Cohort Table Reading
AnalyticsAlso: Cohort Table Analysis · Reading Retention Cohorts
Quick definition
Cohort table reading is the skill of interpreting a cohort table, a grid that tracks a group of customers acquired in the same period against how many are still active over time. It reveals retention rate patterns that a single average churn rate figure hides completely.
How it varies across Australia
How steep the drop looks in a cohort table varies sharply by business model. Subscription products with genuine daily utility flatten quickly after the first month. Low-frequency purchase categories decay for longer before settling. There's no universal shape, only the shape typical of your category.
Compare retention patterns across Australian industries →The three things a cohort table is built from
Each row is a cohort, a group of customers who joined in the same week or month.
Each column is a period since acquisition, month one, month two, and so on.
Each cell shows what percentage of that cohort was still active at that point.
What it actually means
A cohort table looks intimidating the first time you see one, a grid of percentages fading from dark to light as you move right. Most people try to read it like a spreadsheet, scanning left to right along a row. That's the wrong direction.
The useful read is down a column, or diagonally across the newest cohorts. Reading down month three tells you whether retention rate is improving or worsening cohort by cohort, which is the question that actually matters. A single average churn rate figure blends good and bad cohorts into a number that describes nobody.
This is why cohort table reading sits next to attribution and segmentation as a skill rather than a metric. The table doesn't calculate anything new. It just refuses to let you average away the truth. A business can have a stable overall churn rate while every new cohort quietly performs worse than the one before it, and a lifetime value forecast built on the average will be wrong in exactly the direction that hurts.
Once you can read the shape, you can diagnose. A cliff in month one points at onboarding. A slow bleed across every column points at product fit or pricing.
A single churn rate number is a rumour. A cohort table is the evidence.
How it shows up
Cohort table reading shows up whenever someone asks 'is retention getting better or worse?' and the honest answer requires looking at newer cohorts against older ones rather than one blended figure. It shows up in board decks that quietly swap a churn rate trendline for a cohort heatmap once someone senior starts asking harder questions. It also shows up in customer success reviews, where a flattening curve after a specific month points straight at an onboarding or usage problem worth fixing.
The Australian context
Australian subscription and SaaS businesses often run smaller cohort sizes than US equivalents, purely because of market size. Smaller cohorts mean noisier percentages, a single lost customer can swing a monthly cohort by several points. Read cohort tables for shape and direction here, not for precision, and be cautious drawing conclusions from cohorts under fifty customers.
Where people get this wrong
Related terms
Common questions
What's the difference between a cohort table and a churn rate figure?
Churn rate is a single blended number for a period. A cohort table breaks that same population into groups by acquisition date so you can see whether retention rate is improving or worsening cohort by cohort, something the blended figure hides.
How many customers do I need before a cohort table is meaningful?
There's no fixed threshold, but cohorts under fifty customers tend to swing widely from single cancellations. Widen your cohort window to quarterly instead of monthly if your customer volume is low.
Should I read a cohort table by row or by column?
Both, but the column read matters more. Rows show expected decay within one cohort over time. Columns show whether that decay is getting steeper or flatter as newer cohorts replace older ones, which is the trend that needs action.
What does a cliff in month one usually mean?
A sharp drop right after acquisition almost always points at onboarding, not product fit. Customers who never reach a first meaningful use of the product tend to cancel fast. Look at activation steps before looking at pricing or features.
Debrief
Get the next one
No spam. No fluff. Just the next article, straight to your inbox.
Keep exploring
About New Rebellion
New Rebellion is a marketing intelligence consultancy. We build tools, score Australian businesses on how their marketing actually performs, and publish Debrief every day. This dictionary is part of how we work in the open.
How we think →