Where Is My Business Actually Losing Money? A Revenue-Leak Audit From Your Own Data

Problem: You can recite your revenue number and your biggest expense line from memory, but if someone asked which single process is quietly bleeding your margin right now, you would guess.

Quick Win: The money is almost never in the line you watch. It leaks in four quiet places: work done but never billed, leads and follow-ups that die on their own, skilled people stuck on repetitive tasks, and discounts that never hit a report. EY research puts just the billing-and-follow-up leak at 1 to 5 percent of realized operating profit (EBITDA) per year, and roughly 42 percent of companies experience some form of revenue leakage, the money a business earns but never actually collects, or quietly wastes (RevenueGrid, summarizing EY and MGI Research). The rule: you do not fix the leaks, you fix the leak, the one real bottleneck, and you rank it by recoverable dollars from your own numbers before you touch anything.

---

The Four Places Mid-Market B2B Firms Leak Money

Most owners hunt for waste in the profit and loss statement, the P&L. Waste that shows up on the P&L is waste you already manage. The dangerous kind never appears as a line item. It is revenue that should have arrived and did not, or hours that produced nothing you can sell.

Across industries, companies lose roughly 2 to 10 percent of earned revenue to systemic errors that never show up on a P&L line, and software businesses alone lose about 3 to 5 percent of yearly revenue (ARR) to it (LeaksShield). Here is where it actually hides.

The time zone deserves a hard number. Eagle Hill Consulting found that 68 percent of US workers say they regularly spend time on low-value, inefficient tasks (Eagle Hill Consulting). That is not a morale stat. It is payroll you are spending on work that produces nothing sellable, and it never shows up as a loss.

Why You Fix THE Leak, Not the Leaks

Here is the part that changes how you run the audit. Once you find four leaks, the instinct is to fix all four. That instinct is expensive, and the math says so.

Eliyahu Goldratt's Theory of Constraints, introduced in his 1984 book The Goal, starts from one blunt claim: every system has a single bottleneck that decides how much work the whole system can finish, and no process can finish work faster than its slowest step (Lean Production). The useful part is what follows. At any given time, one main bottleneck limits performance, and improving anywhere else has little effect (TOC Institute).

Read that again with a revenue-leak audit in mind. If your real bottleneck is a sales team that cannot follow up fast enough, then automating your proposal process is a genuine improvement, but to a part of the system that was never the limit. You spent money and moved zero additional revenue. The chain got no stronger, because you reinforced a link that was not the weak one.

This is why a 20-item efficiency checklist wastes effort, mathematically. Nineteen of those items sit behind the bottleneck, not at it. The audit's entire job is to find the one that is not, size it in dollars, and put it first. Goldratt's method turns this into a sequence: identify the bottleneck, get more out of it, line everything else up around it, expand it, then repeat (TOC Institute). Notice step one is identify, singular. Everything downstream depends on getting that one right.

Build the Leak Map From Data You Already Have

You do not need a data warehouse or a new tool to run this. The evidence for all four leak zones already sits in systems you own. The audit means checking numbers you already have against each other, not a research project.

• Customer-tracking software (CRM). Pull every open opportunity and sort by last activity date. Deals with no touch in 30, 60, 90 days are your deals-in-progress leak, quantified. Sum the value. That is revenue you are already losing, on a timer.
• Invoices and contracts. Line up signed contracts against issued invoices for one quarter. Un-billed work, unenforced price escalators, and under-metered usage surface here. This is the EY leak in your own numbers.
• Time. You do not need timesheets. Ask your three most expensive people to list the recurring work they redo by hand. Multiply hours by loaded cost. That is the time leak, priced.
• Won-deal data. Average sale price against list price, split by salesperson and by segment, exposes the margin leak. The spread is discount you gave without deciding to.

Four pulls, four numbers. None of them requires buying anything. The reason most companies never do this is not access. It is that nobody owns the reconciliation, so it never happens, and the leak keeps running because no one is looking at the two lists side by side.

The Evidence Test: A Leak You Cannot Source Is a Guess

This is the line that separates an audit from an opinion. Every leak on the map has to trace to a figure from your own data. If it cannot, it does not go on the map.

"I think we lose deals because follow-up is slow" is a hypothesis. "We have 140 open opportunities worth 2.1 million dollars with zero activity in 45 days" is a leak. The first is a feeling that will get argued down in a meeting. The second is a number your CFO can act on before lunch.

The reason this matters is political, not just analytical. Audits die in the room where they are presented. The moment a finding rests on a consultant's benchmark ("firms like yours typically lose 4 percent here") instead of your own figure, someone in the room says "we are not like those firms," and the finding is dead. A number pulled from your own CRM cannot be waved away. It is your data, describing your company, and the only argument left is what to do about it.

So the test for every line on the map is one question: can you point to the source figure. Benchmarks from EY or anyone else are useful for sizing the category and knowing the leak is normal, but the finding itself has to come from inside the building.

Rank by Recoverable Dollars, Not by How Annoying It Feels

The most common way a good audit goes wrong: the team ranks leaks by how irritating they are day to day. The thing everyone complains about gets fixed first. That is almost never the biggest leak.

Rank by recoverable dollars against effort to recover. A leak that annoys everyone but traps 40,000 dollars a year loses to a silent one trapping 600,000. The map has to force that comparison instead of letting the loudest complaint win.

Only the third row respects the bottleneck. It puts the single highest-return leak at the top and everything else in a queue, which is exactly what the Theory of Constraints says to do: fix the real bottleneck, then and only then move to the next one (TOC Institute). The ranking is the deliverable. A leak map with no order is just a longer list of problems.

The One-Pager Beats the 40-Slide Deck

If your revenue-leak audit produces a 40-slide deck, it already failed. Not because the analysis was wrong, but because nobody acts on 40 slides. They get filed, referenced once, and forgotten, which is how most operational audits end.

The deliverable that changes anything is a ranked one-pager. Four to six leaks, each named in a sentence, each tied to a number from your own data, sorted by recoverable dollars, with the one to fix first sitting at the top. A CFO can read it in two minutes and a COO can assign it before the meeting ends. That is the difference between an audit that produces motion and one that produces a document.

The deck is optional supporting evidence. The one-pager is the product. If the person who commissioned the audit cannot look at the top line and know exactly what to fix first and roughly what it is worth, the work has not landed, no matter how good the underlying analysis is.

When a Revenue-Leak Audit Is the Wrong First Move

Honesty about failure modes is the difference between a method and a pitch. A revenue-leak audit is the wrong first move in three situations.

Before product-market fit. If you have not proven people will pay you repeatably, you do not have a leak, you have a demand problem. Improving a process you might scrap in six months is effort spent in the wrong place. Find the demand first.

Below roughly 1 million dollars in revenue. At small volume, the leaks are real but the recoverable dollars are too small to justify the audit's cost, and the fix is usually "sell more," not "reconcile invoices." The audit pays off when there is real volume, real data, and a margin worth protecting.

When you already know the bottleneck. If it is glaringly obvious that follow-up is the bottleneck and everyone agrees, you do not need an audit to tell you. Go fix it. The audit earns its keep when the leaks are hidden or contested, not when they are staring you in the face.

There is also a quieter failure mode: running the audit and then not acting on the ranking. A leak map that produces a ranked one-pager which then sits in a drawer is worse than no audit, because now you know exactly what you are losing and chose not to fix it. Do not commission the diagnosis unless someone owns the fix.

What We Install: A Ranked, Evidence-Backed Bottleneck Map From Your Own Figures

This is the first of our five outcome pillars, and it is deliberately first because everything else depends on knowing where the value actually leaks before you spend a dollar chasing it.

We build the leak map from your data, not a benchmark set: your customer-tracking software, your invoices, your contracts, your team's own account of what they redo by hand. Every leak on it traces to a figure you can point to. The output is a ranked one-pager, ordered by recoverable dollars, with the single real bottleneck at the top and a fix tied to each line. Not a deck. A decision.

See exactly what that deliverable looks like on the bottleneck diagnosis page, and if you want the framework behind the ranking, we wrote it up in why the Theory of Constraints beats a 20-item efficiency checklist. If your instinct is to build this reconciliation internally, read why buying the outcome usually beats building the tool first.

Frequently Asked Questions

Where do most mid-market businesses actually lose money?

Not in the expense line they watch. In four quiet places: work delivered but never invoiced, leads and follow-ups that die on their own, skilled people stuck on repetitive manual work, and discounting that never hits a report. EY research puts the billing-and-follow-up leak alone at 1 to 5 percent of realized operating profit (EBITDA) per year, with roughly 42 percent of companies experiencing some form of revenue leakage (RevenueGrid, summarizing EY and MGI Research).

How do I find the bottleneck in my business?

Stop looking for a list. The Theory of Constraints says every system has one bottleneck that decides how much work the whole system can finish, and improving anywhere else has little effect (TOC Institute). So the job is to find the single real bottleneck, size the dollars trapped behind it from your own data, and fix that one first. The audit ends in a ranking, not a to-do list.

What should a revenue-leak audit actually deliver?

A ranked one-pager. Each leak named, each tied to a number from your own customer-tracking software (CRM), invoices, and time data, sorted by recoverable dollars, with the one to fix first at the top. If the deliverable is a 40-slide deck or a benchmark set from someone else's companies, it already failed. The test for every line: can you point to the source figure that proves it.

When is a revenue-leak audit the wrong first move?

Before you have found product-market fit, and below roughly 1 million dollars in revenue. If you have not proven repeatable demand, your problem is sales, not leakage, and improving a process you might scrap wastes the effort. The audit pays off when you have real volume, real data, and a margin worth protecting.

---

If you can name your revenue and your biggest cost but not the one process bleeding your margin, that gap is the whole problem, and it is fixable in a way that produces a number, not an opinion. We build a ranked, evidence-backed leak map from your own figures, hand you the one-pager that says what to fix first and what it is worth, and leave you with something you act on the next morning. See what we install for companies →