Table of contents

• How do you define funnel stages?
• Where is the real leak?
• What breaks funnel measurement?

How do you define funnel stages?

Stages must be observable events, not intentions. "Considering a demo" is not a stage. "Viewed the pricing page" is.

A workable B2B SaaS funnel:

1. Session started
2. Reached a product or pricing page
3. Reached the demo request form
4. Started the form
5. Submitted the form
6. Demo attended

Two rules keep it honest. Every stage must be measurable without inference, or the number is a guess dressed as data. And stages must be sequential, meaning step four is only reachable through step three, or the drop-off percentages describe nothing.

Real behavior is not linear. Visitors loop, leave, and return over weeks. The funnel is a deliberate simplification of that, useful as a diagnostic frame rather than as an account of how people buy.

Where is the real leak?

The biggest absolute drop is usually at the top, and it is usually the wrong place to start.

Ten thousand sessions producing 500 pricing-page views is a 95% drop, and most of those people were never buyers. Fifty form starts producing five submissions is a 90% drop among people who declared intent, and every one of them is recoverable.

Priority comes from three things multiplied together: the size of the drop, the value of the people dropping, and how fixable the cause is. A late-funnel step with high intent and an obvious cause outranks a top-funnel step with more raw volume.

What breaks funnel measurement?

Steps that share a URL. Multi-step forms and modals change state without changing the page, so every step registers as one event unless you fire custom ones.

Cross-device journeys. A visitor researching on mobile and converting on desktop is two sessions, and the funnel reports one abandonment and one direct conversion.

Entries mid-funnel. Paid traffic landing directly on pricing skips stages one and two, which inflates later-stage rates unless segmented by entry point.

Time windows. B2B buyers return over weeks. A session-scoped funnel counts every visit as a fresh failure.

What breaks funnel tracking on Webflow?

Collection pages generate the URLs, not you. A stage like "reached a product page" spans every item in a Collection, so define it as a path pattern such as /products/ rather than as a list of URLs. The Designer template behind those pages is never served, so it will never appear in the data.

The last stage needs an event. Webflow's form success state renders in place with no page load, so a funnel defined on URLs has no bottom step. Event tracking covers the two ways to fix that.

Related terms

Funnel analysis · Drop-off rate · Micro conversion · Form abandonment · Event tracking

Deeper reading: CRO in 2026, the complete guide. Service: Conversion Rate Optimization.

FAQ

How many stages should a conversion funnel have?

Four to six for most B2B SaaS sites. Too few hides where the loss happens, too many fragments the sample so each step lacks the volume to read reliably.

Is a conversion funnel the same as a sales funnel?

No. A sales funnel describes the commercial process from awareness to closed deal, including offline stages. A conversion funnel is the measurable on-site portion, usually ending at a form submission or purchase.

What is the difference between a conversion funnel and a user journey map?

A funnel counts people at defined steps and produces percentages. A user journey map describes what a person is trying to do at each stage and what gets in the way, without counting anyone. The funnel tells you which step loses people. The map suggests why that step is hard.