What Is Cohort Analysis? Retention & Revenue Cohorts

Cohort analysis is a way to study groups of customers (or users) who began together, then follow that same group across weeks or months to see retention, revenue, and engagement trends. Instead of blending everyone into one average, you compare January buyers to February buyers, or mobile signups to web signups, on equal footing.

For D2C brands, SaaS teams, and any business with repeat purchases, cohort views answer whether product, pricing, or acquisition changes are actually improving outcomes, not just headline revenue.

This page defines cohorts, explains retention and revenue cohort views, and shows how FireAI turns transactional data into live cohort dashboards. For customer analytics workflows in Indian D2C, see D2C e-commerce customer use cases.

What Is a Cohort?

A cohort is a set of customers who share a defining event and time window, such as:

• Acquisition cohort: first order month or first paid subscription month
• Behavioral cohort: first app install week, first campaign response, or first purchase from a specific channel
• Product cohort: customers who first bought a given SKU or bundle

The cohort label is the cohort period (for example, "2025-09 acquisitions"). Every later period measures what share of that original group is still active, still ordering, or still generating revenue. That is different from RFM analysis, which scores where each customer stands today; cohort analysis emphasizes how a fixed group evolves through time.

Retention Curves

A retention curve plots the share of a cohort that remains active at period 0, 1, 2, and so on after the start event.

Typical steps:

1. Define "active." For e-commerce, often "placed an order in that month"; for subscriptions, "renewed" or "logged in."
2. Anchor each customer to one cohort based on first qualifying event.
3. For each period after the start, calculate the percentage of the original cohort still meeting the active rule.

A curve that stays flat and high indicates strong repeat behavior. A curve that drops steeply after period 1 often signals onboarding, quality, or channel issues. Comparing curves across cohorts (for example, Meta ads vs organic) shows which acquisition sources bring buyers who actually stick around.

Revenue Cohorts

Revenue cohort analysis tracks how much money each acquisition group generates over time, not only whether they are still active.

Common views:

• Cumulative revenue per cohort: total rupees from the September cohort through month 3, 6, and 12
• Average revenue per retained customer: avoids masking a small group of whales
• Payback views: when cumulative gross margin crosses CAC for that cohort (often paired with unit economics)

Revenue cohorts are especially useful when AOV or repeat rate changes. Retention might look stable while revenue per cohort drifts down because order sizes shrank. Seeing both curves keeps finance and growth aligned.

How FireAI Builds Cohort Dashboards from Order Data

Manual cohort reporting usually means exporting orders, building pivot tables, and refreshing late. FireAI connects to sources such as Shopify, marketplaces, and payment data so cohort dimensions stay current.

What teams typically automate:

• Acquisition cohorts by month, week, or channel
• Retention tables and curves with configurable "active" rules
• Revenue and margin cohorts when COGS or returns are available in the data model
• Breakdowns by product category, discount cohort, or geography for Indian multi-channel brands

Natural language questions: Ask which cohorts improved after a pricing change or which channel's Month 6 retention lags, without rebuilding a new sheet each time. That is the practical bridge between predictive analytics planning and day-to-day monitoring.

For operators who want the full D2C analytics picture in India, D2C brand analytics sits alongside cohort work as a broader hub.