to make progress?
Export orders as CSV
Cohorts compare customers who started together,
not all customers at once.
Start month × months since start.
A cohort is a group that started in the same period, such as all first-time customers in January. Instead of one average across everyone, a cohort table shows whether newer groups stay longer than older ones. This lesson builds such a table in three steps. Example (fictional): 200 first-time customers in January, 30 of whom buy again the following month (15%) and 22 in the second month (11%). The February cohort sits one row below and is directly comparable.
What you need.
- A list of customer and order date, for example as a CSV export from Peak Analytics.
- A written definition of "active" that everyone shares.
- A spreadsheet; the formulas are simple counts and divisions.
- January cohort: 200 first-time customers (example Month 1: 30 (15%) · Month 2: 22 (11%))
- February cohort: 160 first-time customers (example Month 1: 28 (17.5%) · Month 2: 19 (about 12%))
The learning path
Starting group, activity, comparison.
Find each customer's start month
For every customer, the date of their first order, first receipt or contract signature. That is their cohort, and it never changes.
- Export orders as CSV
- Note each customer's first order as their cohort
- Every customer has exactly one start month
Every customer has exactly one start month
If you get stuck
When the number doesn't fit.
The newest cohorts are short
In July, the May cohort has only two months. Only compare columns that every cohort you look at has already reached.
Small cohorts jump
With 20 customers, one customer is five percentage points. Show the cohort size at the start of each row and merge very small months into quarters.
What the preview shows today
Cohort analysis is planned in Peak Analytics. Today the preview shows orders per month split into new and returning customers. That tells you how much business comes back, but not from which starting group.
Your quality check
Before you apply it to your numbers.
Tick what you have checked. Your selection stays in this view only.
0 of 3 points checked
What this article is based on.
Editorial page with fictional examples (Nordwind Travel Goods (fictional), Taktwerk Software (fictional), Harbourview Agency (fictional)). The guide describes an approach; where Peak Analytics covers a step today, the page says so explicitly. No measurements, no customers, nothing is sent or saved.
- Editorial demo
Synthetic numbers in the browser. No real data is read, computed or saved.
- Checked evidence
Peak Analytics is a preview: pivot over orders, receipts, tasks, goals and files; count, sum, average, minimum, maximum; day to year; CSV export; sharing a released snapshot; largest value as a task.
peak-one app/w/[scope]/apps/analytics, lib/bi/model.ts, lib/views/pivot.ts, lib/bi/share.ts · checked 2026-10-29
Same start, fair comparison.
In lesson 5 the cohort table becomes a retention rate you report every month.
Frequently asked questions
Do I need Peak Analytics for this lesson?
No. The lesson works with any spreadsheet. Peak Analytics is a preview; where it covers a step today, the page says so.
What is a cohort?
A group that started in the same period, such as all first-time customers in January. Its start month never changes.
How do I read a cohort table?
Column by column, top to bottom: the same column, such as month 3, shows for each cohort the share still active at that point.
Can Peak Analytics analyse cohorts?
Not yet; cohort analysis is planned. The preview shows orders per month by new and returning customer, but not which starting group they come from.