How to Audit Your Own Revenue System in One Afternoon
You don’t need to hire anyone to find out where your revenue is leaking. You need four hours, access to your own numbers, and a willingness to write down answers you won’t like.
What follows is the same sequence we use at the start of a client engagement, stripped down to what an owner can run alone. It won’t be as thorough as a full assessment. It will be thorough enough to tell you which part of your business to look at first, which is the decision most manufacturers get wrong. This is one piece of a broader picture of where revenue leaks out of an industrial business.
What is a revenue system audit?
A revenue system audit is a structured walk through every stage a customer passes on the way from stranger to repeat buyer, checking each stage for a defined owner, a defined process, and a number you can actually see. Anywhere all three are missing, revenue is leaking.
The word system is doing real work in that sentence. Most owners audit a function, usually sales, and find that sales is doing roughly what sales does. The leaks are almost never inside a function. They’re in the handoffs between them, which is exactly where nobody’s job description ends up pointing.
What you need before you start
Gather these first. Hunting for them mid-audit is what turns four hours into two weeks.
- Last 12 months of revenue by customer
- Last 12 months of revenue by product line or service category
- Every quote or proposal issued in the last 6 months, with outcome where known
- Your CRM’s total record count, plus how many were touched in the last 90 days
- Your marketing spend by channel for the last 12 months
- A list of customers who bought in the prior year and not this one
- Your last 20 new customers and where each one came from
If you can’t produce that seventh list, stop and note it. Not knowing where your customers come from is itself a finding, and a significant one.
Block the time. Turn off your phone. This doesn’t work in twenty-minute pieces between operational fires, which is the same reason it hasn’t happened yet.
The eight stations
Work them in order. The framework below is TPG’s Predictable Revenue Framework, which maps the full path revenue travels through a business. At each station, answer three questions: who owns this, what’s the defined process, and what number tells us it’s working. Write down every place you can’t answer all three.
1. Opportunity analysis and strategy
Ask: Which customer segments produced your highest-margin revenue in the last 12 months, and does your current sales effort point at those segments?
Pull the revenue-by-customer list and sort it by margin rather than by revenue. Most manufacturers find that their largest customer by volume isn’t their best customer by contribution, and that sales attention correlates with volume rather than with margin. That gap is the finding.
Leak signal: Your top three customers by revenue aren’t your top three by margin, and nobody has looked at this in the last year.
2. USP discovery
Ask: If a prospect asked three of your salespeople why they should buy from you rather than your closest competitor, would you get the same answer three times?
Actually run this. Ask three people separately, in writing, and compare. Then ask two customers the same question about why they chose you.
Leak signal: Your team’s answers differ from each other, or your customers’ answers differ from your team’s. When the reason to buy isn’t consistent internally, every deal gets re-argued from scratch and price becomes the default tiebreaker.
3. Front-end management
Ask: Where did your last 20 customers come from, and who is responsible for each of those sources continuing to produce?
This is the station where the seventh item on your gather list earns its keep. Map each of the last 20 to a source. Then check whether anyone owns that source as part of their job.
Leak signal: Referrals and repeat business account for most of the list, and nobody owns generating either one. That’s not a healthy source mix, it’s a business coasting on relationships built years ago.
4. Offer and conversion
Ask: Of the quotes issued in the last six months, how many are still open with no scheduled next action?
Count them. Then count how many were followed up more than twice.
Leak signal: A meaningful stack of quotes sitting in an undefined state. This is the most common and most recoverable leak in an industrial business, because the cost of acquiring those opportunities is already sunk. If you also find that your CRM’s touched-in-90-days count is a small fraction of its total record count, you have the same leak upstream.
5. Stick programming
Ask: What happens in the first thirty days after a new customer places their first order, and is any of it deliberate?
Stick programming is the term for what you do immediately after the sale to make the decision hold: confirming the customer chose correctly, setting expectations, and building the operational relationship before anything goes wrong. In manufacturing this is usually the gap between the sales rep who won the account and the operations team who now services it.
Leak signal: The answer is that the order goes to production and someone calls if there’s a problem. First orders that go quiet often mean a customer who tried you once and drifted back to their incumbent, and nobody noticed because the account was never technically lost.
6. Revenue expansion
Ask: How many of your customers buy from only one of your product lines, and does anyone have the job of changing that?
Sort your revenue-by-customer data against your product line data. Count single-line customers.
Leak signal: A large share of single-line customers with no owner for cross-sell. This is typically the largest identifiable opportunity in a mid-size manufacturer and the one least likely to be anybody’s responsibility, because it falls between the salesperson who owns new business and the service team who owns the account.
7. Retention
Ask: Which customers bought last year and not this year, and can you say why for each one?
Pull that list. For each name, write the reason. Where you can’t, write unknown.
Leak signal: More than a couple of unknowns. Silent attrition is the most expensive leak in the business because it removes revenue you’d already paid to acquire, and it doesn’t trigger any alert. Nobody files a report saying a customer stopped calling.
8. Referrals
Ask: In the last 12 months, how many referrals did you receive, and how many did you ask for?
The second number is the one that matters.
Leak signal: You receive referrals and never request them. Most manufacturers have genuinely satisfied customers who would refer if asked and are never asked, because asking feels awkward and no process exists to make it routine.
Scoring what you found
Go back through your eight sets of notes and mark each station red, yellow, or green. Red means no owner, no process, or no visible number. Yellow means one of the three exists. Green means all three.
Now the part that determines whether this was worth the afternoon: do not fix the reds in order of how much they bother you. Fix them in the order the framework runs.
Revenue moves through those eight stations in sequence. A leak at station four contaminates everything downstream of it, and fixing station seven while station four is bleeding just means you’re retaining fewer customers more carefully. Your first project is the earliest red on the list, not the loudest one.
The single most common mistake owners make after an exercise like this is going straight to station three and buying more lead generation, because generating leads feels like growth. If stations four through seven are red, more leads pour into a container with holes in it and the marketing spend gets blamed for a problem it didn’t cause.
What this audit can’t tell you
Three limits worth naming.
It won’t tell you the size of each leak in dollars. You’ll know where the holes are, not how much is running out of each. Sizing requires digging into individual opportunities and customers, and that’s the work that takes longer than an afternoon.
It relies on your own reporting, which may be part of the problem. If your CRM is untrusted, the numbers you pull from it inherit that.
And it’s a self-assessment, with the blind spots that implies. The stations you’re most confident about deserve a second look, because confidence and visibility aren’t the same thing.
None of that makes the exercise less worthwhile. Knowing which of eight stations to work on first is worth considerably more than a precise measurement of the wrong one.
For the fuller picture of how these leaks develop and what fixing them involves, see our complete guide to revenue leaks in industrial businesses.
Frequently asked questions
How long does a revenue system audit really take? The version in this article takes about four focused hours if you gather your data first. Gathering the data takes longer than the audit at most companies, which is a finding in itself.
Who should run the audit? The owner or general manager, alone. Running it as a group meeting produces consensus rather than findings, because nobody marks their own station red in front of colleagues.
What if we don’t have a CRM? Run it anyway using quotes, invoices, and your customer list. The absence of a CRM will show up as a red at several stations, which tells you something useful about sequencing.
How often should we do this? Annually as a full pass, with a quarterly check on whichever stations you marked red. More often than that and you’re measuring rather than fixing.
What’s the difference between this and a revenue leak assessment? Scope and depth. This finds which stations are broken. A full assessment quantifies each leak, examines the underlying data rather than your summary of it, and produces a sequenced plan with dollar figures attached.
When you want the sized version
If you run this and come out with four reds and no clear sense of which one is costing you the most, that’s the normal outcome. Prioritizing by dollar impact requires getting into the underlying data. Schedule a Discovery Call and we’ll talk through what you found and what it would take to put numbers on it.