Every leak in a roofing business stays hidden for the same reason: nobody's job is to look. Here's what that costs and how to fix it without hiring.
It's February, and it's slow. For the first time in about eight months, the owner has a quiet afternoon.
So he sits down with the numbers. And he finds something.
Maybe it's that a marketing line he's been paying for since 2024 has never produced a traceable job. Maybe it's that the estimator he's happiest with is closing eleven points below the other one. Maybe it's four hundred estimates sitting open, some of them two years old.
Whatever it is, it's been happening the entire time. Not for a month. For years.
That's the leak underneath all the other leaks. Not one broken thing — the absence of anybody whose job includes noticing that things are broken.
What's missing is somebody who owns the numbers — a person responsible for asking whether the business is working and looking at the answer regularly. Not a department. Not software. A responsibility.
Here's the size of it. Every other problem in a roofing company has a duration attached to it, and duration is decided by how long it takes someone to notice. A leak caught in a month costs a month. The same leak caught in a slow February costs two years.
This article covers why these problems stay invisible, the questions most roofing owners can't answer about their own business, the handful of numbers that actually matter, and how to fix this without adding a salary.
Table of Contents
- Key takeaways
- Why roofing problems stay invisible
- Why having reports doesn't fix this
- The 8 questions most roofing owners can't answer
- Why the owner can't be the one watching
- This isn't a hire, at least not yet
- Which numbers actually matter for a roofing company?
- Why more software doesn't solve it
- What does having nobody watching cost?
- How to check whether anyone owns your numbers
- What does good look like at your size
- How this role fits into the bigger revenue picture
- Find your roofing revenue leaks
- Roofing business metrics FAQs
Key takeaways
- Most problems in a roofing company don't announce themselves. They stay hidden until somebody deliberately goes looking.
- The cost of a leak is mostly its duration, and duration is decided by whether anyone is watching.
- Having reports isn't the same as having someone who reads them and acts on what they say.
- The owner is the worst candidate for this job, because he's only free to look during the months when it's already too late.
- For most roofing companies, this starts as a recurring hour and a handful of numbers, not as a hire.
Why roofing problems stay invisible
The expensive problems in a roofing business don't generate complaints, errors, or alerts. They generate silence.
Think about what you'd actually notice. A crew no-show, you hear about. A furious customer, you hear about. A supplier shorting a delivery, you hear about.
Now think about what produces nothing at all:
- A phone number that stopped reaching anyone after hours
- A homeowner who searched, didn't see your company, and called somebody else
- An estimate delivered in March that nobody followed up on
- A marketing line item that hasn't produced a job in a year
- A crew losing ninety minutes a day to drive time
- Material prices that drifted past your estimating template
None of those rings a phone. Nobody files a complaint about a conversation that didn't happen. The business feels normal because, from the inside, it is normal — this is just how it's been.
That's what makes these problems different from the problems roofing owners are good at solving. An owner who can diagnose a leak from a ceiling stain has no equivalent instinct for a marketing channel that quietly died, because there's no stain.
Why having reports doesn't fix this
Most roofing companies have more reports than they use, and the gap between the two is where this leak lives.
Your accounting software produces reports. Your production software produces reports. Your ad platform emails you one every month. Somewhere there's a dashboard somebody set up two years ago.
They don't help, for three reasons.
They answer the platform's question, not yours. An ad report tells you cost per lead because that's what the platform can see. It can't tell you cost per sold job, which is the number you'd actually make a decision with.
Nobody reads them on a schedule. A report that's available is not a report that's been read. Most get opened when something already feels wrong.
They report what happened, not what's happening. Revenue last month is history. The estimate that went cold three weeks ago is still recoverable, and no standard report will surface it.
The useful version of this isn't more reporting. It's one person asking a small number of specific questions on a regular schedule and doing something about the answers.
The 8 questions most roofing owners can't answer
Not because they're bad owners. Because nothing in the business is set up to produce the answers.
Try these on your own company:
- How many leads came in last week, and from where?
- How long does it take, on average, for a person to contact a homeowner who reached out?
- How many estimates are open right now, and how many have a next step with a date?
- What's your close rate by lead source?
- Which marketing produced your last ten sold jobs?
- What did your most recent job actually cost, including materials pulled from the yard?
- How many productive hours did each crew get last week?
- How many customers left a review last month?
Most roofing owners can answer two or three of those confidently.
The point isn't that all eight need a dashboard. It's that each unanswered question is a place where something could be going wrong indefinitely without anyone finding out.
And notice what they have in common. None of them are hard questions. They're just questions nobody is assigned to ask.
Why the owner can't be the one watching
In most roofing companies, the owner is the estimator, the salesperson, the escalation path, and the final decision on everything. That's four jobs, and watching the numbers becomes the fifth.
Fifth jobs don't get done.
There's a worse version of the problem, though, and it's the one in the story at the top of this article. The owner does look at the numbers — in February, when it's slow enough to sit down.
Which means the business only gets examined during the months when there's least to examine, and problems from the busy season get discovered six months after they started costing money.
The months you most need someone watching are the months nobody has time to watch. Storm season, peak summer, the weeks when volume is highest, and every gap is multiplied by the most opportunities.
That's not a discipline failure. An owner on a roof at 2 pm on a Tuesday is doing the right thing with his afternoon. The problem is structural, and structural problems need a structural answer rather than a resolution to try harder.
This isn't a hire, at least not yet
For most roofing companies, owning the numbers starts as a recurring hour on somebody's calendar.
The instinct when you name a gap like this is to think about adding a person, and for a company doing 150 jobs a year, that's usually wrong. What's needed first is much smaller.
Pick a person — often the office manager, sometimes a spouse, sometimes the owner with an appointment he actually keeps. Pick a day. One hour, same time every week or every month.
In that hour, they answer the same handful of questions and write the answers down somewhere they can be compared to last time. That's it. No analysis, no dashboards, no software project.
The value isn't in the sophistication. It's in the repetition, because that's what turns a number into a trend, and a trend is what makes a problem visible before it's a year old.
The person doesn't need to be analytical by nature. They need to be consistent, and they need permission to bring an owner bad news.
It becomes a real role later, if the company grows to where the coordinating is a fractional or full-time job. Most roofing companies get most of the benefit long before that.
Which numbers actually matter for a roofing company?
Six, and you can stop there.
The temptation with measurement is to build something comprehensive. Resist it. A dashboard with thirty numbers gets ignored exactly as thoroughly as no dashboard at all.
-
Leads by source, weekly. Where the work is coming from, counted the same way every week.
-
Time to first human contact. How long a homeowner waits before a person actually speaks to them.
-
Open estimates, and how many have a next step. The second half is the one that matters.
-
Close rate by source. Not lead volume by source. Which channels produce sold jobs.
-
Productive hours per crew per day. The cleanest measure of whether scheduling is working.
-
Material cost as a percentage of job revenue. When this drifts, something upstream changed, and nobody told you.
Six numbers, checked on a schedule, written down so this month can be compared to last month.
What makes them work isn't precision. It's consistency. A roughly-right number tracked the same way for six months will show you a problem. A perfect number calculated once will not.
Why more software doesn't solve it
Most roofing companies with this problem already have plenty of data. What they don't have is anyone turning it into a decision.
This is worth saying plainly, because the usual response to "we can't see what's happening" is to buy something.
Software makes the data available. It doesn't decide which questions matter, notice when a number moved, or walk into the owner's office and say the thing nobody wants to hear about the estimator's close rate.
There's also a practical trap. A company that buys a system without anyone owning the numbers usually ends up with the same blind spots plus a subscription. The reports exist. Nobody reads them. Six months later, the conclusion is that the software didn't work.
The sequence that works is the opposite. Decide what you need to know. Start answering it however you can, even badly, even on paper. When the manual version becomes the bottleneck, buy something that removes that specific bottleneck.
Software is very good at making a working process faster. It's very bad at inventing one.
What does having nobody watching cost?
Everything the other leaks cost, multiplied by how long they run.
This one doesn't have a cost of its own. It sets the duration on all the others, which makes it the most expensive item on the list.
These are illustrative numbers. Use your own.
A paid phone number that stops working costs nothing on day one. Over three weeks, at $6,000 a month in ads, it's around 22 calls that reached nobody. Caught in a week, it's a nuisance. Caught in February, it's a season.
An estimator closing eleven points below his colleague costs a few thousand dollars a month. Nobody compares their numbers, so it runs for two years before anyone notices, and by then it isn't a coaching conversation anymore.
Material costs drifting 8% past your estimating template costs a few hundred dollars a job. Over 150 jobs, that's real money, and the books will show it as a soft year rather than as a fixable problem.
Each of those is survivable on its own. The reason they add up to something serious is that none of them get found early, and none of them get found early because finding things isn't anybody's job.
The fix costs an hour a week.
How to check whether anyone owns your numbers
This one takes about 10 minutes and the answers tend to be uncomfortable.
Test 1: Name the person
Whose job is it to notice when something in this business stops working?
If the answer is "mine, I guess," ask the follow-up: when did you last do it, and was it on a calendar or because something felt wrong?
Test 2: Answer the eight questions
Go back to the list earlier in this article and answer as many as you can without looking anything up or calling anyone.
Count them. That's your current visibility.
Test 3: Find the last thing you caught late
Think of a problem you discovered that had been going on a while. A marketing spend that wasn't working, a process nobody was following, a number that had drifted.
How long had it been happening before you found it? What would have surfaced it sooner?
That's your detection time, and it applies to everything you haven't found yet.
Test 4: Check when you last looked
Look at your own calendar for the last three months. Is there a recurring block for reviewing the business?
Not a meeting about this week's jobs — time to look at whether the business is working.
Test 5: Ask your office manager what worries her
Whoever handles your scheduling and paperwork sees more of the business than anyone except you. Ask what she notices that nobody ever does anything about.
The list will be specific, and most of it will be things she's mentioned before.
Found a visibility gap?
The free Roofing Revenue Leak Check looks for the other places where problems may be hiding because nobody can see the numbers soon enough.
What does good look like at your size
Somebody specific looks at a small set of numbers on a schedule and has permission to say when something's wrong.
One to three crews
One hour a month, one person, six numbers, written down where last month's can be seen next to this month's.
That person is usually the office manager. The owner's job is to make the hour real — on the calendar, protected, and not the first thing cancelled during a busy week.
The habit matters more than the rigor. Six roughly-right numbers, monthly, for a year, will catch nearly everything in this series.
Four or more crews
At this size, the coordination itself becomes a job. More people, more handoffs, more simultaneous work, and more places for something to go quiet.
This is usually where a company either promotes someone into an operations role or brings in outside help on a recurring basis. The work is the same, just more of it: keeping the numbers current, spotting the drift, and making sure decisions get made on evidence rather than on whoever spoke last.
It's also where the systems start to matter, because manual collection stops being viable and the bottleneck moves from "nobody's looking" to "looking takes too long."
Either way
Two things have to be true, regardless of size. The looking happens on a schedule rather than when something feels wrong. And the person doing it can tell the owner something he doesn't want to hear without it being a problem.
The second one is where most of these efforts actually fail.
How this role fits into the bigger revenue picture
This is the last stage in the process and the one that determines whether any of the others get fixed.
Lead source → Lead capture → Response → Qualification → Appointment → Estimate → Follow-up → Sale → Measurement
Every leak in this series lives at one or two of those stages. This one lives at the end, and it's the only one that makes the others visible.
That's the thing worth sitting with. You could fix every other problem described in this series and, without somebody watching, they'd come back. Numbers change, staff turn over, a campaign gets renamed, a form gets rebuilt, a process quietly stops being followed.
A roofing company without anyone owning the numbers doesn't have a permanent problem with any one thing. It has a permanent condition where problems run until somebody stumbles across them.
This is the last of fifteen revenue leaks that waste good roofing leads, and it's the one that keeps the other fourteen alive.
Find your roofing revenue leaks
You may not need more leads. You may need somebody to actually look.
The free Roofing Revenue Leak Check is one outside pass across your lead-to-revenue process — from how homeowners find and contact you through response, appointments, estimates, follow-up, sales, and measurement.
We start with one short call. Then I look at what's public and, where it helps, what's happening inside your process. You get a plain-English list of what may be leaking, roughly what it may be costing you, and what I'd fix first. If nothing's leaking, you'll know that for certain instead of guessing. No pitch buried inside it.
Check your roofing revenue leaks →
Roofing business metrics FAQs
What numbers should a roofing company track?
Six are enough: leads by source, time to first human contact, open estimates and how many have a next step, close rate by source, productive hours per crew per day, and material cost as a percentage of job revenue. Tracked consistently, those will surface nearly every common problem. A larger dashboard gets ignored as completely as no dashboard.
Who should be responsible for reviewing a roofing company's numbers?
Someone other than the owner, in most cases, because the owner is estimating, selling, and handling escalations, and only has time to look during slow months. It's often the office manager. What matters more than analytical skill is consistency and permission to deliver bad news.
How often should a roofing company review its numbers?
Monthly is enough for most companies with a few crews, weekly once there are several crews running simultaneously. The schedule matters more than the frequency, because reviewing when something already feels wrong means you only find problems after they've been expensive for a while.
Do I need software to track roofing business metrics?
Not to start. Most companies with visibility problems already have plenty of data, but nobody turning it into decisions. Start by answering a small set of questions however you can, even on paper, and buy something when the manual version becomes the bottleneck. Software makes a working process faster; it doesn't create one.
When does a roofing company need an operations role?
Usually, when coordination itself becomes a full-time job — several crews running at once, multiple people handing work between them, and enough simultaneous activity that nobody can hold it in their head. Before that point, an hour a month on somebody's calendar delivers most of the benefit.
Comments