A funnel is a sequence of steps to an outcome, and funnel analysis pinpoints where users drop off and what each drop costs. Every product has several funnels — not one — each governing a different journey.
▸ Try the interactive toolA funnel is a sequence of steps users must complete to reach a desired outcome, and funnel analysis identifies exactly where users drop off and quantifies the business cost of each drop-off point. Every product has not one funnel but several, each governing a different critical journey — and treating them as one is a common mistake.
The major funnel types include acquisition (visitor → sign-up), activation/onboarding (sign-up → first value), conversion (free → paid), and feature-adoption (aware → habitual user). Each has different steps, different drop-off causes, and different business stakes. The skill is mapping the right funnel for the question you're asking, quantifying the drop-off at each step (in users and in business value), and recognising that the biggest drop isn't always the most valuable one to fix.
Acquisition — visitor → sign-up
Activation — sign-up → first value
Conversion — free → paid
Feature adoption — aware → habitual
Each has distinct steps, drop-off causes, and stakes.
Treating a product as having a single funnel blurs distinct journeys with distinct problems — and quantifying drops only in user counts misses where the real business value leaks.
| The shortcut | What it costs | What it gives you instead |
|---|---|---|
| Treating it as one funnel | Distinct journeys with distinct problems get blurred together. | Mapping each funnel type isolates the real drop-off and cause. |
| Counting drops in users only | The biggest user-drop isn't always the costliest. | Quantifying drops in business value prioritises correctly. |
| No funnel for the question | Analysing the wrong journey for the problem at hand. | Choosing the right funnel type targets the actual issue. |
| Ignoring downstream funnels | Fixing acquisition while conversion silently leaks. | Seeing all funnels reveals where value actually leaks. |
Acquisition, activation, conversion, or feature adoption? The journey you analyse must match the problem you're solving — don't analyse onboarding for a monetisation question.
Lay out the specific steps users move through to the outcome. Clear step definitions are what make drop-off measurable.
Measure the percentage lost at each step and the business value of that loss. A step losing few users but high-value ones can matter more than a high-volume, low-value drop.
The most valuable fix is the drop that costs the most business value relative to the effort to fix it — not automatically the largest percentage drop. Prioritise by value, not volume.
Once you've found the drop worth fixing, the drop-off diagnostic (Tool 12) is how you move from 'users drop here' to a prioritised experiment to fix it.
A team mapped its conversion funnel and saw the biggest percentage drop at an early step. They instinctively prioritised fixing it — it was the largest number, after all. But quantifying the drops in business value rather than raw users changed the picture entirely.
A later step lost fewer users, but those users were far higher-value — they'd shown strong intent, and losing them cost much more per head. The biggest volume drop was low-value early-funnel tyre-kickers; the most valuable drop was the smaller, later one. Fixing the high-value drop returned far more than chasing the big early percentage would have.
The deliverable is the right funnel for the question, with each step's drop quantified in both users and business value — prioritised by value.
| Funnel | Journey | Stakes |
|---|---|---|
| Acquisition | Visitor → sign-up | Top-of-funnel volume |
| Activation | Sign-up → first value | Retention foundation |
| Conversion | Free → paid | Direct revenue |
| Feature adoption | Aware → habitual | Depth & expansion |
Every product has multiple funnels, and the analytical skill is twofold: choosing the right funnel for the question, and quantifying its drops in business value rather than mere user counts — because the biggest drop and the most valuable drop are often different.
The value-weighting insight is the one that changes prioritisation. It's natural to attack the largest percentage drop — it's the most visible — but a large drop of low-value users (early-funnel browsers who were never going to convert) can matter far less than a small drop of high-intent, high-value users deeper in the journey. Quantifying each drop in business value, not just volume, is what surfaces the fix worth making. And recognising that a product has several distinct funnels — acquisition, activation, conversion, adoption — prevents the error of optimising one while another silently leaks. Funnel analysis identifies where and how much; the drop-off diagnostic then handles the why and the fix.
Distinct journeys blur together. Map each funnel type separately.
The largest drop isn't always the costliest. Quantify in business value.
Match the funnel type to the question — don't debug onboarding for a revenue issue.
Funnels show where users drop; use the diagnostic (Tool 12) to find why and fix it.
Funnels find the drop worth fixing; the diagnostic (Tool 12) explains and fixes it.
The activation funnel gets its own deep treatment (Tool 13) as the highest-leverage one.
Quantifying drops in business value ties funnel steps to outcomes (Tool 02).
The free→paid funnel connects directly to SaaS revenue and economics (Tools 14, 15).
Imagine a conversion funnel where the biggest percentage drop is at the first step and a smaller drop is at the last. Why might the smaller, later drop be the more valuable one to fix?
Sketch how you'd quantify each drop in business value, not just user count, to decide.
If the smaller later drop — high-intent users lost near the finish — turns out more valuable than the big early drop of browsers, you've learned why funnels must be measured in value, not volume.