Funnels fail at the joins, not the stages. Here is where we look first when the numbers stop making sense.
Funnels fail at the joins, not the stages. When the numbers stop making sense, the stage everyone stares at is usually working fine and the handover into it is not.
Look at the transitions first
Take the conversion rate between each pair of adjacent stages rather than the rate of each stage on its own. A 3% visitor-to-lead rate means nothing until you know that 40% of the people who started the form abandoned it at the phone field.
The three joins that break most often
- Ad to landing page. The promise in the ad is not the headline on the page, so the visitor has to work out whether they are in the right place. Most will not.
- Form to follow-up. A lead that waits two days for a first reply has usually already spoken to someone else. This is an operations problem that presents as a marketing problem.
- Trial to paid. Users who never reached the feature that delivers the value churn on schedule, and no amount of renewal emailing fixes it.
Instrument the joins before optimising the stages
Most analytics setups measure arrivals and conversions and nothing in between, which makes every diagnosis a guess. Event tracking on field-level abandonment, time-to-first-response and first-key-action will tell you where to spend within a fortnight.
Fix in order of cheapness
Message match costs a copy change. Response time costs a process change. Product onboarding costs engineering. Working in that order gets the compounding gains first and buys the time to do the expensive one properly.
“The best marketing decision we made was agreeing what success meant before anyone spent anything.”
