Jake McMahon
Led by Jake McMahon 8+ years B2B SaaS · Behavioural Psychology & Big Data

Retention analysis for SaaS teams.

Retention analysis should show why customers stay, drift, or leave so the team can act before churn is final.

This page is for teams trying to answer:

✓ Which behaviors predict retention ✓ Where churn risk rises ✓ Which intervention to try
See the churn analysis program See the next step

Plain English first. Intervention and prevention second.

Retention Analysis, Broken Down

01 — Signals Which usage, support, and billing signals matter most
02 — Cohorts Which groups are retaining, slipping, or leaving
03 — Views Churn, cohort, and risk views tied to decisions
04 — Action What the team changes next because the signal is clear

If those four layers do not connect, the setup stays descriptive instead of useful.

RETENTION SIGNAL LEAD TIME 30 days

The usage patterns that predict 90-day retention are visible within the first 30 days of an account's lifecycle — if the instrumentation is designed to surface them.

MOST USEFUL COHORT SPLIT By activation milestone

Accounts that reach the activation milestone retain at 2–4x the rate of those that don't. That gap is where the intervention should be designed.

RETENTION CURVE SHAPE Tells you everything

Whether your retention curve flattens or keeps declining after month 3 is the single most diagnostic signal in a B2B SaaS analytics stack.

WHY RETENTION ANALYSIS STALLS

"Our retention chart shows a number but we can't explain the shape"

"We track 90-day retention by cohort. It's around 68%. But when leadership asks 'which cohorts are better and why,' we don't have a clean answer. The number exists. The explanation doesn't."

Head of Product — B2B SaaS, $25M ARR
"We know some accounts retain better but don't know what they do differently"

"Enterprise accounts retain at 84%. SMB accounts retain at 51%. We know the gap but we don't know what behaviour separates the ones that stay from the ones that leave. Until we know that, we can't change it."

VP Product — PLG SaaS, Series B
"We're measuring retention but it's the wrong metric for our product"

"We're a B2B tool used once a month for a specific workflow. Our 'weekly active user' retention looks terrible. But our actual renewal rate is 91%. We need retention metrics that match how our product is actually used — not generic DAU/WAU curves."

Founder/CEO — Vertical SaaS, $8M ARR
"Retention improved for one cohort but we don't know what we changed"

"Q2 retention was up 7 points over Q1. We shipped four things that quarter. Two of them were product changes, one was a pricing change, one was an onboarding sequence update. We have no idea which one drove the improvement — or whether it was a seasonal effect."

Growth Lead — B2B SaaS, $40M ARR

What It Is

Retention analysis is not just churn reports.

Retention analysis is the practice of measuring whether customers come back, stay active, and expand over time. The point is not to count more cancellations. The point is to make better decisions with less guessing.

A useful retention analysis setup helps your team answer a small set of questions clearly. Which cohorts are leaving? Which signals predict churn? Which interventions actually move the number? Which segments are most at risk right now?

When the setup is working, retention analysis gives product, success, support, and leadership the same view of where loss is coming from. When it is not working, the team gets averages, cancellation logs, and no clear next move.

Where Teams Get It Wrong

Most setups answer activity questions, not retention questions.

The tools are usually there. The gap is between what the team tracks and what the team actually needs to know.

The team tracks cancellations, not warning signals.

Plenty of setups log the final cancellation event. Much fewer are built around the earlier signals that show churn is building.

Dashboards exist, but nobody changes the retention plan because of them.

That usually means the views are descriptive but not decision-ready. The team can observe movement, but not what to fix, test, or save next.

The churn model is not connected to action.

If the model does not point to an intervention, it is just a forecast. Retention analysis needs a next step.

The setup explains the past, but not the next intervention.

Retention analysis is most valuable when it shortens the time between "something changed" and "the team knows what to do next."

What Good Looks Like

Three signs the setup is actually useful.

01 — Clear Definitions

The team agrees on retained, at-risk, and lost states.

Active use, renewal status, churn event, and expansion signals are defined in plain language. Product, success, and leadership are not using different meanings for the same metric.

02 — Trusted Instrumentation

The underlying signal layer is stable enough to trust.

Usage, support, billing, and success signals stay consistent. New instrumentation makes the system sharper instead of noisier.

03 — Decision-Ready Views

The dashboards point to a next action.

The team can look at a retention, risk, or segment view and know whether to intervene, extend, or escalate next.

How ProductQuant Approaches It

Start with the question, not the spreadsheet.

Most retention debt starts because tracking was added signal by signal, not question by question.

ProductQuant approaches retention analysis from the business questions backward. First define what the team needs to know. Then map usage, support, billing, and success signals that answer those questions. Then build the views and intervention process that keep the setup usable as the customer base changes.

That means definitions, dashboards, and review cadence all serve the same goal: fewer arguments, clearer priorities, and better decisions.

01 — Define

Start with the retention question

Who is leaving, who is at risk, and which intervention is worth trying. Name what the team actually needs to understand.

→
02 — Map

Design the signal layer

Choose the behaviors and properties that answer the question without turning the system into clutter.

→
03 — View

Build the right analysis layer

Retention views, cohorts, dashboards, or risk segments should point to a concrete next action, not a reporting ritual.

→
04 — Run

Keep it usable over time

Ownership, QA, naming discipline, and decision reviews stop the setup from drifting as the customer base evolves.

A cleaner setup means each new risk pattern is easier to answer than the last one.

Related Guides And Proof

Go deeper from here.

These are the most relevant ProductQuant assets if you want implementation detail, churn context, or a clearer retention foundation.

Guides
The 6 Types of SaaS Churn
A practical read when you need to separate churn types before acting.
Guide Start here →
Retention Moats
A useful read on why some products keep customers longer than others.
Guide Read →
Why Customers Churn in B2B SaaS
A useful read when you need the cleanest explanation of retention loss.
Guide Compare →
Proof
Churn Prevention Case Study
A retention rebuild showing how to save accounts at the moment of intent.
Case Study See →
Churn Early Warning Dashboard
A useful example of turning warning signals into something the team can act on.
Case Study See →
Reduce Churn with PostHog
An engineering-led retention playbook with a measurement angle.
Guide Use →

CLIENT WORK

Healthcare SaaS — Signal Design
14 days
earlier churn detection with behavioural signals

Retention Signals: Identifying At-Risk Accounts Early

Built a behavioural retention signal system for a healthcare SaaS — usage decline, session shortening, and feature abandonment patterns identified accounts at risk 14+ days before cancellation events.

Read the case study →
B2B SaaS — Cohort Analysis
13
dashboards covering activation, retention, and feature adoption

Retention Analysis From Scratch: Full Dashboard Build

Full retention analytics implementation for a healthcare SaaS — 13 dashboards covering activation milestone tracking, cohort retention curves, and feature adoption correlations with 37 UX issues surfaced from session replay data.

Read the case study →
Jake McMahon — retention analysis consultant

WHO DOES THIS WORK

Jake McMahon

Founder, ProductQuant · MSc Big Data & Business Analytics · BSc Behavioural Psychology · 8+ years B2B SaaS

Jake has built retention analysis systems for B2B SaaS teams at every stage from $5M to $80M ARR. The work starts from a core question: what does a retained account actually do differently in the first 30 days, and is the instrumentation layer designed to surface that difference?

Retention analysis is most useful when it isolates the product behaviours that predict retention at the account level — not just the aggregate curve. That means cohort analysis, activation milestone tracking, and feature adoption correlations built to answer specific questions, not to produce charts for the board deck.

Retention analysis Cohort analysis Activation analytics Feature adoption Retention curve diagnosis Early warning signals B2B SaaS PostHog

COMMON QUESTIONS

Retention analysis: what it is and what it should produce

Questions about your specific situation? Book a call →

What is retention analysis and what should it produce?+
Retention analysis measures which accounts stay active over time and why. Done well, it should produce: cohort retention curves that show how different signup groups behave over time, activation milestone correlation showing which early product actions predict long-term retention, and a clear brief on what retained accounts do differently in their first 30 days. It is not useful if it stops at a single aggregate retention percentage.
How do you measure retention for a B2B SaaS product?+
Cohort analysis by signup or activation date is the most reliable method. Track rolling 30/60/90 day active rates and measure at the account level rather than the user level for B2B products. The critical design decision is defining “active” based on a meaningful product action — not just login. Login-based retention hides real disengagement and overstates how well the product is actually being used.
What is a good retention rate for B2B SaaS?+
Benchmarks vary by ACV tier: enterprise ($50K+) should see 90%+ at 12 months; mid-market typically lands at 80–85%; SMB is often 70–75%. But these aggregate numbers are less useful than comparing within your own segments. The question that actually drives decisions is: which of your cohorts retains best, what did they do differently, and can you replicate that pattern?
What product behaviours predict long-term retention?+
The most consistent predictors across B2B SaaS products are: activation milestone completion, feature depth (3+ features used in the first week), collaboration or team actions (inviting colleagues, sharing outputs), and integration setup. These vary by product — a daily-use tool has very different predictors than a monthly reporting tool. The only way to know which behaviours matter for your product is to build the correlation analysis from your own event data.
How is retention analysis different from churn analysis?+
Retention analysis looks forward: which accounts are staying, what they do differently, and what product behaviours predict long-term engagement. Churn analysis is retrospective: which accounts left and what pattern did they fit. Both are needed, but retention analysis is more predictive — it points to interventions that can be designed before the cancellation decision is made. If you only run churn analysis, you are always explaining the past rather than changing the future.
How do you improve retention once you've identified the gap?+
The intervention depends on the gap type: activation failure requires onboarding redesign targeted at the milestone that predicts retention; value decay requires feature discovery nudges and re-engagement sequences; plan mismatch requires a better upgrade path at the right moment; product-fit failure requires segment repositioning or ICP refinement. A single generic retention program cannot address all four. The analysis must identify which archetype applies before the intervention is designed.

Best Next Step

Pick the step that matches the gap.

This page is educational first. If you want help turning the ideas into a working setup, these are the most relevant ProductQuant paths.

Churn Prediction
Best when the team needs early warning on account risk.
Churn Prevention Emails
Best when the team needs a direct intervention path after the risk is clear.
Kill Churn Workshop
For teams that need a focused session to diagnose and reduce churn.
Churn Diagnosis Playbook
For teams that want a repeatable playbook for identifying churn drivers.

Retention analysis should tell you who is leaving before they leave.

If your team has churn data but still cannot tell why it is happening, start with the program or the workshop.

See the program See the workshop