On this article

Why Too Many Steps Break Your Conversion Funnel (and How to Fix It)

Long funnels can collapse to zero because every step compounds drop-off. Why it happens, and how a critical-path funnel fixes it.
This is some text inside of a div block.

Why Too Many Steps Break Your Conversion Funnel (and How to Fix It)

The short version: A conversion funnel shows too few — sometimes zero — conversions when it has too many steps, because a funnel only counts users who complete every step in order. Each extra step multiplies the drop-off, and any single low-traffic or mis-fired step can collapse the whole funnel toward zero even when people are genuinely converting. The fix is to measure the critical path: the handful of high-signal events that actually represent progress, not every micro-interaction. In one real case, a 13-step funnel reporting zero conversions was replaced with a 4-step funnel that immediately surfaced 56 conversions that had been there all along.

What is Conversion Funnel? - Learn How to Optimize your Conversions

The symptom: a funnel that says (almost) nothing

You open your funnel report and it says zero conversions. Or a number so low you know it can't be right — you have customers, so where are they? The instinct is to blame the tool: it must have a limit, or a bug. Almost always, it doesn't. The funnel is working exactly as designed; the problem is how many steps you asked it to check.

Why too many steps break a funnel

A funnel measures one specific thing: how many users completed a defined sequence of events, in order, within a time window. That "every step, in order" requirement is where long funnels get you, in two ways.

Drop-off compounds. Even gentle drop-off multiplies. Ten steps that each retain 90% of users leave you at 0.9¹⁰ ≈ 35% — and real steps aren't all 90%. Stack enough of them and the end of the funnel rounds toward nothing.

One weak step can sink everything. This is the real killer. If an intermediate step is a rarely-fired, low-traffic, or mis-sequenced event, the funnel still requires it — so users who genuinely converted but didn't happen to trigger that one event get dropped from the count. The funnel isn't reporting "how many people converted." It's reporting "how many people fired all of these exact events in this exact order." The more steps you add, the more chances a real converter has to fall out on a technicality.

If funnels are new to you, our primer on funnel analysis — what it is and how to use it covers the fundamentals.

A real example: 13 steps to zero, 4 steps to 56 conversions

This isn't hypothetical. A business-formation platform we worked with had Amplitude live and a 13-step formation funnel returning zero conversions. The team assumed Amplitude couldn't handle it. The actual cause was mathematical: low-traffic intermediate events were collapsing the funnel before it could ever resolve.

The fix took days, not weeks. Replacing the 13-step funnel with a 4-step critical-path funnel immediately revealed 56 conversions that had been invisible the whole time. Nothing about the product changed — only what the funnel was asked to measure. (The full story is in the Bizee case study.)

The fix: measure the critical path

The solution isn't a different tool — it's a shorter, sharper funnel. A critical-path funnel includes only the events that unambiguously mark real progress: roughly started → key milestone → succeeded. Each step should be a high-signal event that essentially every genuine converter fires. Everything in between is diagnostic detail, not part of your headline conversion metric.

Use the long, detailed funnel as a magnifying glass — to investigate where people drop once you know your real conversion rate — not as the number you report. The critical-path funnel tells you whether you're converting; the detailed one helps you find out why you're not.

So how many steps should a funnel have?

As few as it takes to capture the real journey — usually three to six. The test for including a step is simple: does nearly every real converter trigger this event? If the answer is no — it's optional, it's low-traffic, or it fires inconsistently — it doesn't belong in your primary funnel. Milestones, not micro-interactions.

Ecommerce Conversion Funnel: Definition, Stages & Optimization Tips | Yespo

Other reasons a funnel collapses

Too many steps is the most common cause, but a few cousins produce the same "zero conversions" symptom:

  • An event that isn't firing at all. If a required step was never implemented correctly, the funnel dead-ends there. Solid event tracking is the prevention.
  • Front-end/back-end sequencing. When a server-side event and a client-side event land out of the order your funnel expects, a strict-order funnel breaks even though the user did everything right.
  • Strict order vs. any order. Requiring "this exact order" when the real journey is flexible will understate conversions. Testing with "any order" often reveals the conversions immediately.
  • No tracking plan. Ad-hoc events with inconsistent names mean your funnel may be pointing at the wrong event entirely — one of several Amplitude configuration mistakes that quietly corrupt reporting.

How to fix it in Amplitude

  1. Build a short critical-path funnel — three to five milestone events, in the order a real customer hits them.
  2. Check the event volume at each step. A step showing near-zero events is your culprit; that's the one collapsing the funnel.
  3. Switch ordering to "any order" as a test. If conversions appear, your sequencing or order requirement was the problem.
  4. Confirm each event actually fires using a live event view before trusting the report.

Getting a trustworthy funnel is usually a foundation problem

A funnel that lies is almost always a symptom of a shaky measurement setup underneath — no tracking plan, mis-sequenced events, or a funnel built to mirror the UI instead of the real journey. If your reports don't match reality and you're not sure why, a Data Stack Audit pinpoints the root cause and gets your numbers trustworthy again.

👉Book a call with our team →

Frequently asked questions

Why is my Amplitude funnel showing zero conversions?
Almost always because the funnel has too many steps, and one of them is a low-traffic or mis-fired event that's required to complete the funnel. Since a funnel only counts users who complete every step in order, that one weak step drops real converters and collapses the total toward zero. Rebuilding it as a short critical-path funnel usually surfaces the conversions instantly.

How many steps should a conversion funnel have?
Usually three to six — only the milestones that nearly every real converter passes through. If a step is optional, low-traffic, or fires inconsistently, leave it out of your primary funnel and use a longer, detailed funnel separately for diagnostics.

Does adding more steps to a funnel lower the conversion rate?
It lowers the reported rate, yes. Each step compounds drop-off, and any step that doesn't fire reliably removes real converters from the count. More steps rarely tell you more; they mostly add ways for the number to understate reality.

What is a critical-path funnel?
A funnel built from only the essential events that mark genuine progress — start, key milestone, success — rather than every interaction along the way. It gives you an accurate read on whether users are converting, which a long, over-detailed funnel often can't.

Why does my funnel work with "any order" but not "this order"?
Because your events aren't firing in the strict sequence the funnel expects — often due to front-end and back-end events landing out of order. If "any order" shows conversions and "this order" shows zero, sequencing is your problem, not the conversion itself.

Related articles

Guide
5min

Running Experiments in Statsig: Layers, Holdouts & Power Analysis

How experimentation works in Statsig: reading results, power analysis and MDE, and using layers and holdouts to run many tests at once.
Guide
5min

Statsig Feature Management: Feature Gates, Dynamic Configs & Parameter Stores Explained

Feature gates, dynamic configs, and parameter stores let you turn features on, tweak configs, and update content, no deploy needed.
Guide
5min

How to Track Ad Performance and ROAS in Amplitude

Sync Google, Meta, and LinkedIn ads to Amplitude and connect spend to conversions—so you can measure CPC, CAC, and true ROAS.

Get in touch!

Adasight is your go-to partner for growth, specializing in analytics for product, and marketing strategy. We provide companies with top-class frameworks to thrive.

Gregor Spielmann adasight marketing analytics