How to optimise conversion funnels

Detect where you lose users and design actions to improve every funnel stage

9 min

A conversion funnel is the sequence of steps a user takes from arrival to final conversion. At each step, a percentage of users drops off. Funnel optimisation involves identifying the steps with the highest drop-off and reducing the friction that causes it.

This guide explains how to analyse funnels, detect drop-off points, use micro-conversions as progress indicators and design specific tests for each funnel stage.

Defining your funnel steps

The first step is mapping your actual business funnel. In ecommerce: visit → product page → add to cart → checkout → purchase. In SaaS: visit → sign-up → activation → paid conversion. In lead generation: visit → service page → form → submission.

Each step must correspond to a measurable event in your analytics tool. If you cannot measure a step, you cannot optimise it. Configure events in GA4 or your analytics platform before attempting to analyse the funnel.

Drop-off analysis: where users are lost

Drop-off analysis identifies the steps with the highest abandonment rate. In GA4 funnel explorations, you can see the percentage of users advancing from one step to the next and the percentage that abandons.

Knowing where they abandon is not enough: you need to understand why. A high drop-off on the checkout page could be caused by unexpected shipping costs, missing payment methods, an overly long process or technical issues. Combine quantitative analysis (GA4) with qualitative analysis (heatmaps, recordings, surveys).

  • Identify the step with the highest absolute abandonment percentage
  • Calculate the conversion rate between each pair of consecutive steps
  • Segment by device, channel and audience to find patterns
  • Cross-reference with qualitative data to understand abandonment reasons

Identifying and removing friction points

Friction is any obstacle that makes it harder for the user to advance to the next step: an overly long form, a price hidden until the last moment, mandatory registration, a slow page or confusing design.

The most common friction points vary by business type, but there are universal patterns: unexpected costs are the #1 cause of cart abandonment (47%), followed by mandatory account creation (25%) and an overly long process (18%).

  • Unexpected costs: show shipping costs as early as possible
  • Mandatory registration: offer guest checkout
  • Long process: reduce steps, combine screens, eliminate unnecessary fields
  • Lack of trust: add security badges, return policy and reviews
  • Technical issues: validation errors, slow loading, broken mobile design

Micro-conversions as progress indicators

Micro-conversions are intermediate actions that signal progress towards the final conversion: viewing the pricing page, adding to cart, starting checkout, completing step 1 of a form. Measuring them allows you to understand funnel health with greater granularity.

If you optimise a micro-conversion (for example, increasing the percentage of users who add to cart), the effect propagates downstream and increases the final conversion. Micro-conversions are easier to optimise because they depend on fewer variables than the final conversion.

Specific tests for each stage

Each funnel stage has its own optimisation levers and test types. At the top (awareness/attraction), test headlines, value propositions and CTAs. In the middle (consideration), test sales arguments, social proof and information structure. At the bottom (decision/conversion), test the checkout process, forms and trust elements.

Testing priority should go to the step with the highest potential impact, which is usually the one with the greatest drop-off and highest volume. A small percentage improvement on a step that 10,000 users pass through has more impact than a large improvement on a step with 500 users.

  • TOFU: headlines, hero images, value proposition, primary CTAs
  • MOFU: sales arguments, social proof, comparisons, FAQs
  • BOFU: forms, checkout, payment methods, trust messages

Segmented funnels: not all users are equal

The aggregate funnel hides important differences between segments. A Google Ads user on mobile has a radically different funnel from an organic desktop visitor. Analysing funnels by segment reveals where the real opportunities lie.

The most useful segments for funnel analysis are: device (mobile vs desktop), traffic source (organic, paid, referral, direct), user type (new vs returning) and audience (GA4 or CDP segments). The biggest opportunities typically lie in segments with the worst relative performance but the highest volume.

Key Takeaways

  • Define each funnel step as a measurable event in analytics
  • The step with the highest drop-off and volume is the optimisation priority
  • Combine quantitative analysis (GA4) with qualitative (heatmaps, recordings)
  • Micro-conversions allow you to optimise with greater granularity
  • Analyse funnels by segment: the aggregate hides opportunities

Where are you losing users in your funnel?

We analyse your conversion funnel, identify the highest drop-off points and design optimisation actions with measurable impact.