What most web shops measure wrongly in their funnel analysis
You think your funnel is clear. But what you measure is often only part of reality. And that is exactly what costs you conversion.
Funnels are often used as the way to analyse conversions. From product page to basket, to checkout, to purchase. But reality is more complex, and your data usually does not show that. In this blog we look at the most common mistakes in funnel analysis. And more importantly: how to avoid them. So you measure what really happens and optimise on behaviour, not on assumptions.
Contents
- The funnel as a model: why it is too simple
- Common measurement mistakes in GA4 funnels
- What you miss if you only measure pages
- How to make your funnel analysis actually valuable
- Conclusion
The funnel as a model: why it is too simple
A classic funnel suggests a logical order: view product, add to basket, check out, done. But in reality visitors click back and forth, compare, save, come back, drop off and start again. That makes the funnel model inadequate on its own.
Many web shops keep using this linear model in their analytics. They base their optimisation on ‘drop-offs’ between step 2 and 3, without realising that many users do not take the steps in that order, or skip them altogether.
The result? You optimise for a route that many customers never take. And that creates blind spots in your analysis and missed opportunities in your conversion.
Common measurement mistakes in GA4 funnels
GA4 offers extensive funnel reports. But if you set them up wrongly, you do not get insight, you get distortion. These are the most common mistakes:
- Only using pageviews as funnel steps (while behaviour within pages is what matters most)
- Measuring events without parameters, so context is missing
- Basing funnels on sessions instead of users (so returning visitors disappear)
- No breakdown per device, while mobile behaviour is completely different
If you only measure whether a page was viewed, you still do not know whether a visitor took action. Did they scroll, click, add something to the basket? Without that interaction data your funnel analysis stays superficial.
What you miss if you only measure pages
Visitors do not always navigate along the route you intended. They arrive on a PDP via Google, click back to a category, open a blog or leave the site only to return an hour later via email. If your funnel only measures pages, you will not see this behaviour.
That is why your funnel analysis needs to be set up at event level. You do not just want to know which page was visited, but what happened there: which buttons were used, which modules were viewed, was anything added to the basket?
If you really want to understand behaviour, combine data from GA4 with tools such as Clarity or Hotjar. There you can see whether a user gets stuck on your filter, hesitates at your shipping option or clicks away through a lack of trust.
How to make your funnel analysis actually valuable
If you want your funnel analysis to give direction, you need to measure at behaviour level, segment on relevance and filter on quality. That means:
- Work with events: add_to_cart, begin_checkout, form_submit, purchase
- Use parameters: which product, which device, which channel
- Compare converters and drop-offs on behaviour: where is the difference?
- Create segments: new vs. returning visitors, mobile vs. desktop
- Combine funnel analysis with session recordings for context
Only once you have insight into how users behave, not just where they land, can you optimise effectively. Funnel data then stops being a list of numbers and becomes a roadmap.
Want to know more about behaviour and conversion analysis? See how we design funnels based on behavioural data.
Conclusion
Funnel analysis is valuable, if you do it right. Measuring pages without behavioural context gives you false precision. By switching to event-based funnels, analysing behaviour and applying segmentation smartly, you get a grip on where you are really optimising.
Let us look at your funnel structure or book an analysis. So you do not just measure where customers drop off, but also why, and what you can do about it.
Melle Sikma
Helmsman
melle@brandmonks.nl
