Find the step where people give up.

Line up the events between a first visit and a paid plan. See how many people make it through each step, how long it takes, and what the people who left did instead.

events per funnel, each with its own filters
2–8
day conversion windows: 7, 14, 30 or 90
7–90
ways to read a result: steps, over time, time to convert
3
people per drop-off CSV export
50k

What it does

A funnel is a saved sequence of events your whole team can open. ClickClacks counts who entered, who finished inside the conversion window and where the biggest drop sits. When a funnel crosses from your marketing site into your app, it also shows how many people kept the same ID along the way.

Builder

Build it from events, see it before you save.

Pick autocaptured or custom events and filter any step by its properties. A live preview re-runs as you edit, so you see the counts at every step before the funnel exists.

  • Start from a template: Signup → activation, Browse → checkout or Trial → paid.
  • Count unique people, sessions or companies, in order or in any order.
  • A step that matched nothing is flagged before you save.

Results

One funnel, read four ways.

Steps shows the bars and the drop between each one. Over time shows whether conversion is moving. Time to convert shows how long people take. Break it down by a property of the first step to compare sources or landing pages.

  • Each broken-down value shows its gain or loss in points against the overall rate.
  • A step table under the chart: from previous, from start, dropped, median time.
  • Compare with the previous period for a change in points.
  • Or compare up to five saved segments side by side.

Drop-off

Follow the people who left.

Click any drop to see what those people did instead, ranked, including “Nothing — session ended”, and the median time before they gave up. Then act on them.

  • Save them as a People segment: did step 2, never did step 3.
  • Open the list in People, or export it as a CSV.

Identity

Check that identity carries from your site into your app.

When a funnel runs from your marketing site into your app, it shows how many people kept the same ID across the hop. If the ID isn’t carrying across, the funnel says so and explains how to fix it.

  • Measured on people who followed a link between your domains and arrived within 30 minutes.
  • Tagged links (?_ccid=) carry the ID to the first pageview on the other domain.

Questions it answers

Ask it the way you’d ask a colleague.

How ClickClacks answers it

Make a pageview filtered to /pricing the first step and trial_started the second. The result leads with overall conversion and how many of the people who entered converted inside the window.

18.2%

Overall conversion · 2,271 of 12,480

Who it’s for

Three ways teams use it.

  • Agencies

    Build Browse → checkout across the site you designed and the store it sells into, and show the client which step loses the most people and who left there.

    One funnel per client organization

  • Indie developers

    Send two custom events, trial_started and subscription_created, and the Trial → paid template shows whether trials turn into paid plans, and how long it takes.

    Template: Trial → paid

  • Product teams

    Save the onboarding funnel once so everyone reads the same numbers. When a step slips, save the people who dropped as a segment and hand it to research.

    Drop-off → Save as segment

How it works

Under the hood

  1. 01Choose the stepsTwo to eight events, autocaptured or custom. Any step can carry property filters, so a pageview of /pricing is a step of its own.
  2. 02Set the rulesPick a conversion window, a counting mode and whether order matters. The live preview runs against the last 30 or 90 days while you edit.
  3. 03Read the resultOverall conversion, converted, median time to convert and the biggest drop sit on top, with the chart and the step table below.
  4. 04Follow the people who leftOpen a drop to see what people did instead and how long they lasted, then save, open or export them.

Spec sheet

from the product

Steps
2–8 events
Filters per step
Up to 10, with is, is not, contains and more
Conversion window
7, 14, 30 or 90 days
Counting
Unique people, sessions or companies
Order
In order or any order
Live preview
Last 30 or 90 days
Breakdown
Top 3 values, 500+ entrants
Compare
Previous period, or up to 5 segments
Drop-off export
CSV, up to 50,000 people

FAQ

Funnels, in detail.

  • Yes. A step is any event plus optional property filters, up to ten per step, each with its own comparison: is, is not, contains and more. A pageview filtered to /pricing, or a click on one button, is a step of its own.

  • Counting unique people enters each person once. Counting sessions treats each person and session pair as a separate funnel entry, so someone who tries twice in two visits counts twice.

  • Yes, when your events carry a company. Switch Count by to companies and each company is counted once, and the step and drop-off lists become lists of companies. The choice is saved with the funnel.

  • Yes. Open a step in Flows and choose Open as funnel. The funnel builder opens with that path already filled in as its steps.

  • With order on, steps must happen in the sequence you listed. With it off, any order counts, as long as every step happens inside the conversion window.

  • Open the drop and save it as a segment (did the step before, never did the next one), open that group in People, or export it as a CSV of person ID, last event and last seen, up to 50,000 people.

  • A funnel that crosses from one source to another shows how many people kept the same ID across the hop, so you can see whether it did. Once both domains are listed in Sources, links between them are tagged automatically with ?_ccid=, so the ID travels with the click.

  • Small groups swing wildly, so only the top three values with at least 500 entrants are charted. Every other value is still listed in the table, with a note saying why it isn’t charted.

Start hearing your clicks.

Install one script tag and see your first report in minutes.

About 5 KB · pageviews, clicks and scroll depth, automatically