After an ice storm the phone beats the office, and a missed call is somebody else's roof
Demand here arrives in bursts around weather, and the constraint in those bursts is never marketing, it is who answers.
Ice and wind events put a county's worth of worried homeowners on the phone inside forty eight hours. The contractor who answers gets the work, and the one who returns calls on Thursday gets the leftovers.
Pull the phone records after the next event and count the unanswered calls. Almost nobody does this, and the number is usually the largest single leak in the business.
Build the overflow plan before the weather, not during it. Somebody on backup, a script for the office manager, and a clear rule about what gets promised while the ice is still on the roof.
After hours calls need a real path. An answering service is fine if it is briefed properly and takes an address, a roof age and a callback number. An answering service that says somebody will be in touch is a delay dressed up as coverage.
Speed matters most on the emergency calls and matters more than most contractors think on the ordinary ones. The first company to speak to a homeowner has an advantage that later, better companies have to overcome.
A Wilson County address and a Belle Meade address deserve different scheduling promises
Qualification is not screening people out, it is making sure the appointment you book is one you can keep well.
Metro Nashville and Davidson County are consolidated while Williamson, Rutherford, Sumner and Wilson are not, and your operational reality changes at the line as surely as the government does. Drive time, crew availability and what you are willing to send a truck for are all different in Mount Juliet than in Green Hills.
Ask the address first. Then roof age, what is happening, and whether it is actively leaking. Four questions, in that order, before anybody promises a day.
Group appointments by geography rather than by call order. An estimator who spends the morning crossing the county sees fewer roofs than one who spends it in one part of it, and estimators who see fewer roofs sell fewer roofs.
Give different promises honestly. Same day for an active leak nearby, later in the week for a distant inspection on a fifteen year old roof. A kept modest promise beats a broken generous one every time.
Track no shows and cancellations by county. If one area produces most of them, the problem is usually the promise made on the phone, not the homeowner.
Most proposals die quietly between Thursday and the following Tuesday
The proposal follow up is the cheapest unbuilt asset in almost every roofing company we look at.
A homeowner who takes three bids is not making a decision on the day. They are waiting for the third estimator, talking to a spouse, and looking at the number again on the weekend.
Nothing structured usually happens in that window. One call on day two, maybe, then nothing, and the job is recorded as lost without anybody knowing why.
Build a sequence and run it the same way every time: a same day summary while the visit is fresh, a call within two days, something useful rather than a nudge later in the week, and a final check that gives the homeowner an easy way to say no.
Useful means answering the question they did not ask in the driveway. What the tear off week actually looks like. How the decking gets assessed. What happens if it rains mid job. What the warranty covers and what it does not.
Record the reason for every loss in one field with a short list of options. Within a quarter you will know whether you are losing on price, on timing, or on the third bid being faster.
The insurance question gets asked on every call, and a vague answer ends it
After any storm the homeowner's first real question is whether insurance will pay, and how your office answers determines whether they stay on the phone.
Two failures are common. Overpromising, where somebody says the claim will be covered, which is a statement nobody in your office is in a position to make. And retreating into vagueness, which the homeowner hears as evasion.
The workable script explains the process without predicting the outcome. What an inspection generally looks at, what documentation usually helps, how deductibles generally work in principle, and a clear line that coverage decisions rest with the insurer and are worth confirming with them or with their own counsel.
Write it down and train it. Consistency here is worth more than eloquence, and an inconsistent answer across three staff members is a trust problem before it is a legal one.
The same discipline belongs in the follow up. Never send a written statement that a claim will be paid.
Homeowners who moved here recently often have no experience of a claim on this kind of weather at all. Explaining the process plainly to them is a differentiator, not a chore.
Count signed roofs per estimator week, because everything upstream lies
Leads, form fills and calls all inflate for reasons that have nothing to do with revenue, and only the signed job is real.
The chain worth measuring is short: inquiries received, inquiries reached, appointments booked, appointments kept, proposals issued, contracts signed, and the value of those contracts. Six numbers, tracked weekly.
Attach a source to every one of them so the marketing spend can be judged on contracts rather than on clicks. Without that, the cheapest source always looks best and the cheapest source is usually the worst.
Segment by county. A blended close rate across Davidson, Williamson and Rutherford hides the county that is quietly carrying the others and the one that is losing money on drive time.
Separate repair from replacement in every one of those numbers. They convert at different rates and different values, and blending them produces an average that describes no job you have ever done.
We will not quote you an industry benchmark for any of this. Benchmarks in this trade are mostly invented, and the only meaningful comparison is your own numbers last quarter.
You will never have the volume to prove a button color, so test the phone script
One metro does not generate enough inquiries to detect a small effect, which changes what is worth testing at all.
Split testing works when there is enough traffic to separate a real difference from noise. A roofing contractor covering Middle Tennessee generally does not have that, and a test that runs for months while conditions change is not a test.
Test the large, obvious things instead, one at a time, and judge them over a full season rather than a fortnight. Who answers the phone and what they say. Whether the estimate summary goes out the same day. Whether the form asks for roof age. Whether the emergency path exists at all.
Call recordings are worth more than any experiment. Listen to twenty calls that did not become appointments and the pattern is usually visible by the eighth.
Talk to the customers who chose you and, harder but more useful, a few who did not. Small numbers, direct answers, no statistics required.
Change one thing at a time and write down when you changed it. Half the confusion in this work comes from three changes landing in the same month.
Questions we actually get
- Where do roofing inquiries usually leak first?
- At the phone, and it is rarely close. Unanswered calls during busy weeks, slow callbacks, and after hours inquiries that get nothing more than a promise of contact. Pull the call records after your next weather event and count. The second largest leak is the proposal that goes out and then gets no structured follow up.
- How fast do we really need to respond to an inquiry?
- Faster than you currently do, and fastest on active leaks. We will not quote you a magic number of minutes, because the studies people cite for that are mostly unverifiable. The practical standard is that a roofing inquiry has a short shelf life and the first company to have a real conversation carries an advantage the later ones have to overcome.
- Is split testing worth doing for a single market roofing company?
- Rarely, in the classic sense. One metro does not send enough traffic to prove a small difference, and a test that runs long enough to reach significance has been contaminated by seasons and storms. Test large changes one at a time over a full season, and spend the effort you saved on call recordings and customer conversations instead.
- What should we track if not cost per lead?
- Inquiries received, inquiries reached, appointments booked, appointments kept, proposals issued, contracts signed and contract value, with a source attached to each and the whole thing segmented by county and by repair versus replacement. Cost per lead rewards whichever channel produces the most people who will never buy.
- How should our office answer questions about insurance coverage?
- With a written, trained script that explains the process without predicting the result. Describe what an inspection generally looks at and what documentation usually helps, and state clearly that coverage decisions rest with the insurer and are worth confirming with them or with the homeowner's own counsel. Never say a claim will be paid, in a call or in writing.
