Which side of the river is a required field, not a nice-to-have
If your CRM does not record state and county on every inquiry, every number downstream is a guess.
Jefferson County, Oldham County, Bullitt County and the Southern Indiana counties across the river behave differently on close rate, drive time and job size. A metro average hides all of it.
Routing depends on it too. Which estimator, which day, what you can promise on scheduling. A dispatcher guessing from a phone number is guessing.
Put the address question immediately after the name in the intake script. The address settles the state, the county, the drive and half the qualification.
If you are not set up to work Indiana, say so within the first thirty seconds and refer the caller to somebody who is. That call costs you a minute. A canceled appointment in Clarksville costs a half day and a review.
A 6 a.m. missed call is a shift worker who never calls back
Missed calls, not weak websites, are where most roofing money quietly leaves this market.
The freight and logistics base here means people are awake and dealing with their houses at hours an office does not cover. Those callers are not patient, and they have three more numbers on the results page.
Pull the last ninety days of call data before you argue about it. Count missed calls, count after-hours calls, count average time to answer, and look at the hour-by-hour spread.
Fix in order. Somebody answers during business hours, always. Then an answering service with a script that captures the address and the water question. Then automatic text-back on every missed call.
An automatic text within a minute costs almost nothing and gives the caller a reason not to dial the next roofer while they wait.
Leave a number at the house or leave the door open for the next roofer
The gap between the inspection and a proposal that arrives days later is where the job goes to the second bidder.
Homeowners collect three bids. Whoever puts a written scope and a number in their hands on the day becomes the benchmark the others are measured against.
Equip the estimator to do it: measurements on a tablet, pricing built on site, options written out with what each one includes and excludes.
When something is genuinely unknown, such as decking under an old covering, say what is conditional, give the range, and state how the change would be priced. Vagueness reads as a setup for a change order.
Track same-day proposals as a rate per estimator and coach on it. It is one of the few numbers you can move in a month.
Between the second bid and the third, stop sending check-ins
Follow-up that repeats just checking in teaches the homeowner to stop opening your messages.
Send something that helps them compare bids: what a scope should specify, why ventilation and decking matter, what a low bid usually leaves out. You are arming them for a conversation with your competitors, and that is the point.
Send the photos from their own roof. Nobody else will, and it is the single most persuasive thing you have.
A workable cadence is day one, day three, day seven, day fourteen, then monthly through the season. Stop when they ask you to.
Stay neutral on insurance. Describe your process, never predict coverage, and point them to their insurer or their own counsel on anything about the policy itself.
A six week backlog quietly cancels the jobs you already paid for
Contracts that wait months for a crew fall apart, and marketing usually takes the blame.
Winter events and the spring rush stack the schedule fast. A homeowner who signed in a panic and then heard nothing for five weeks starts calling around again.
Give a real date or a real window at signature, and update the customer when it moves. Silence is what kills it, not the wait.
A one-line note each week during the wait costs nothing and holds the job. Assign it to somebody by name or it will not happen.
Measure cancellation rate after signature, split by how long the wait was. If it climbs past a certain point, the fix is scheduling and communication, not another campaign.
One change a month, judged on signed roofs, is your whole test program
A single metro does not generate enough inquiries for a roofing company to prove a small difference, so test large things slowly.
Classic split testing needs volumes you will not see. Running one anyway produces a result that looks conclusive and is noise, and then somebody rebuilds a page around it.
Change one meaningful thing per month instead: the intake script, the form length, the offer, the proposal format, the follow-up sequence. Big levers, one at a time.
Judge on signed jobs over a matched period, and respect the season. December compared against June proves nothing about anything.
The best evidence in a market this size is not a dashboard. Listen to twenty call recordings and phone five homeowners who chose somebody else. They will tell you the answer in an afternoon.
Questions we actually get
- How fast do we need to call a new inquiry back?
- Faster than the roofer they contact next. We will not quote a study figure, because we do not have a verified one for this market. What holds up is simple: the first company to reach a homeowner with water coming in usually gets the inspection, and everyone else is arguing with a decision already half made.
- What should we measure instead of leads?
- Completed inspections, proposals delivered, signed contracts, and cost per signed contract, each split by state and county. Lead counts flatter the campaigns that generate the most form fills, which is rarely the same as the ones that fill the crew calendar.
- Can you run A/B tests on our site?
- Rarely with statistical confidence, and we will tell you when the volume does not support it. The honest approach in a market this size is sequential: one substantial change per month, judged on booked work over a comparable period, with the season accounted for.
- Should we use an answering service?
- It beats voicemail if the script is right. It needs to capture the address, whether water is entering, and a callback commitment, and it needs to know which side of the river you serve. A service that promises appointments you cannot keep is worse than the voicemail was.
- How do we handle homeowners waiting on an insurance decision?
- Stay useful and stay neutral. Keep the follow-up going with information about scope and materials, document what you observed, and let the insurer and the homeowner handle the claim. Do not predict an outcome, and point policy questions to their insurer or their own counsel.
