How to run a revenue leak audit on your own business
The same diagnostic we run, written out so you can do it yourself in an afternoon. No tools, no spend, one spreadsheet.
Most sales advice assumes you already know what is broken. This does the opposite — it is a procedure for finding out, and it works before you have decided anything.
It takes an afternoon. You need your CRM or your inbox, a spreadsheet, and a willingness to write down numbers you may not like.
Before you start: pick a window
Choose the last 90 days, or the last 30 deals, whichever is bigger. Shorter than that and one unusual month distorts everything; longer and you are auditing a business that no longer exists.
Write the window at the top of the sheet. Every number below has to come from the same one, or the ratios mean nothing.
Step 1 — Count the stages
Six columns. For your chosen window, count how many people reached each:
- Arrived — visited your site, or entered a conversation
- Enquired — filled in something, replied, asked
- Booked — a call in the calendar
- Attended — actually turned up
- Proposed — received a number
- Closed — paid
Absolute counts, not percentages, and if you cannot get a number exactly, estimate it and mark it as an estimate. An honest estimate is more useful than a blank.
Step 2 — Calculate the step-down rates
Between each pair of stages, work out what share survived:
enquired / arrived = capture rate
booked / enquired = booking rate
attended / booked = show rate
proposed / attended = qualification rate
closed / proposed = close rate
Do not compare these to industry benchmarks. Benchmarks are averages of businesses that are not yours, and chasing one is how people fix things that were never broken. Compare them to each other.
Step 3 — Find the cliff
Read the five rates in order. You are looking for the one that is dramatically worse than its neighbours — not slightly worse, dramatically. That discontinuity is your leak.
A chain that degrades gently at every step is a business with no single problem and a general quality issue. That is rarer than it sounds. Most sheets have a cliff.
Step 4 — Ask what the cliff means
The stage where the drop happens tells you which link is broken:
- Capture — message or offer. They understood and did not care.
- Booking — funnel. They cared and the path was too hard.
- Show — funnel or message. The booking was not real to them.
- Qualification — leads. Wrong people, arriving efficiently.
- Close — conversion or pricing. This is the expensive one.
Note that two very different problems can produce the same cliff, which is why the next step is not optional.
Step 5 — Read the losses in their own words
Take the ten most recent deals that died and find what the person actually said. The email, the call note, the last message before silence.
Do not summarise them into categories yet. Read them as sentences. Patterns in real language are far more diagnostic than patterns in a dropdown field, because a CRM stage called "Lost — no budget" hides three different failures and the email that preceded it usually names the real one.
Step 6 — Check the end of the chain
Two questions that take five minutes and that almost every audit skips:
How many clients in the window referred someone? If the answer is none, the ecosystem link is broken, and everything upstream is working harder than it should to replace people who should have brought others.
What share of revenue came from existing clients? A business winning only new logos is a business paying full acquisition cost for every pound it earns.
Step 7 — Write one sentence
At the bottom of the sheet, finish this:
Our revenue is constrained at the ___ link, because ___, and the evidence is ___.
If you cannot fill in the third blank from the numbers above, you have a hypothesis, not a diagnosis. Go back to step 5 and read more losses.
What to do with it
Fix the named link. Only that one. Then re-run this in 90 days with the same window length and see whether the cliff moved.
The discipline is in the "only that one". The temptation after an audit is to fix everything you noticed on the way, and that is how you end up unable to tell what worked.
If the sheet does not give you a clean answer
Sometimes it does not. Two stages look equally bad, or the volumes are too small for the ratios to mean anything, or the losses do not read as a pattern.
That is worth an outside pair of eyes — someone who has read a few hundred of these knows which ambiguities matter. The audit is free, takes 45 minutes, and there is no pitch at the end: you get the name of the link and the fix whether you hire us or not.
Related reading: the seven places B2B revenue leaks and why "we need more leads" is usually the wrong diagnosis.