If you run paid search for a roofing company, the ground moved under you this month.
Google is retiring the standalone Local Services Ads product and folding it into Performance Max with a pay-per-lead goal. Roofing is named explicitly in the first wave of US categories, alongside plumbing, HVAC, electrical, appliance repair, house cleaning, lawn care, pest control, and moving. That rollout started in August 2026. Broader advertiser groups follow later this year, and non-US accounts in 2027.
So the honest version of this article is no longer “should roofers use PPC.” Of course they should — it is the only channel where you reach a homeowner in the twenty minutes between finding the stain on the ceiling and calling someone. The real question in 2026 is who should be running it, and against what numbers.
What actually changes with the LSA migration
If you have been running LSAs, most of what you knew about managing them is about to stop being true.
Your budget math changes. Weekly budgets convert to daily averages — your historical weekly number divided by seven. Monthly spend is then capped at that daily average multiplied by 30.4. If you were deliberately front-loading budget into storm weeks, that lever works differently now.
Manual bidding goes away. Manual bidding and vertical-level Target CPA are being deprecated. If your account was built on hand-set bids by service line, that structure does not survive the move.
Your reporting history does not come with you. This is the one that will actually hurt people. Historical LSA performance reports do not transition. Google’s guidance is to download them from the redirection page before your account migrates. Past customer leads and message history do carry over, and the Google Business Profile sync continues, but your performance history — the thing you need to know whether this year is better than last year — is yours to save or lose.
Some features are simply gone. Better Business Bureau callouts are no longer supported.
You get 14 days’ notice. Google sends an email two weeks before your account migrates, and campaign setup transfers automatically.
If you do nothing else after reading this article, do this: screenshot or export your LSA performance reports now. Once the migration runs, that baseline is unrecoverable, and you will spend the next year unable to answer whether the change helped or hurt you.
What roofing PPC actually costs in 2026
Cost-per-lead numbers get thrown around in this industry with no sample behind them, so here is one with its work shown.
A Q1 2026 analysis of 15 US roofing contractors, $310,000 in non-branded Google Ads spend, and 2,491 tracked leads found:
Campaign type
Average cost per lead
Non-branded search
$124
Branded search
$44
Performance Max
$64
The spread on non-branded matters more than the average: 25th percentile $80, median $125, 75th percentile $256. That is a 3x gap between a well-run account and a poorly-run one on the same service in the same country. The difference is not budget. It is structure, negative keywords, and landing pages.
Seasonality is real and it is large. In that same sample, non-branded CPL fell 23% from January ($145) to March ($111) while lead volume rose 52%. Demand and efficiency improved together going into spring.
Two caveats I would want if I were reading this: fifteen contractors is a small sample, and storm-driven markets with heavy insurance-claim volume behave nothing like markets where replacement is elective. Use these as a sanity check on your own numbers, not as a target.
The number that actually decides this: if roughly 15% of your leads close, a $124 lead is a $827 customer acquisition cost. Whether that is good or catastrophic depends entirely on your average job value and gross margin — not on whether $124 sounds expensive. Run that math before you judge any agency’s performance, including mine.
Why DIY roofing PPC usually costs more than it saves
The mechanics look simple: pick keywords, write an ad, set a budget. The reason it goes wrong is that roofing is one of the most expensive, most competitively bid categories in local search, and the errors are expensive and invisible.
Broad keywords bleed money quietly. Bidding on “roof” or “roofing” without tight match-type control buys you clicks from people writing school papers, researching DIY shingle replacement, and shopping for materials. You do not see the waste; you see a high CPL and assume the channel is expensive.
Negative keywords are the whole game. “DIY roof repair,” “free roofing materials,” “roofing jobs,” “roofing supplies,” “how to.” Every one of those is a click you pay for and can never close. A serious negative list runs to hundreds of terms and gets pruned monthly. This single discipline explains most of the gap between the 25th and 75th percentile above.
Performance Max hides where the money goes. With LSAs folding into PMax, more of your spend now sits inside an automated campaign type that reports at a higher level of abstraction than search campaigns did. Getting useful signal out of it requires deliberate asset-group structure and conversion-value inputs. Set up carelessly, PMax will happily spend your budget on the cheapest conversions it can find, which are rarely roof replacements.
The landing page is usually the actual problem. Most roofing accounts I have audited were not losing on bid strategy. They were sending high-intent clicks to a homepage. If the click lands anywhere other than a page about the specific service the person searched for, with a phone number visible without scrolling and a form under six fields, the ad spend was already wasted upstream.
It is not set-and-forget. Bids, budgets, search terms, and creative all need monthly attention minimum, weekly during storm season. That is real hours, and they are hours you are not spending on crews, sales, or collections.
What it costs to hire someone — and what you are actually buying
Market rates as of 2026:
•
Freelance PPC specialists: roughly $750–$2,500/month in management fees, depending on account size and whether they specialize in home services.
•
PPC agencies: typically $1,500–$5,000/month, or 10–20% of ad spend. Percentage-of-spend is worth scrutinizing — it pays your vendor more when your costs go up.
•
Hourly consultants: $100–$250/hour for audits, one-time builds, or fixing someone else’s account.
•
One-time setup fees: $500–$3,000, covering keyword research, campaign build, conversion tracking, and landing page work.
Here is the part most agencies will not tell you: a PPC manager optimizes the campaign. They do not own whether the lead becomes a job.
If your CPL is fine but your close rate is 8%, PPC management is not your problem. Your intake is. Nobody answers the phone in under sixty seconds, nobody follows up on a form fill within five minutes, and nobody is scoring which lead sources actually produce signed contracts. Buying better campaign management to fix that is like buying a better faucet to fix a hole in the bucket.
How to decide who runs it
A short decision tree.
Hire a PPC specialist or agency if: your campaigns are structurally sound, your intake and follow-up already work, and you need someone to run the account week to week. This is an execution problem and you should buy execution.
Bring in a fractional marketing leader if: you do not know your cost per acquired customer, your lead sources are not attributed to closed revenue, you have never had a marketing strategy that survived contact with a slow month, or you are about to hand a five-figure monthly budget to a vendor and have no way to evaluate their work. This is a leadership problem, and buying more execution will not solve it.
Do it yourself if: your budget is genuinely small, you enjoy this work, and you have the discipline to look at search terms every week. Some owners do run good accounts. They are rarer than the internet suggests.
The migration to Performance Max sharpens this. As Google removes manual levers, the value of a person who can hand-tune bids goes down, and the value of a person who can define what a good lead is worth, feed that back into the platform, and hold the whole funnel accountable goes up.
Where I fit
I am a fractional CMO, not a PPC agency. I run paid search when it is the right lever, but I am generally brought in earlier than that — when an owner has been spending on marketing for a while and cannot tell what it bought.
For home services companies that usually means: getting attribution honest, finding out what a signed job actually costs to acquire by source, fixing the intake gap between lead and appointment, and then deciding whether the media plan needs changing at all. Often it does not. Often the campaigns are fine and the follow-up is broken.
If your LSA account is migrating in the next few weeks and you want a second set of eyes on the transition before it runs — or you want someone to tell you honestly whether your current PPC spend is working — that is a conversation worth having.
Sources: Google Ads Help — Local Services Ads transition to Performance Max, Searchlight Digital 2026 roofing CPL benchmarks. Rate ranges reflect observed 2026 US market pricing. Last reviewed August 2026.