Roofing companies sell the same product in two completely different markets: the planned residential replacement a homeowner researches for weeks, and the storm job that did not exist until the hail hit. Running both through one Google Ads strategy wastes money in both directions, because everything that makes retail ads work (patience, financing, comparison-proof landing pages) is wrong for storm, and everything that makes storm work (speed, surge budgets, insurance fluency) is wasted on retail.
Here is how to run each lane on its own terms, and how to measure an account where the average job runs $8,000 to $40,000.
The two businesses inside a roofing account
Residential retail is considered purchase behavior. The homeowner knows the roof is old, gets two to four bids, and decides over 2 to 8 weeks. Tickets run $8,000 to $25,000, demand is roughly steady, and the winner is usually the company that handled the estimate process best, not the one with the cleverest ad.
Storm restoration, the insurance restoration side of the business, is event-driven. Demand appears overnight in specific zip codes, out-of-town stormers flood the auction within days, insurance rules the conversation, and tickets run $10,000 to $40,000 and up. The window is weeks, not quarters.
Different buyers, different tempo, different economics: they need separate campaigns, separate budgets, and separate landing pages, always.
The retail playbook
- Campaign lanes: brand, replacement intent (“roof replacement cost”, “roofing companies near me”), repair intent, and specialty materials (metal roofing, tile) if you sell them. Repair converts cheaper than replacement; give replacement its own protected budget or repairs will eat the account. The lane discipline is the same one the HVAC and plumbing structure guides walk through; roofing just carries a wider gap between the cheap lane and the one that pays.
- Landing pages sell the estimate, not the roof. Financing options, what the inspection covers, real project photos, and proof you will still exist in ten years. The 2-to-8-week sales cycle means your follow-up and lead handling matter as much as the click.
- Bid with the sales cycle in mind. A conversion today closes next month. Judge campaigns on closed-deal revenue by cohort, not this week’s cost per lead, or you will kill campaigns that were quietly winning.
The storm playbook
- Pre-build, do not scramble. Storm campaigns should exist before the storm: geo-targeted campaign shells, insurance-claim landing pages, and ad copy ready to activate. The companies that win the first 72 hours built their campaigns in the calm.
- Surge the budget where the hail fell. Tight geo-targeting on affected zips, budgets that follow the damage maps, and bids that respect what a $30,000 insurance job is worth. This is the one context where aggressive spend on day one is the conservative play.
- Sell the inspection and the insurance fluency. The homeowner’s real question is “will insurance cover this and who do I trust to handle the claim?” Copy that answers that beats copy about quality craftsmanship.
- Expect junk and filter it. Storm auctions fill with tire-kickers and adjuster-chasers. Qualified-call grading and quick disqualification keep the crews pointed at real claims, and call tracking with outcome tagging turns those grades into data the bidding can learn from.
Running both lanes and not sure the budget split is right? The free audit maps your spend against your closed jobs by lane and quantifies what the mismatch costs. Request a Free Audit. Five business days. No cost. No commitment.
Measurement is harder in roofing, and worth more
Roofing’s long cycle and insurance mediation mean the platforms almost never see the real outcome on their own. A form fill in March becomes a signed contract in April and a paid invoice in June. If your bidding optimizes on form fills, it is learning from the weakest signal in the chain. Most roofing accounts are not even at that stage: in our August 2026 scan of 1,000 contractor websites, only 38 percent of roofing sites were wired to measure ads at all, the best of the four trades and still fewer than two in five.
The fix is routing closed-deal revenue from your CRM back to the ad platforms: JobNimbus and AccuLynx both support this, and it changes what the account chases. With deal values flowing, the retail lane learns which searches sign $25,000 contracts instead of which ones fill forms, and the storm lane can prove it earned its surge budget. As a working rule, value-based bidding stabilizes around 15 valued conversions in a rolling 30 days; storm lanes usually clear that during events, retail lanes may need conversions-based bidding between.
Frequently asked questions
What should a roofing company spend on Google Ads?
Typical range: $5,000 to $50,000 a month, with storm-active periods justifying the top of the range. At $10,000-plus average tickets, a single incremental job a month pays for most of these budgets; the lead value guide runs that break-even math on a $16,500 replacement. Published roofing cost per lead sits around $228, the highest of the trades with a citable figure; the benchmarks page carries the source and sample size, and treats it as a range check, not a target.
Should we run storm campaigns if we are primarily retail?
Only with the operational capacity to serve storm work: insurance paperwork, supplement fluency, crew surge. Storm leads without storm operations produce refunds and reviews you do not want. This is a strategy question before it is an ads question, and we will say so on a discovery call.
How do we compete with out-of-town stormers in the auction?
You mostly should not try to outbid them; you out-credential them. Local address, local reviews, “still here next year” messaging, and pre-built campaigns that were live before they arrived. Stormers rent the auction; you own the market.
Do Local Services Ads work for roofing?
Yes, especially for repair and retail replacement intent, and the Google Verified badge is a trust signal out-of-town stormers cannot easily get. The LSA versus Google Ads breakdown covers how the two share the page, and the migration now moving LSA into the Google Ads interface, with roofing in the first wave.
Retail roofing and storm restoration reward opposite instincts, and the accounts that treat them as one business do both badly. Separate the playbooks, pre-build the storm response, and route closed-deal revenue back so the bidding learns from contracts instead of form fills. For the roofing-specific version of everything we do, start there, or request the free audit: a 20-minute discovery call, then five business days to a quantified answer.