Separate the work
Build around the services, locations, and urgency that the business can serve profitably.
Service-area example: Roofers
Buy roofing businesses customers at a price you can check.
High-ticket work can absorb acquisition cost, but only if the estimate and booked-job path is visible.
- Month to month
- You own the account
- Spend cap and verdict date in writing
- Conservative attribution
The return math
What a roofing click is worth.
A roofer who pulls back on bids when clicks hit $30.00 may be making a mistake. If that click has a 10% chance of turning into a quote, and you close 30% of quotes, the math on a $12,000.00 job still works at $50.00 per click. With job values commonly running $8,000.00 to $25,000.00, the click price alone can never tell you whether the spend is working. Job value, quote rate, and close rate decide that, and those three numbers are set with you before launch. Figures illustrative; the division is the point.
Storm response is built before the storm: campaigns prepared and held in reserve, ready to launch when demand spikes instead of being assembled while competitors are already live. Insurance-claim work is its own positioning, because a homeowner with a damaged roof is searching for someone who can navigate the claim process, not for the cheapest quote.
The commercial reality
Campaign structure should follow the work.
Storm demand and longer sales cycles require disciplined follow-up.
Before more budget is committed, identify the service lines, location limits, response path, realistic capacity, and the job outcome that makes the spend worth it.
What the campaign needs
Control before scale.
Keep the handoff visible
Track calls and forms, then make it easy for the business to confirm which ones turned into booked work.
Reconcile monthly
Use the Ledger to decide whether to hold, improve, scale, or stop. Not a dashboard mood.
The Payback Window
A defined maximum downside and a dated verdict.
Every engagement gets a proving period with a spend cap and a verdict date, set in writing before launch. Proving periods are set per trade, because a drain-cleaning call and an insurance roof replacement do not pay back on the same calendar.
The most this can cost
Before launch, the all-in maximum for the proving period is written down: fee plus ad spend. Hoping has a budget and the budget has a ceiling.
Long enough to be fair
The window is matched to how your trade actually books and pays. Long enough for a true answer, short enough that a wrong bet stays small.
The day hoping ends
On the verdict date, the Ledger says continue, adjust, or stop. If the numbers say stop, I will say stop out loud. That is the deal.
Questions roofing businesses ask
The answers, before a sales call.
What should a roofing campaign measure after a lead arrives?
The quote and the close. Roofing sales cycles are long, so a lead count in week two proves little. The path from enquiry to estimate to signed job is what gets tracked, and the verdict date is set far enough out to judge it fairly.
How does storm demand get handled?
With campaigns built in advance and held in reserve. When a storm hits, the account launches prepared coverage the same day instead of spending the surge week building.
Are insurance-claim jobs worth targeting?
Often they are the best jobs in the book, and the ads should say what those homeowners need to hear: experience navigating the claim, not a discount. That positioning changes the ad copy, the landing page, and which searches are worth paying for.
“I count conservatively. Every untracked job is your upside, not my credit.”
The first decision
Set a bounded first bet.
Before any campaign launches, set the spend cap, proving period, and verdict date in writing.
Talk through the numbers
Free, 30 minutes, no obligation. I reply within 12 hours.