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

Funnel analysis for SaaS teams.

Funnel analysis should show where people drop out between first touch and value. If it only shows conversion rates with no context, it is not enough.

This page is for teams trying to answer:

✓ Where drop-off starts ✓ Which step is broken ✓ Which change is worth testing
See the activation deep dive See the next step

Plain English first. Conversion path second.

Funnel Analysis, Broken Down

01 — Entry Where the funnel starts and who actually enters it
02 — Friction Which steps slow people down or push them out
03 — Breakpoints Where the loss is concentrated and what changed
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.

FUNNEL ACCURACY IN SELF-BUILT SETUPS Often wrong

Impossible conversion rates (>100% at a step) appear in roughly 40% of self-implemented B2B SaaS funnels — because the event design didn't account for multi-device, multi-session, or re-entry behaviour.

TYPICAL FUNNEL DIAGNOSTIC VALUE 3–5 drop points

A well-instrumented activation funnel typically reveals 3–5 meaningful drop-off points. Most teams can only see 1–2 because critical steps are not tracked.

TIME FROM DIAGNOSIS TO ACTION 1 sprint

When funnel analysis produces a clean drop-off point, the team can design and ship a fix in one sprint — without it, the debate about "where to focus" continues indefinitely.

WHY FUNNEL ANALYSIS MISLEADS

"Our funnel shows 140% conversion on one step and nobody can explain it"

"Step 3 of our activation funnel has a 140% completion rate. That's been there for eight months. Engineering says it's technically possible if users go back and repeat the step. Product uses the funnel for weekly reporting anyway. Nobody trusts it but nobody has fixed it."

Head of Product — B2B SaaS, Series A
"The funnel shows where users drop out but not why"

"We have a funnel. We can see that 60% of users who start setup don't finish. But the funnel doesn't tell us whether they left because the step was confusing, because they didn't have the information they needed, or because they just got distracted. We're guessing at the fix."

Growth PM — PLG SaaS, $14M ARR
"We're optimising a funnel built around UI steps, not value moments"

"Our activation funnel tracks form completions, wizard steps, and button clicks. But nobody defined what value the user is supposed to feel at the end of the flow. We've optimised the mechanics of the setup without improving whether the product actually delivers something useful."

VP Product — B2B SaaS, $20M ARR
"Different tools show different numbers for the same funnel"

"PostHog says our activation is 68%. Amplitude says 74%. Our CRM says 61% of signups reach 'active' status. We have three funnel numbers and none of them agree. Every product review starts with a debate about which number to use."

Director of Analytics — SaaS, $35M ARR

What It Is

Funnel analysis is not just a conversion chart.

Funnel analysis is the practice of measuring where people move, pause, and drop out between the start of a journey and the moment value appears. The point is not to count more steps. The point is to make better decisions with less guessing.

A useful funnel analysis setup helps your team answer a small set of questions clearly. Where do people drop? Which step causes the biggest loss? Is the problem traffic, onboarding, setup, or value delivery? What changed after the launch?

When the setup is working, funnel analysis gives product, growth, and leadership the same view of where the loss is coming from. When it is not working, the team gets stage arguments, vague conversion numbers, and no clear fix.

Where Teams Get It Wrong

Most setups answer activity questions, not funnel 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 top-line conversion, not the broken step.

Plenty of setups log overall signups and completions. Much fewer are built around the specific step where users stall or leave.

Dashboards exist, but nobody changes the journey 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 remove next.

Step definitions are inconsistent across teams.

If everyone defines the steps differently, the funnel becomes a reporting argument instead of a useful diagnostic.

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

Funnel 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 the funnel stages.

Entry, step completion, and value moment definitions are written in plain language. Product, growth, and leadership are not using different meanings for the same stage.

02 — Trusted Instrumentation

The underlying journey layer is stable enough to trust.

Events, properties, and step order stay consistent. New instrumentation makes the funnel sharper instead of noisier.

03 — Decision-Ready Views

The dashboards point to a next action.

The team can look at a step view and know whether to investigate onboarding, routing, value delivery, or form friction next.

How ProductQuant Approaches It

Start with the step, not the dashboard.

Most funnel debt starts because tracking was added step by step, not journey by journey.

ProductQuant approaches funnel analysis from the business questions backward. First define the journey the team needs to understand. Then map the steps that answer those questions. Then build the views and QA process that keep the setup usable as the product changes.

That means step naming, dashboards, and tooling all serve the same goal: fewer arguments, clearer priorities, and better decisions.

01 — Define

Start with the journey question

Signup, trial, demo, onboarding, or activation. Name what the team actually needs to understand.

→
02 — Map

Design the step layer

Choose the events and properties that answer the question without turning the journey into clutter.

→
03 — View

Build the right analysis layer

Funnels, step views, dashboards, or segment views 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 journey evolves.

A cleaner setup means each new journey is easier to evaluate than the last one.

Related Guides And Proof

Go deeper from here.

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

Guides
Activation vs. Onboarding
A useful boundary when the funnel is stalling but the diagnosis is fuzzy.
Case Study Start here →
What Is Activation Rate?
A simple read on the metric that often sits behind funnel confusion.
Case Study Read →
SaaS Activation Rate Benchmarks
A benchmark read when the team needs context for the funnel gap.
Guide Compare →
Proof
PPC Platform Activation Strategy
A conversion rebuild showing how to move users from sign-up to value.
Case Study See →
PPC Platform Onboarding
A routed onboarding system that matched users to the right path.
Case Study See →
Onboarding Diagnostic
A funnel diagnosis that located where users were leaking value.
Guide Use →

CLIENT WORK

Healthcare SaaS — Funnel Design
114
events designed with clean funnel tracking

Activation Funnel From Scratch: 114 Events, Zero Ambiguity

Designed a full activation funnel for a healthcare SaaS — 114 custom events with JTBD-aligned naming, per-step conversion tracking, and a funnel definition the whole team could agree on and act from.

Read the case study →
B2B SaaS — Funnel Accuracy
3–5
actionable drop-off points surfaced per funnel

Funnel Diagnosis: From Broken Numbers to Actionable Drop-offs

Rebuilt a self-implemented funnel that was showing impossible conversion rates. Identified and fixed the event design gaps, then surfaced 4 distinct drop-off points the team could prioritise in a single sprint.

See the deep dive →

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.

Activation Deep Dive
Best when the funnel is stalled and the team needs a direct diagnosis.
Onboarding Review
Best when the issue looks like onboarding but needs a cleaner read.
Free Activation Teardown
A lighter first step if you want a quick read on where the funnel leaks.
Onboarding Teardown Kit
For teams that want a more complete teardown system for the journey.
Jake McMahon — funnel analysis consultant

WHO DOES THIS WORK

Jake McMahon

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

Jake has designed and rebuilt funnel analysis systems for B2B SaaS teams where the existing setup was technically running but analytically broken. The approach starts from the business question — what exactly is the team trying to measure — and works backward through event design, funnel definition, and interpretation to produce a number the team can actually trust and act on.

Funnel analysis Activation funnel Conversion rate optimisation Event design Funnel accuracy PostHog funnels Onboarding analytics B2B SaaS

COMMON QUESTIONS

Funnel analysis: what it is and what it should produce

Questions about your specific situation? Book a call →

What is funnel analysis?+
Funnel analysis is the practice of tracking completion rate at each step of a user journey. It identifies where users drop out between the start of a flow and a conversion goal. It is most useful when tied to a specific conversion goal — activation, upgrade, or checkout — rather than used as a generic engagement chart.
What is a conversion funnel in B2B SaaS?+
A conversion funnel in B2B SaaS is the sequence from signup to activation to paid — but it also applies to any specific in-product workflow. Each step is a potential drop-off point with a specific fix. The funnel is only useful when each step is clearly defined, consistently tracked, and tied to a decision the team can act on.
How do you build a funnel analysis in PostHog?+
Define each step as a specific event. Add filters for the relevant time window and user segment. PostHog shows per-step conversion rates and median time between steps. The critical design decision is whether the events are defined correctly — if the underlying event taxonomy is inconsistent, the funnel will produce numbers the team cannot trust.
What drop-off rate is considered normal in a SaaS onboarding funnel?+
20–40% drop-off per step is common in most B2B SaaS onboarding funnels. Anything above 50% at a single step is a clear problem worth diagnosing. The activation milestone itself should convert at 25%+ of total signups for a healthy setup.
How do you fix a high drop-off rate in a funnel?+
First confirm the event tracking is correct — false drop-offs from tracking gaps are common and are often mistaken for real product problems. Once the data is clean: reduce required steps, add progress indicators, and trigger contextual guidance at the specific drop-off point. Fix the measurement before you fix the product.
What is the difference between funnel analysis and cohort analysis?+
Funnel analysis measures conversion through a sequence at a point in time — it shows what percentage of users complete each step. Cohort analysis measures the behaviour of a group over time — it shows how retention or usage changes for users who started in the same period. Both are useful and the most complete picture comes from using them together.

Funnel analysis should tell you where value leaks.

If your team has conversion data but still cannot tell where the leak is, start with the activation deep dive or the review.

See the activation deep dive See the onboarding review