The Funnel Playbook Find the hole before you carry more water. Stop guessing.
Growth is slow, so you post more. More traffic into a process that loses most of what enters costs you the one thing you cannot buy more of: evenings. This guide shows where people stop, why, and what to fix first.
More traffic is the most expensive lever you have.
Between a stranger arriving and a customer paying there is a sequence of steps, and at each one some people stop. Those drop-offs have specific, findable causes: a confusing page, a form that asks too much, a promise the product does not keep, a doubt with nothing to settle it. Traffic is additive and has to be re-earned every week. Conversion is multiplicative and applies to every visitor you will ever have. You cannot find the leak by intuition, because you know where every button is and you have never arrived at your own homepage with no context.
None of them is "optimise the website".
Improve one step and the gain applies to every visitor who ever arrives, through every channel, including the ones you have not started. Nothing on the traffic side works like that.
"Growth is slow" cannot be acted on. "Seventy per cent of people who start the signup do not finish it" has a location, a likely cause and a next task.
Add the place where the most interested people are failing to your list of requests, competitor features and your own conviction. It is often unglamorous and worth more than the feature.
A sharp drop right after someone commits is usually a mismatch, not a usability problem. Marketing sees the conversions, product sees the inactive users. Only the funnel shows they are the same people.
A channel with poor conversion may be sending the wrong people rather than being a bad channel. You cannot tell the difference without both views.
Waiting for revenue can take a full sales cycle. Watching the step you changed can take days. Measure the steps and you learn weekly instead of quarterly.
Fix the leak. Then get the traffic.
Not an optimisation project. A way to define one funnel properly, measure it simply, rank the losses by people and decide what to fix first, with an honest note on when the funnel is the wrong instrument.
Five parts, in order.
Six mechanisms, and the honest limit: a funnel converts the interest you have. It cannot rescue an offer people do not want.
Steps as states, the customer's decision end to end, cohort progression, a time window, segments, and one funnel nearest the money.
Events where the action is not a page, naming conventions, joining the visitor to the user, walking it yourself, five or six steps and a month of baseline.
Rank by the largest absolute loss, weight by effort, diagnose friction, doubt or mismatch, add qualitative evidence, change one thing, and look for where the constraint moves.
Define the funnel, instrument and verify, and the monthly improvement loop. Each line is a box you either can or cannot tick.
Nine ways funnel work turns into busywork. Every one is a failure of definition or prioritisation, not of tooling.
Five or six states in order, ending at the outcome nearest the money. Not pages. If someone can reach step four without step two, step two is not in this funnel.
From a device and network that are not yours, then again on mobile. Confirm your own journey appears in the data in the right order. Twenty minutes saves a month of confident wrong decisions.
Multiply drop-off by volume at every step. A step converting at ten per cent with two hundred entering loses a hundred and eighty. One at sixty per cent with five thousand loses two thousand.
- You run a one-person business and want to grow it without spending more evenings on promotion.
- You are willing to count cohorts and not just totals.
- You will change one thing at a time and say honestly when your volume is too small to read the result.
- You want a method for deciding what to fix first, not a growth hack.
- You want more traffic and a funnel is the last thing on your mind.
- You expect optimising the sequence to rescue an offer nobody wants.
- You want a tool recommendation or an instrumentation project that takes a quarter.
- You plan to change five things at once and call it an experiment.
Both your evenings and your appetite for promoting yourself run out.
The second runs out first. Every hour spent chasing traffic buys visitors who leak away at the same rate as the last batch. Every hour spent on the leak makes all the traffic you will ever get worth more. Find the hole before you carry more water. Then go and get the traffic, and keep it.
Luis Gonçalves
Twenty years in digital product development. Led the agile transformation at Nokia. Teaches at Porto Business School. Has spent the last decade working with founders across Saudi Arabia, Portugal, Brazil, and the broader emerging-market startup ecosystem.
He calls himself the Dark Entrepreneur because he believes the founder world is allergic to honesty. No success theater. No highlight reels. Just the real work that has to happen.