Topic
Funnels
How to build funnels that measure what users actually do: naming steps by intent, keeping them few, reading drop-off, and checking it against real sessions.
A funnel is an ordered list of steps and a count of how many people made it from each step to the next. That simplicity is why funnels are the most commonly built report in product analytics, and also why so many of them mislead: a funnel faithfully measures whatever steps you gave it, including steps that exist only in your team's vocabulary.
These articles are about building funnels that hold up. Name each step after what the user is trying to do, not the screen where it happens. Keep the count small enough that a drop between two steps points at one thing. Read the drop-off before the overall rate, and when a step leaks, watch the sessions behind it before you redesign anything.
Reading order
Where to start
- Build funnels from what users actually do
The core method: name steps by user intent, validate each one by watching real sessions, and keep one activation funnel that nobody edits casually.
- What an onboarding funnel is, and how to define its steps
A worked case. Onboarding is the funnel teams most often get wrong, because they mix up conversion and activation and add too many steps.
- Activation funnel benchmarks are mostly useless. Build your own.
Before you compare your conversion rates with someone else's: why benchmark numbers rarely transfer, and why drop-off and time between steps tell you more.
- When product teams need more than Google Analytics
Where funnels fit in a wider stack. Traffic tools tell you who arrived; this covers what product teams add when they need to see what those visitors did next.
Funnels
Every article on this topic
What an onboarding funnel is, and how to define its steps
A signup is a conversion, not an activation. The difference decides which steps belong in the funnel and which number you should actually watch.
Activation funnel benchmarks are mostly useless. Build your own.
Why published SaaS activation rates rarely transfer, and how to define an activation event and baseline that actually guides decisions.
When product teams need more than Google Analytics
How to evaluate a Google Analytics alternative when you need behavior analytics, session recordings, heatmaps, and feedback tied to product decisions.
Build funnels from what users actually do
Funnels work better when the steps come from observed user intent, not the internal names your team gives product screens.
Related