A business came to us wanting to launch. When we sat down to run the calculation, there was nothing to run it on. No sales history worth the name — there was barely a sales function at all. Every number in the model was going to be a guess.
We could have refused on those grounds. Instead we built the model on assumptions, wrote each one down, and launched a small campaign specifically to test them. A plan built on assumptions beats no plan, as long as everyone knows which parts are assumptions.
One of them was the close rate. We assumed 65% — inquiries that would become jobs.
The real figure came back under 30%.
Not a miss. Less than half. And notice which side of the table it came from: the market behaved roughly as expected. It was the business’s own number that fell apart.
What was actually happening
We looked at what happened to an inquiry after it arrived.
Calls went unanswered. Callbacks were slow when they happened at all. Nobody followed up. There was no second attempt on anyone who didn’t buy immediately.
The people were not lazy or incompetent. They simply weren’t selling. They were doing something else that looks similar from a distance.
Think of walking into a corner store for a loaf of bread. You ask for the bread, someone takes it off the shelf, hands it to you, rings it up and tells you the total. That is a complete and correct interaction. Nobody involved has done anything wrong. Nobody has sold anything either.
That was the sales department. Order-taking dressed as sales. It works fine when the customer has already decided — and it converts under 30% of the people who haven’t.
That distinction is worth more to most owners than any advertising advice. Ask yourself honestly whether your team sells, or hands things across a counter. The tell is simple: what happens to the person who says let me think about it? If the answer is “nothing,” you’re a shop.
What we did about it
Not a campaign change. Not a landing page test. Not a bid adjustment.
The sales team was replaced. Not retrained — replaced, and rebuilt.
That’s an uncomfortable sentence to publish, and it’s here because softening it would misrepresent what actually fixed the problem. The gap between 65% and 30% was not going to be closed by a script.
The advertising, meanwhile, was paused. There was no point buying more inquiries for a process that was losing seven out of ten of them.
What happened
After the rebuild, the close rate came back at just over 50%.
Below the 65% we’d assumed. Short of the plan.
And it was enough. The economics closed at 50%. The business became viable, the advertising was worth running, and the project continued.
This is the part that most contradicts how people read a forecast:
Your assumption is a hypothesis. The threshold is the decision.
Missing your assumption is not failing. Falling below the threshold is failing. Those are different lines, and confusing them causes two opposite errors. One owner hits 50% against a 65% forecast and shuts down a working business because he “missed target.” Another hits 90% of a forecast that was below threshold to begin with, and celebrates.
Go back to your two-point test. The number that matters isn’t the one you assumed. It’s the one where the economics close.
The order of repair
When month one refutes your calculation, the instinct is to reach for the advertising, because that’s the thing that just changed and the thing you can adjust today.
Wrong order, almost always. Here’s the order that works — cheapest and highest-leverage first.
1. Whoever answers the phone
Free, immediate, and the most common culprit by a distance.
How many callers reach a human? How fast is the callback on a form? What happens on the second attempt — is there one?
One appliance repair company had leads at a quarter of the affordable price and a dispatcher converting about one in ten of them into a booked visit. The advertising was excellent. The business was still throwing away nine of every ten customers it paid for.
Nothing else on this list will out-earn fixing this, and it usually costs nothing but attention.
2. The sales conversation
Once people are reaching a human, what does the human do?
The most useful diagnostic here costs almost nothing: be a customer of your competitors. Call three of them the way a homeowner would, and take notes. Then have someone call you the same way.
The differences are never subtle. Here’s what you’ll typically find on the other side of those calls. They are more persistent. They talk more. They drop the professional register and speak like a person. They follow up without being asked. They offer options instead of a price. And they do not end the call when the customer says they’ll think about it.
That comparison tells you more about your close rate than any analytics tool, and it tells you in an afternoon.
One caveat worth insisting on: you know your trade better than any consultant does. When an outsider says your close rate looks low, they’re comparing to other markets and they may be wrong. The mystery-shop is what settles it, because it replaces opinion with two transcripts sitting next to each other.
3. The website
If people arrive and leave without asking for anything, the site is the constraint.
Compare against competitors, feature by feature. Is there a callback request? A chat? Prices, or at least ranges? Photos of actual work rather than stock images?
And when you find a gap, the objection is usually operational rather than technical. We can’t add chat, there’s nobody to staff it. Fine — then it isn’t a chat, it’s a form that looks like a chat. The visitor leaves a message, nobody answers in real time, and you call them back within the hour. You’ve captured the contact, which is the entire point. Or put an assistant on it trained on your own material to handle the first questions.
Almost every “we can’t because of how we operate” has a version that captures the contact anyway. Look for that version.
4. The offer and the price
Slower, and the most powerful.
Raising your price raises your maximum proportionally, which is the fastest way to make advertising affordable — and it’s the move owners resist hardest, because it feels like it costs customers. Often the customers it costs are the ones you were losing money on.
Changing what you sell can move more than price. The panel-bathroom remodeler is the clearest version: same margin percentage, same crews, a different product — and a maximum that went up without anyone being hired or any bid being touched. It was reached through the product, not the ad account, which is why it belongs on this list at all.
5. The advertising
Yes, it’s last.
Not because it doesn’t matter — it does, and the difference between a managed account and an unmanaged one is real money. But because in a first month that refutes your calculation, the advertising is rarely what broke. And every week spent tuning campaigns against an economic gap is a week the actual problem goes unexamined.
The test that tells you which
Here’s how to know whether you’re facing an optimization problem or a business problem.
Find the current market range for clicks. Compare it to your maximum.
If competitors are comfortably paying about twice what you can afford, that is not recklessness on their part. One competitor miscalculating is possible. Several of them sustaining it is not a mistake — it’s information.
It means their close rate, their average job, their repeat business or their website is beating yours. Possibly all four. You are not looking at an advertising gap. You are looking at a scoreboard.
And it has a mechanical consequence people miss. At half the market rate you don’t get a smaller version of the campaign — you get a broken one. Ad platforms need a certain flow of conversions to learn who to show you to. Bid far below market and the system never gets enough data to work, so you lose money and never find out whether the channel could have worked.
Which is why the answer at that point is to stop, not to try harder.
Sometimes the fix removes the need entirely
One storage business spent a month on its own sales process, tripled its revenue, and never ran a campaign at all.
The conversion step was the constraint. Fixing the constraint released demand that was already there and already paid for. Advertising would have added more demand to a process that was losing it — which is to say, it would have bought more of the problem.
You may reasonably wonder why an agency would engineer that outcome, so here’s the answer. If we’d launched as things stood, the advertising wouldn’t have worked — the economics didn’t close. If we’d launched and then fixed the sales process, the need for advertising would have evaporated anyway, and we’d have spent the client’s money proving it. We’d have ended up telling that story too, with a different spin — they asked us to pause the campaigns because they couldn’t keep up — and it would have been a short relationship. We try to avoid short relationships.
That case isn’t typical. It matters because of what it implies about the ones that are: when a business converts at half of what it should, advertising isn’t the thing that’s missing. It’s a way of paying to avoid a conversation you could have had for free.
The rule
Everything here is one idea in different clothes:
Fix the business, not the advertising.
Advertising is a catalyst. Catalysts don’t fix reactions. They make whatever is already happening happen faster. If what’s happening is that seven out of ten interested customers hang up without buying, a better campaign delivers more people to hang up.
Fix the reaction. Then add the catalyst. And when the numbers do hold, remember that month one is a measurement — which side of the table missed is the whole question.
This chapter is drawn from Win in the Spreadsheet First. The full method and the same treatment per trade are in the book.
Frequently asked questions
- Why aren't my leads converting into jobs?
- Usually because nobody is selling. There's a difference between selling and order-taking that looks identical from a distance: someone asks for the thing, you take it off the shelf, hand it over and state the total. That's a complete and correct interaction in which nothing has been sold. It works fine when the customer has already decided, and it converts under 30% of the people who haven't. The tell is simple — what happens to the person who says let me think about it? If the answer is nothing, you're a shop, not a sales team, and no campaign change will fix that.
- What's the cheapest way to diagnose a low close rate?
- Be a customer of your competitors. Call three of them the way a homeowner would and take notes, then have someone call you the same way. The differences are never subtle. What you'll typically find on the other side: they're more persistent, they talk more, they drop the professional register and speak like a person, they follow up without being asked, they offer options instead of a price, and they don't end the call when the customer says they'll think about it. That comparison tells you more about your close rate than any analytics tool, and it tells you in an afternoon.
- In what order should I fix things when the first month misses?
- Cheapest and highest-leverage first. One, whoever answers the phone — free, immediate, and the most common culprit by a distance. Two, the sales conversation, once people are reaching a human. Three, the website, if people arrive and leave without asking for anything. Four, the offer and the price — slower, and the most powerful, because raising a price raises your maximum proportionally. Five, the advertising. It's last not because it doesn't matter, but because in a first month that refutes your calculation, the advertising is rarely what broke, and every week spent tuning campaigns against an economic gap is a week the actual problem goes unexamined.
- How do I know whether I have an optimization problem or a business problem?
- Find the current market range for clicks and compare it to your maximum. If competitors are comfortably paying about twice what you can afford, that isn't recklessness on their part. One competitor miscalculating is possible; several of them sustaining it is not a mistake — it's information. It means their close rate, their average job, their repeat business or their website is beating yours, possibly all four. And there's a mechanical consequence people miss: at half the market rate you don't get a smaller version of the campaign, you get a broken one, because the platform never gets enough conversions to learn from. Which is why the answer at that point is to stop, not to try harder.
- Should I judge my campaign against the forecast or against break-even?
- Against break-even, and confusing the two causes two opposite errors. Your assumption is a hypothesis; the threshold is the decision. In the case above we assumed 65%, rebuilt the sales team, and the close rate came back at just over 50% — below plan, and enough, because the economics closed at 50%. One owner hits 50% against a 65% forecast and shuts down a working business because he missed target. Another hits 90% of a forecast that was below threshold to begin with, and celebrates. The number that matters isn't the one you assumed. It's the one where the economics close.