Home/Blog/Conversion Funnels/How to Fix Funnel Drop Off: Friction, Do…
Conversion Funnels

How to Fix Funnel Drop Off: Friction, Doubt or Mismatch

How to fix funnel drop off by naming the cause first: friction, doubt and mismatch need different fixes, and talking to people beside the numbers shows which.

Cover of the FIKR Space guide: How to Fix Funnel Drop Off: Friction, Doubt or Mismatch

To fix funnel drop off, first decide whether the cause is friction, doubt or mismatch, because each one has a different fix and the wrong fix does nothing. Then look at real people going through the failing step, since the numbers show where people leave and almost never why.

Finding the step that loses the most people is only half the work. A founder who stops there tends to reach for the nearest remedy, which is usually to simplify something. Sometimes that is right. Often it is not. This post is about the choice between three causes, how to tell them apart, and what to do about each. If you have not yet located the step, start with how to find where a funnel loses people.

How to fix funnel drop off: name the cause before the fix

People leave a step for one of three reasons, and the reasons call for different action. Treating them as one problem is the most common way to spend a month improving the wrong thing.

CauseWhat it meansHow it is fixed
FrictionIt is too hard, too long, too slow, has too many fields, or it is unclear what to do nextRemove steps and reduce effort
DoubtThey understand and are not yet convinced: missing evidence, pricing anxiety, an unanswered riskAdd the right information at exactly the point of hesitation
MismatchThey were never right for this, arrived with a wrong expectation, or the previous step promised something this one does not deliverFix it upstream, in the message or the targeting

Read the third row twice. Mismatch is the only cause where working on the step itself is guaranteed to fail. You can polish a page for as long as you like, and if the problem was created before people reached it, the page will not repair it.

Why founders assume friction

Founders default to friction because it is the most actionable. You can shorten a form this afternoon. You can remove a screen, tidy a layout, cut a field. It feels like progress and it is easy to measure afterwards.

The cost is that it gets applied to problems that are not friction. A page can be made simpler and still lose people, because the real problem was that nobody believed the claim. Simplifying does not supply evidence. Likewise, no amount of simplification at a step will fix a mismatch that was set up by an earlier promise. In both cases the effort is real, the result is nil, and the next conclusion drawn is wrong, since the team decides the step is hopeless when it was only mis-diagnosed.

The habit to build is a pause. Before you touch the step, write down which cause you believe it is, and why. If you cannot say why, you are guessing, and the next section tells you how to stop.

Free guide, 30 pages: The Funnel Playbook. Send me the guide.

Look at people, not only percentages

The funnel tells you where. It very rarely tells you why. The quantitative view finds the location and sizes the prize. The qualitative view supplies the cause.

There are two simple ways to get it. You can watch a handful of people go through the failing step. Or you can ask a few of the people who abandoned it. Either approach generates better hypotheses in an hour than a week of staring at percentages.

EvidenceWhat it gives youWhat it cannot do alone
NumbersThe location and the size of the lossSay why people leave
Watching and askingThe cause, in people's own behaviour and wordsTell you how big the prize is, or whether it generalises

Teams that use only one of the two go wrong in opposite directions. Those with only numbers optimise blindly. Those with only conversations chase vivid anecdotes that do not generalise. Using the numbers to choose where to look, and the conversations to understand what you see, avoids both.

Reading the evidence for each cause

When you watch someone use the step, or hear from someone who left it, you are listening for a particular kind of signal. The three causes tend to sound different.

  • Friction sounds like effort. The person is confused about what to do next, finds the step long, or struggles with the work it asks of them. They are willing, and the step is in the way.
  • Doubt sounds like a question. The person understands the offer and hesitates. They are looking for evidence, worrying about the price or wondering about a risk that has not been answered.
  • Mismatch sounds like surprise. The person expected something else. What they met did not fit what they were promised, or they were never the right person to begin with.

These are prompts for your listening, not a script. What matters is that you go in with the three possibilities in mind, so that what you hear can be sorted, rather than merely collected.

Fixing friction

When the cause is friction, the remedy is to take things away. Remove steps. Reduce the effort the person must make. Cut what is too long or too slow, and make what to do next obvious.

Because this is the most actionable fix, it can usually be done fast. The discipline is to do it only where the evidence points to friction, and to make one change at a time so that you can see its effect. How to run changes without tangling them is covered in how to run a funnel improvement loop.

Fixing doubt

When the cause is doubt, the person is not stuck. They are unconvinced. The remedy is to add information, and the placement is the whole trick: it has to appear at exactly the point of hesitation. Evidence that sits on a page the person never reads at that moment does not help them with a question they have right now.

Look for the questions that go unanswered at the failing step: missing evidence, anxiety about the price, a risk nobody has addressed. Answer them where they arise. Notice that this fix often makes a step longer, not shorter, which is why it is the opposite of a friction fix and why the two must not be confused.

FIKR Content, multi-channel publishing: reads, listens, opens, clicks in one dashboard. See FIKR Content.

Fixing mismatch

When the cause is mismatch, the problem began before the step. Either the people arriving are not the right people, or the earlier message set an expectation that the step does not meet. The fix lives upstream, in the message or the targeting.

This is the finding that is hardest to see from inside any one team. Marketing sees conversions, and product sees users who did not engage, and only the funnel shows they are the same people. A sharp drop straight after someone commits, after a signup or a first purchase, is a typical sign. It is usually not a usability problem.

Segments can help here. If the loss is concentrated in one audience, for example one traffic source, that tells you the people sent are the issue and not the step. That is covered in how to segment a funnel.

A signal that you fixed the wrong thing

Sometimes a fix works at its own step and the loss simply moves to the next one. The people you saved were never going to convert anyway. The guide says this usually indicates a mismatch that the fix did not address.

That is useful information, not a failure. It tells you to stop working at the step and to look upstream. It is also a good reason to check the improved step again some weeks after the change, to see whether the gain held.

When the numbers and the people disagree

Occasionally what you see will not match what you expected. A step that the numbers say is losing people may look smooth when you watch someone use it, and nobody you ask seems to remember a problem. Do not argue with either source. Treat the disagreement as a prompt to check the basics: whether the step counts cohorts, whether your own walkthrough appears correctly, and whether the loss belongs to one audience and not the step. A drop that turns out to be an instrumentation gap is a cheap finding, and far better made now than after a redesign.

Free guide, 30 pages: The Funnel Playbook. Send me the guide.

A short routine for any failing step

  1. Confirm the step really is losing people, using cohort counts.
  2. Watch a few people go through it, or ask a few who left.
  3. Name the cause: friction, doubt or mismatch.
  4. Write down your hypothesis before you change anything.
  5. If it is mismatch, work upstream. Otherwise change the step, one thing at a time.
FIKR Content, multi-channel publishing: reads, listens, opens, clicks in one dashboard. See FIKR Content.

Frequently asked questions

How do I know which cause I have?

Watch a few people use the step or ask a few who left, then sort what you hear. Effort points to friction, an unanswered question points to doubt, and surprise or a wrong audience points to mismatch.

Is simplifying the page ever the wrong move?

Yes. If people understand and simply do not believe the claim, simplifying does nothing. That is doubt, and it needs information at the point of hesitation.

Can I fix a mismatch at the step itself?

No. It is the one cause where working on the step is guaranteed to fail. It has to be fixed upstream, in the message or the targeting.

Do I need numbers if I am talking to people?

Yes. Conversations alone can lead you to chase vivid anecdotes that do not generalise. Numbers show where to look and how large the prize is.

The full method, with the checklists, is in The Funnel Playbook, a free 30-page playbook.

LG

Luis Goncalves

// Founder & CEO at FIKR Space

Three-time founder. Built and exited Evolution4All before this. Now building FIKR Space — the operating infrastructure underneath every innovation ecosystem (startups, accelerators, governments, investors). Lisbon-based, works global.

Claim your founding member spot.

Every founding member is hand-reviewed. Tell us who you are and how you want to co-create — we get back inside 48 hours.

Role
Job Title
Country
Cookies

We use cookies.

Essential ones keep the site working. The rest only run if you say so. No dark patterns, no pre-ticked boxes.Privacy policy

Co-create with us. It's about building it together.Co-Create With Us