The funnel self audit: count what you can see

I was on a call about a marketing job at a firm that buys lead lists. New names were arriving every week. I asked the recruiter one thing: who contacts them? "That I don't know," the recruiter said. "Probably some other people in their team in office." The list was paid for. Nobody on the call could say who was contacting them.

This self audit works the other way around. The audit opens with made-up example numbers so you can see how it works. Change a number and that step becomes yours. Flip any step you can't count to Dark. It shows you which stages of your funnel you can see and what the leaks you counted cost you a month. A dark step stays dark, and the tool tells you so. The audit is right below the steps, it's free, and it asks for no email.

Think of a radar screen. It only shows what its beam has swept. A dark part of the screen isn't empty. It just hasn't been checked.

Most funnel calculators fill in numbers you never counted. On October 21, 2026, I searched Google for "funnel audit calculator". I opened the four calculators on the first page. Three of them filled in rates the owner never typed, either as defaults or as "industry benchmarks". Take the funnel leak calculator at Technovier. It starts every visitor at a 50% contact rate and 35% qualified. Then it assumes 50% booked, a 65% show rate, and 25% closed. None of the four told the owner which stages he couldn't see.

That matters because a filled-in number looks like a measured one. Ours starts with example numbers too, but they are marked Example, and they never count toward your score or your dollar figure.

If you can't count a stage, no calculator can either. It can only guess, and the guess comes with decimals.

Grab last month's numbers, even rough ones. Then go step by step. We use the same six steps as our post on why most funnels are a form and a hope, and we count calls and messages too, not just the form:

1. Count the people who visited your website.

2. Count the people who started to reach out: started a form, or tapped to call or chat.

3. Count the people who reached out: sent a form, called, or messaged you.

4. Count the people who got a reply from you.

5. Count the people who got a reply within an hour.

6. Count the real sales talks, like quotes sent or estimates booked.

Flip any step you don't track to Dark. Don't guess. Then set your hire rate and what an average job is worth. Until you do, the dollar figure uses example numbers and says so.

It means the beam reached four parts of your funnel. You have real counts there. The other two are dark.

A dark stage isn't a failure. It's the most useful thing the audit can tell you, because a stage you can't see is a stage nobody is fixing. Form sends are often the easiest count, because the form emails you. The call back is harder to count, and the call back within the hour is harder still.

Before you trust any dollar figure, look at how many steps you counted. The tool only works out costs from the stages you counted.

On that call, I asked who contacts the leads, and the recruiter didn't know. That's a handoff with no name on it: a lead moves to the next step and nobody is holding it. If you can't write a name next to a handoff, nobody owns it.

In an interview, one leader put it this way: "we've had single points of people doing all of this." If a handoff has one person's name on it and that person is out, nobody else holds it. No report says so.

Before you run the audit, write a name next to each of the six steps. A step with no name beside it is a step nobody is watching, even if you know the number.

The math only uses your counts. Take the people you lost between two steps you counted. Multiply by your hire rate, then by your average job.

Say 30 people sent your form and only 18 got a call back. That's 12 people lost at one handoff. With a 25% hire rate and a $9,000 average job, those 12 people are worth about $27,000 in jobs a month. The audit shows that number for your leakiest counted handoff, and it leaves the dark stages out of the total.

The tool also checks one of those guesses against your count. A typical calculator assumes 50% of people who reach out get a reply, which is Technovier's default. The tool shows the percent you counted next to it. The gap between them is the part nobody measured.

One company I spoke with sells only to other businesses. Its leader told me why each lost lead hurts in that market: "There's only so many of them." In a small market, a dark stage costs more. You can't buy more traffic to make up for it.

Someone on a marketing team told me about advertising they ran in the spring. "We weren't ready. Our systems weren't quite ready, like the backend systems." Count the stages after the form before the ad money goes out.

If you can't count a single stage yet, this audit won't help you much. It will only tell you that the whole screen is dark. I can't tell you what a good rate looks like for your business either, and nobody can until you count.

Pick the first dark stage after the form. A shared inbox and a timestamp are enough to count a call back. The same two things count a call back within the hour. You don't need new software.

If you'd rather have someone light those stages for you, request a free site analysis. We'll look at your form, your follow-up, and the stages you can't see yet.

An industry average is someone else's count. It can't tell you where your business loses people, and it makes a step nobody counted look counted. The audit only counts the numbers you type in. The starting numbers are marked Example and never count.

A rough count from last month works. It beats a perfect average from someone else's business. Round to the nearest five if you need to.

Your numbers stay in your browser, and nothing you type is sent anywhere. The page only counts how many steps you counted, never the numbers you typed in them.