Are Google Ads Worth It for Roofers? The Honest Math, Margins, and Math-Behind-the-Math

Every roofing owner who has ever logged into Google Ads has felt the same two things in the same week: a flush of hope when the first call comes in, and a sinking feeling when the monthly invoice lands and the math does not obviously work. You spent $4,200. You closed two jobs. Were those two jobs because of the ads, or would they have called anyway? Did the $4,200 buy you margin, or did it buy you the privilege of competing with three other roofers for the same impatient homeowner who was going to pick whoever answered the phone first?

The honest answer to "are Google Ads worth it for roofers" is the answer no agency wants to give you, because it does not fit on a pitch deck: it depends entirely on numbers you probably are not tracking yet, and for a large share of roofing companies, the honest answer is not the way you are currently running them. That is not the same as "never." Plenty of roofers print money with paid search. But they win on a small set of fundamentals — close rate, average ticket, geographic density, follow-up speed, and lead-source attribution — and they lose on the exact same fundamentals when those are weak.

What follows is the math a working contractor actually needs: how to calculate whether paid search pays, the channels inside "Google Ads" that behave nothing alike, the fraud and waste that quietly eats 15-30% of a roofing budget, the seasonality that turns a winning account into a losing one for four months a year, and the situations where your dollar is far better spent somewhere else entirely. No hype, no fake case-study numbers. Just the operating reality.

The one calculation that decides everything

Before you argue about keywords, headlines, or which agency to hire, you need one number: your maximum profitable cost per acquired job. Almost no roofer can tell you theirs off the top of their head, which is exactly why so many run ads on vibes.

Here is the chain. Work it for your own shop with your own figures.

1. Average job value (revenue). Not your dream job. The blended average of what actually closes off paid search. Retail residential reroofs and small repairs skew this down hard. Say your blended paid-search job is $11,000.
2. Gross margin on that job. After materials, labor, dumpster, permits, and crew. A healthy residential reroof runs 35-50% gross; a lot of competitive retail runs thinner. Use 40% → $4,400 gross profit per job.
3. The slice of gross profit you are willing to spend to get the job. This is a business decision, not a marketing one. Many roofers will spend 20-30% of gross profit on customer acquisition and still be healthy. At 25% of $4,400, that is $1,100 you can spend to win one job and stay in the green.
4. Your close rate on paid-search leads. This is where dreams die. Paid-search roofing leads close at a lower rate than referrals or door-knocked storm work — these are shoppers, often price-shoppers, often calling three companies. A realistic paid-search close rate is 20-35% for residential, sometimes lower if your follow-up is slow. Use 25% → it takes 4 leads to make 1 job.
5. Therefore your maximum profitable cost per lead is $1,100 ÷ 4 = $275 per lead.

Now hold that $275 against reality. In competitive metros, a single click on a high-intent roofing search term — "roof replacement near me" or "emergency roof repair" — can cost $25 to $75+. If your landing page and call handling convert clicks to leads at 8-12% (a generous range for cold paid traffic), you are paying for 8 to 12 clicks per lead. At $40/click and a 10% conversion rate, that is $400 per lead — well north of your $275 ceiling. You are losing money on every job and making it up in volume, which is the oldest joke in business.

That single comparison — max profitable CPL versus actual CPL — is the entire question. Everything else in paid search is about moving one of those two numbers until they cross in your favor.

A worked example, both directions

Shop A — paid search works. Average paid job $14,000, gross margin 45% ($6,300), willing to spend 25% of gross ($1,575) to acquire, close rate 33% (3 leads per job) → max CPL $525. Their actual CPL, with a tuned account and fast call answering, is $310. They make roughly $215 of headroom per lead, times hundreds of leads a year. Paid search is a profit engine for them.

Shop B — paid search bleeds. Average paid job $8,500 (lots of repairs and small retail), gross margin 35% ($2,975), willing to spend 25% ($744), close rate 22% (4.5 leads per job) → max CPL $165. Their actual CPL in the same metro is $290. Every job acquired through ads costs them $125 in pure overspend before they have paid a single salesperson. They feel "busy" and wonder why cash is tight.

Same city, same Google, same keywords — opposite outcome. The difference is not the ads. The difference is ticket size, margin, and close rate. Fix those and the ads follow; ignore them and no agency on earth can save the account.

"Google Ads" is three different products wearing one logo

A huge source of confusion is treating Google Ads as one thing. For roofers, it is at least three distinct products with different economics, different fraud profiles, and different reasons to use or avoid them.

1. Local Services Ads (LSA) — pay per lead, sits above everything

Local Services Ads are the green-checkmark Google Guaranteed units that appear at the very top of local-intent searches, above the regular text ads. You pay per lead (a call or message), not per click, and you only get charged for leads that are arguably relevant. Google also runs the Google Guaranteed verification — license and insurance checks, background checks in many categories — and ranks you partly on proximity, responsiveness, and review profile.

For most residential roofers, LSA is the first dollar you should spend in paid search, and often the only paid-search dollar worth spending early. Reasons:

• You pay for leads, not clicks, so you are insulated from the click-cost arms race that makes regular Search brutal.
• You can dispute bad leads. Spam, wrong service area, out-of-scope (someone wanting a metal carport, a tenant who can't authorize work, a robocall) — you flag them and Google credits a meaningful share if your disputes are legitimate and well-documented.
• The Guaranteed badge is real trust. Homeowners click it.
• Ranking rewards behavior you control — answer the phone fast, gather reviews, keep your service area tight — rather than just out-bidding everyone.

The catch: LSA rewards speed and reviews ruthlessly. Miss calls and your rank and lead quality both fall. Let your review velocity stall and you slide under competitors. And in some markets LSA lead prices have climbed steeply as more roofers pile in. But pound for pound, LSA is the highest-floor, lowest-ceiling product — hard to lose your shirt, harder to scale to dominance.

2. Search (PPC) — pay per click, the classic, the dangerous one

This is what people usually mean by "Google Ads": text ads on the results page, bid per keyword, charged per click whether or not that click ever calls you. This is the product with the seductive upside and the genuine downside.

Search can absolutely win — it is how Shop A above prints money — but it demands real competence: tight keyword and negative-keyword lists, call tracking, fast landing pages, geographic precision, dayparting, and relentless waste-cutting. Run it loose and it is the single fastest way to set money on fire in roofing marketing. Run it tight and it scales past what LSA alone can deliver. We will spend most of the rest on doing it right, because this is where the "worth it" question is genuinely live.

3. Performance Max & Display — usually a trap for roofers (with one exception)

Google pushes Performance Max (PMax) hard, because it hands Google's algorithm the keys and spends your budget across Search, Display, YouTube, Gmail, and Maps with little visibility into where. For a local service business with a finite service area and high lead value, PMax frequently spends a chunk of budget on cheap, low-intent Display and YouTube impressions that generate clicks and "conversions" that never become real jobs. The reporting looks great. The phone does not ring with closers.

The one defensible use: a roofer with a strong, clean conversion signal (real booked-job data fed back to Google, rather than only form fills) and enough volume to let the algorithm learn, who uses PMax as a supplement to a controlled Search campaign — not a replacement. For most owners reading this, start with LSA, add disciplined Search, and treat PMax with deep skepticism until your data is clean enough to trust an algorithm with your money. Standalone Display ads (banner ads on random websites) are almost never worth a roofer's money for lead generation; their place is brand recall, which is not what you are buying ads to do.

The seasonality nobody prices in

Roofing demand is not flat, and paid-search economics swing with it harder than owners expect. Two patterns dominate.

The replacement/repair cycle. Steady-state "my roof is old / leaking / failing" searches run year-round but peak in spring and fall when homeowners tackle projects and weather is workable. Winter (in cold climates) and the dead of summer often see softer high-intent volume, so your same daily budget buys fewer good leads and your CPL drifts up. An account that pencils out in April can quietly run underwater in January if you do not throttle it.

The storm spike. After a significant hail or wind event, search volume for roof repair and inspection explodes — and so does competition and click cost, because every roofer in a 200-mile radius (including out-of-state storm chasers) floods the same keywords at the same time. You can pay triple your normal click cost for the same lead during a post-storm surge. Worse, a lot of post-storm search traffic is tire-kicking — homeowners checking whether they have damage, not committed buyers — so conversion quality can drop exactly when costs peak.

The operational implication: paid search should not be your storm-response strategy. When a storm hits, the roofers who win are not the ones bidding hardest on "hail damage roof" against forty competitors — they are the ones who already know which neighborhoods and which specific roofs the storm most likely damaged and are knocking those doors, mailing those lists, and booking inspections while everyone else is fighting over the same overpriced clicks. Paid search after a storm is the expensive, crowded lane. Targeted outreach to the actually-affected roofs is the cheap, uncrowded one. (More on closing that data gap below.)

Where roofing budgets actually leak

Most roofers who conclude "Google Ads don't work" were not beaten by Google. They were beaten by waste inside their own account. The big leaks, in rough order of how much they cost:

Click fraud and junk traffic

Roofing is a high-cost-per-click category, which makes it a target. Three flavors of waste:

• Competitor click-draining. A rival (or their agency, or a bot they hired) clicks your ad repeatedly to burn your daily budget so their own ad wins the rest of the day. In a $50-per-click category, a few dozen malicious clicks is real money.
• Bot and scraper traffic. Automated clicks from data center IPs and click farms that will never convert.
• Mis-intent clicks. Not fraud exactly, but pure waste: people searching "roofing jobs" (looking for employment), "roofing materials wholesale," "how to roof a shed," "roofing companies" as students doing research. Without aggressive negative keywords, you pay for all of them.

The Federal Trade Commission and Google both treat invalid-click filtering as Google's responsibility, and Google does refund some invalid clicks automatically — but the automatic filter misses plenty. Estimates for wasted spend in competitive local PPC categories commonly land in the 15-30% range once you account for fraud plus mis-intent. Mitigation: a tight negative-keyword list (build it weekly from your search-terms report), IP exclusions for repeat offenders, and for higher-spend accounts a third-party click-fraud filter. If you are spending under ~$2,000/month, the negative-keyword discipline alone is most of the win.

Sending paid clicks to a slow, generic page

You paid $45 for a click. It lands on your homepage, which takes six seconds to load on mobile, buries the phone number, and talks about your company instead of the visitor's leaking roof. Conversion rate craters and your effective CPL doubles. Paid traffic needs a dedicated landing page: phone number tappable above the fold, a single clear action, proof (reviews, license/insurance, Google Guaranteed badge), a short form, and a sub-three-second mobile load. The page is not a detail; it is half the math. A page that converts at 12% instead of 6% literally cuts your cost per lead in half on identical ad spend.

Slow or missed call answering

The ugliest leak of all because you paid for the lead and then dropped it. Industry call-tracking data across home services repeatedly shows a large share of inbound calls to contractors go unanswered or to voicemail, and a meaningful slice of paid-search calls are missed entirely. Speed-to-lead matters enormously: a homeowner with a leaking roof who reaches voicemail calls the next roofer on the list within minutes. If you are going to run paid search, the single highest-ROI fix is often not in the ad account at all — it is making sure every paid call is answered live, fast, by someone who can book an inspection. LSA's ranking literally penalizes you for missing calls; Search just silently wastes your money when you do.

Bidding outside your profitable geography

Roofing economics are local and drive-time-sensitive. A lead 45 minutes outside your tight service area costs you crew windshield time and often closes worse. Yet default campaign settings, and especially PMax, happily spend in the fringe. Radius and ZIP-level targeting, with bid adjustments for your densest, highest-margin neighborhoods, is basic blocking and tackling that a surprising number of accounts skip.

Counting the wrong conversions

If your "conversions" in Google Ads are form fills and the algorithm is optimizing toward more form fills, it will dutifully find you more form-fillers — including the cheap, low-intent ones that never become jobs. Feed real outcomes back into the system: phone calls of meaningful length, booked inspections, and ideally signed jobs via offline conversion import. Optimize toward dollars, not toward the cheapest possible "conversion." This one change quietly separates accounts that scale profitably from accounts that scale their waste.

A real account structure that holds up

If you decide Search is worth running, here is a structure that survives contact with reality. It is deliberately boring; boring is what works.

Campaign and budget layout

• Campaign 1 — LSA (run first, run always). Separate product, separate budget. Tune your service area tight, drive reviews, answer every call. For many roofers this alone carries the paid program.
• Campaign 2 — Search: high-intent replacement/repair. Exact and phrase match on the terms a homeowner with a real problem types. Examples of the intent (not a copy-paste list): roof replacement, replace my roof, roof repair near me, roof leak repair, emergency roof repair, new roof cost, roofing contractor + your city. Tight geo. Manual or target-ROAS bidding once you have conversion data.
• Campaign 3 (optional) — Search: storm/weather, gated. Only live when there is genuine storm activity and you have inspection capacity. This is the campaign you pause the rest of the year. Keep the messaging strictly on inspection and documentation — "book a free storm inspection," "document hail damage" — and never on promising claim approval, waived deductibles, or a "free roof." (Why that matters legally is its own section below.)
• No standalone Display. PMax only after your offline-conversion data is clean and you have budget to experiment without panicking.

Keyword and negative discipline

Start tighter than feels comfortable. Broad match plus Google's automation is a great way to spray money. Use phrase and exact match on proven terms; expand only into search terms that have actually converted. Build the negative-keyword list relentlessly from the search-terms report — every week, add the junk: jobs, careers, salary, DIY, how to, cheap, free [non-storm], wholesale, supplier, materials, course, license, insurance jobs, competitor names you do not want to pay to defend, and the steady drip of weird off-target queries. Over a quarter this list becomes one of your most valuable assets.

Bidding and budget pacing

New accounts have no conversion history, so the algorithm is guessing. Start with manual CPC or maximize-clicks with a hard CPC cap to control downside, gather 30-50 conversions, then switch to a smart bidding strategy (target CPA or target ROAS) fed by real conversion values. Set budgets you can sustain for at least 60-90 days; paid search punishes start-stop spending because the learning resets. And set a CPL alarm: if your trailing 14-day cost per qualified lead crosses the ceiling you calculated at the top, you throttle or pause — you do not "give it more time to optimize" while it drains the account.

Tracking you must have before spending a dollar

• Call tracking with a dedicated number per campaign, recording call length and ideally outcome.
• Form tracking that fires only on real submissions.
• A back-end tie from lead → booked inspection → signed job, even if it is a spreadsheet your office manager updates. Without this you are flying blind on the only number that matters: cost per job, not cost per click.

The storm-and-claims line you cannot cross in your ad copy

Because so much profitable roofing demand is storm-driven, roofers are constantly tempted to write ad copy and landing pages that wander into dangerous territory. Keep your paid search firmly on the documentation and estimate side of the line, never the claims-handling side.

What you may legitimately advertise and do: inspect a roof, document damage thoroughly with photos and measurements, prepare an accurate, Xactimate-aligned repair estimate for your own scope of work, and hand that documentation to the homeowner. You can state plain facts about what you found and what your repair would cost. That is normal contracting.

What you must not advertise, promise, or imply — anywhere in your ads, headlines, landing pages, or call scripts:

• Do not offer to "handle," "negotiate," "fight," or "manage" an insurance claim for a fee. For a non-licensed party to negotiate a claim on the homeowner's behalf is unlicensed public adjusting in essentially every state.
• Do not interpret the homeowner's policy or tell them what is or is not covered. Coverage is the insurer's determination.
• Do not promise a specific payout, an approval, or that the carrier "will pay."
• Do not advertise that the deductible is waived, absorbed, eaten, or gone — that is illegal in many states and is insurance fraud to actually do.
• Do not advertise a "free roof." It is the single biggest red flag for regulators and a classic deceptive-advertising problem under FTC and state rules.
• Do not represent the homeowner against their insurer.

The compliant frame, which also happens to convert honest homeowners better: we inspect, we document the damage, we write you an accurate repair estimate; you file your claim, your insurer decides coverage, and we do the work we are qualified to do. The homeowner files. The insurer decides. You document and build. Running storm ad copy that respects that line keeps your license, your reputation, and your Google account safe — Google itself will disable ad accounts for deceptive claims, and a "free roof" headline is exactly the kind of thing that triggers it.

When paid search is genuinely not worth it for you

A fair, practitioner answer has to include the cases where the smart play is to spend the money elsewhere. Paid search is probably not your best dollar right now if:

• Your average ticket is low and your margin is thin. If you mostly do small repairs at 30% margin, your max CPL is so low that competitive paid clicks cannot clear it. Either change your offer mix (push toward replacements and higher-value work) or skip Search and lean on LSA, referrals, and reputation.
• You cannot answer the phone live and fast. If paid calls hit voicemail, you are pre-committed to wasting the budget. Fix call handling before you fund ads, not after.
• You have no tracking and no intention of building it. Running paid search without call tracking and back-end attribution is gambling. You will not know what worked, so you cannot improve it.
• You are post-storm and the keywords are on fire. As covered above, the surge is the worst time to start bidding. Targeted door/list outreach to actually-affected roofs is cheaper and less crowded.
• Your service area is tiny and rural. Low search volume means the algorithm starves for data and your CPLs are erratic. Local reputation, referrals, and direct outreach usually beat thin paid search.
• Your reviews and online reputation are weak. Paid clicks land on a company a homeowner then Googles — and bounces from if your review profile is sparse or negative. Earn the reviews first; ads amplify a good reputation and expose a bad one.

None of these are permanent. Each is a fixable precondition. But spending on ads while they are broken is how owners "prove" to themselves that Google Ads don't work, when what actually failed was the foundation underneath.

Paid search versus the alternatives, honestly

"Worth it" is always compared to what? Here is how Search stacks against the other places a roofing acquisition dollar can go.

The pattern: referrals and LSA are the floor everyone should be standing on. Search is a strong amplifier for the right shop. SEO is the asset you build so you are not renting every lead forever. And targeted outreach is the underrated lane — especially around storms — because it lets you pick roofs by condition and likelihood of need instead of paying a premium to interrupt whoever happened to type a keyword today.

Closing the data gap that makes Search expensive

Here is the structural reason paid search is costly: when you bid on a keyword, you are paying to reach whoever is searching right now, with no idea whether their roof is actually due. You are renting intent at auction prices, and you are competing for it with everyone else who wants that same impatient buyer. You have almost no information about the physical roofs behind the searches.

That is the gap RoofPredict is built to close, and it is worth understanding because it reframes the whole "worth it" question. Instead of paying to interrupt searchers, you target the roofs themselves — house by house, by condition and likelihood of need:

• Roof-age range per address from aerial imagery. Not a guaranteed install date — an estimated age range — so you can see which roofs in a neighborhood are aging out of their service life and are statistically due, and prioritize those addresses for outreach. It is a probability signal, not a certificate.
• Storm physics modeled per roof. After a hail or wind event, instead of fighting forty competitors over "hail damage" keywords, you get a per-roof model of which specific roofs were most likely affected — odds, not proof — so your crews knock the doors that are actually worth knocking and your mail hits the addresses most likely to have damage worth documenting.
• Enrichment of your own CRM and mailing list. RoofPredict layers roof-age range and storm-likelihood signals onto a contractor's own list to rank doors, routes, and direct-mail targets. It is not a lead-buying service and it does not replace your sales process — it tells you which roofs to prioritize so you stop spending equally on roofs that are nowhere near due.

The honest limits, because hype helps no one: age is a range, not a precise date; storm modeling gives odds, not a guarantee that a given roof is damaged or that any claim would be approved; and it still takes a real inspection, real documentation, and a real estimate to turn a prioritized address into a job. What it changes is where you point your effort first. A roofer who knocks the most-likely-due and most-likely-affected roofs runs a cheaper, less crowded funnel than one paying surge prices for storm keywords — and can use paid search as the amplifier on top of that targeted base rather than the whole strategy.

The useful way to think about it: paid search rents intent at auction; roof-condition targeting buys you a head start on the roofs that are actually due. The strongest roofing growth machines run both — LSA and tight Search to catch active searchers, plus prioritized outreach to the roofs the data says are aging out or storm-worn — so they are never only dependent on winning an overpriced keyword auction.

Writing ads and landing pages that actually convert clicks to calls

Most roofing ad accounts fail at the cheapest possible place to fix: the words. You can have perfect bidding and the wrong message and still bleed. A few principles that move conversion rate, which is the multiplier on your entire budget.

Match the headline to the search, not to your ego. A homeowner who typed "roof leak repair near me" does not care that you are "family owned since 1998." They care that you can stop water coming through their ceiling today. The ad headline that wins says, in plain words, roof leak repair, same-day inspection, [your city]. Speak to the problem the searcher just typed, then put the trust signals (years in business, licensed and insured, Google Guaranteed) in the supporting lines. People scan for relevance first and credibility second.

Use the ad to pre-qualify. It feels counterintuitive to write copy that repels some clicks, but a click you do not pay for is worth as much as a lead you close. If you do not do small repairs, do not write copy that invites "fix my one shingle" clicks. If your minimum job is a full replacement, hint at scope. You want the homeowner who is ready to spend, not the one fishing for a free patch — and every unqualified click you avoid is money back in the budget.

Put the phone number everywhere and make it the action. Roofing is a phone business. Most high-intent homeowners want to talk to someone, especially with a leak. Use call extensions, a call-only ad variant for mobile, and a landing page where tapping the number is the single most obvious thing on the screen. Forms are fine as a secondary path, but a roofer optimizing only for form fills is leaving the warmest, fastest-closing leads on the table.

Anatomy of a landing page that earns its clicks

The page a paid click lands on is where half your money is won or lost, so treat it like a sales rep, not a brochure. A high-converting roofing landing page has, top to bottom:

1. A headline that repeats the searcher's intent and your city. "Roof Replacement and Repair in [City] — Free Inspection." The visitor should know in one second they are in the right place.
2. A tappable phone number, sticky on mobile. Above the fold, large, with a click-to-call. This single element often outperforms every form on the page.
3. One primary action. Either call or book an inspection. Not five competing buttons. Confused visitors leave.
4. Proof, immediately. Star rating and review count, license and insurance, Google Guaranteed badge if you run LSA, a couple of real photos of your crews and finished roofs. Homeowners are about to let strangers on their house; reduce the fear.
5. A short form as the secondary path. Name, address, phone, problem. Every extra field cuts completion. Ask for the minimum that lets you call back fast.
6. Speed. Sub-three-second load on a mid-range phone over LTE. A slow page silently throws away clicks you already paid for; speed is conversion.
7. No navigation maze. A paid landing page should not be your full website with a menu that lets people wander off. Strip it down to the decision: call, book, or leave.

A page that converts cold paid traffic at 12% instead of 6% does not improve your results by a little — it halves your cost per lead on the exact same ad spend. That is the highest-leverage edit in the whole account, and it costs nothing but attention.

The math of scaling — why "more budget" is not a strategy

Once an account works, owners reach for the budget slider. But paid search does not scale linearly, and understanding why saves you from the classic trap of pouring money into a winner until it stops being one.

Marginal clicks get more expensive and lower-intent. Your first dollars buy the cheapest, highest-intent searches. As you raise budget and bids to capture more volume, you reach further down the intent curve — broader terms, fringe geography, lower-quality times of day. Your average CPL creeps up as you scale, even with a perfectly managed account. The job that penciled out at a $275 CPL might be a $340 CPL at triple the volume. Re-run your kill-switch math at each new spend level; do not assume a winner at $3k/month is still a winner at $9k/month.

Capacity is a real constraint. If you scale lead volume past what your sales team can follow up on within minutes, your close rate falls and your effective cost per job rises — you paid more to close a smaller share. Scaling ad spend without scaling fast follow-up is how accounts "stop working" for reasons that have nothing to do with Google.

Geography and density set a ceiling. A tight, dense metro can absorb a lot of budget profitably. A thin suburban or rural footprint hits diminishing returns quickly because there simply are not enough high-intent searches. When you near that ceiling, the next profitable dollar is usually not more paid search — it is SEO, reputation, referrals, or targeted outreach to roofs you already know are due, which do not bid against you at auction.

What roofing owners get wrong most often

After the math and the mechanics, the failures cluster into a short list of repeat offenders. If you recognize your shop here, fix these before you blame the channel.

• Judging the account on cost per click or cost per lead instead of cost per job. A "cheap" lead that never closes is infinitely expensive. A "pricey" lead that closes at a $14k job is a bargain. Track to signed work or you are optimizing the wrong number.
• Letting calls go to voicemail. It is worth repeating because it is the most common, most expensive mistake. You paid for the call. Answer it live, fast, every time, or do not run the ads.
• Starting and stopping the budget. Pausing when you get busy and restarting when you get slow resets the algorithm's learning and inflates costs each time. Set a budget you can sustain for a quarter.
• Trusting vanity conversions. Counting every form fill and phone tap as a "conversion" lets the algorithm chase cheap junk. Feed it real outcomes.
• Running storm copy that crosses the legal line. "Free roof" and "we handle your claim" headlines feel like they convert — until they cost you your license, a regulator's attention, or a disabled Google account. Stay on the document-and-estimate side.
• Treating paid search as the whole strategy. The roofers who quietly win run paid search on top of referrals, reviews, and outreach to roofs that are actually due — so they are never hostage to a single overpriced auction.

Should you run it in-house or hire an agency?

A practical question once you have decided paid search is worth testing. Neither answer is universally right; it depends on spend level and your own bandwidth.

In-house (or owner-run) makes sense at lower spend — say under a couple thousand dollars a month — and especially for LSA, which is simple enough to manage yourself: verify, set your area, answer calls, gather reviews, dispute bad leads. The fundamentals of a tight Search account (negatives, geo, landing page, call tracking) are learnable, and at small budgets an agency's monthly fee can exceed the waste you would have made yourself.

An agency or specialist earns its fee when your spend is high enough that a few points of efficiency outweigh the management cost, or when you genuinely have no time to maintain the account weekly. If you hire one, vet hard: insist on transparent reporting tied to booked jobs, ownership of your own ad account and data, weekly negative-keyword work, and a clear answer to "how do you handle click fraud and off-target traffic." Walk away from anyone who promises a fixed cost per lead with no knowledge of your margins, talks only about clicks and impressions, or wants to lock your account inside their own agency wrapper so you cannot leave with your history. The good ones obsess over your cost per job; the bad ones bill you for activity.

Whichever path, the non-negotiable is the same: you own the kill-switch math, the call answering, and the back-end attribution. No agency can substitute for those, and no account survives without them.

A 30-day decision plan

If you want a concrete way to find out whether paid search is worth it for your shop without lighting cash on fire, run this:

1. Days 1-3: Do the math at the top. Calculate your real max profitable CPL from your actual ticket, margin, willing spend %, and honest close rate. Write the number on the wall. This is your kill switch.
2. Days 1-5: Fix the preconditions. Stand up call tracking. Make sure every call is answered live and fast. Build one fast, mobile-first landing page with a tappable phone number and your proof up top. Get your Google Business Profile and reviews in order.
3. Days 3-7: Launch LSA first. Verify for Google Guaranteed, set a tight service area, start gathering and disputing. For many roofers this is the whole answer and you can stop here.
4. Days 7-30: Run a tight Search test, capped. Small daily budget, phrase/exact high-intent terms, aggressive negatives rebuilt weekly, manual or capped bidding, dedicated landing page. Track to cost per booked inspection, not cost per click.
5. Day 30: Compare against your kill switch. Is your cost per qualified lead under the ceiling from step 1? Are inspections turning into signed jobs at your expected close rate? If yes, scale deliberately and add offline-conversion import. If no, diagnose in order — call answering, landing page, geo, negatives, ticket size — and either fix the specific leak or move the budget to LSA, referrals, reviews, and targeted outreach.

That plan answers the question with your numbers instead of an agency's promises. And it builds the tracking discipline that makes every future marketing dollar — paid or not — measurable.

So, are Google Ads worth it for roofers?

For the right shop, run the right way: yes, often very much so. A roofer with a healthy ticket, solid margin, fast call answering, a converting landing page, real attribution, and the discipline to cut waste can make Search and LSA into a reliable, scalable profit engine — and many do.

For a shop with thin margins, missed calls, a generic homepage, no tracking, and a habit of starting and stopping the budget, Google Ads are a slow leak that feels like marketing. The product is not the problem; the foundation is. And even the best-run paid program is one lane of a healthy roofing growth strategy — strongest when it sits alongside referrals, reputation, SEO, and targeted outreach to the roofs that are genuinely due and genuinely storm-worn.

The roofers who win the "is it worth it" question are simply the ones who stopped asking Google to tell them which roofs are ready, and started knowing — then used paid search to catch the active searchers on top of a base of roofs they already knew were due. Do the math first. Fix the foundation. Start with LSA. Test Search small and tracked. And aim your most aggressive effort, especially after a storm, at the roofs the data says are actually worth the door — that is where the cheap, uncrowded growth lives.

FAQ

How much do Google Ads cost for a roofing company?

There is no flat price — you set the budget — but the economics are driven by cost per click and cost per lead. In competitive metros, high-intent roofing clicks commonly run $25-$75+, and after you factor in landing-page conversion you can pay several hundred dollars per genuine lead. Local Services Ads charge per lead instead of per click and are usually the better starting point. The number that actually matters is your cost per booked job, which depends on your close rate, not the click price.

Are Local Services Ads better than regular Google Search ads for roofers?

For most residential roofers starting out, yes. LSA charges per lead rather than per click, lets you dispute irrelevant leads, carries the Google Guaranteed trust badge, and ranks you on things you control like responsiveness and reviews. Regular Search has a higher ceiling and can scale further, but it demands tight management, negative keywords, call tracking, and a converting landing page or it wastes money fast. Run LSA first; add Search once your tracking and call handling are solid.

What is a good cost per lead for roofing on Google Ads?

It is whatever sits below your maximum profitable cost per lead, which you calculate from your average job value, gross margin, the share of profit you will spend to acquire a job, and your real close rate. A shop with a $14,000 average job and 45% margin might profitably pay $400-$500 per lead; a shop with an $8,500 job and 35% margin might have a ceiling near $150. The same dollar figure can be great for one roofer and ruinous for another.

Is click fraud a real problem in roofing PPC?

Yes. Roofing is a high-cost-per-click category, which attracts competitor click-draining, bot traffic, and mis-intent clicks from people searching for jobs, DIY help, or materials. Google filters some invalid clicks automatically and refunds them, but wasted spend in competitive local categories commonly lands in the 15-30% range once fraud plus off-target queries are counted. A relentless weekly negative-keyword list, IP exclusions, and for larger budgets a third-party click-fraud filter are the main defenses.

Should I run Performance Max campaigns for my roofing business?

Be cautious. Performance Max hands Google's algorithm control and spreads spend across Search, Display, YouTube, and Gmail with little visibility, and for local service businesses it often buys cheap, low-intent impressions that produce reportable conversions but few real jobs. It is only defensible once you feed real booked-job data back to Google as offline conversions and have budget to experiment. Start with LSA and a controlled Search campaign; treat PMax as a later, optional supplement.

Why do my Google Ads leads not close?

Paid-search leads are shoppers who often call several roofers, so they close at a lower rate than referrals or door-knocked work — typically 20-35% for residential, lower if your follow-up is slow. The biggest fixable cause is call handling: a large share of contractor inbound calls go to voicemail, and a homeowner with a leak calls the next company within minutes. Answer every paid call live and fast, send clicks to a focused landing page, and book the inspection on the first call.

Is it worth bidding on Google Ads right after a hailstorm?

Usually it is the worst time to start. After a storm, every roofer in a wide radius floods the same keywords, so click costs can triple while a lot of the traffic is homeowners merely checking whether they have damage. The cheaper, less crowded play is targeted outreach to the specific roofs most likely affected — knocking those doors and mailing those addresses while competitors overpay for crowded keywords. Keep any storm ad copy strictly about inspection and documentation.

Can I advertise that I will handle a homeowner's insurance claim?

No. Negotiating, adjusting, or handling an insurance claim on a homeowner's behalf for a fee is unlicensed public adjusting in essentially every state. You also cannot advertise a waived deductible, a free roof, or a promised approval or payout — those run into insurance-fraud and deceptive-advertising rules and can get your Google account disabled. The compliant frame: inspect, document the damage, write an accurate repair estimate for your own scope, and let the homeowner file while the insurer decides coverage.

How long before Google Ads start working for a roofing company?

Local Services Ads can produce leads within days once you are verified. Search campaigns need weeks of tuning — gathering 30-50 conversions before smart bidding behaves, building negative-keyword lists, and refining geo and landing pages. Plan to commit a stable budget for at least 60-90 days, because start-stop spending resets the algorithm's learning and inflates your costs. SEO and reputation, by contrast, take months but compound instead of disappearing when you stop paying.

How does RoofPredict relate to running Google Ads?

Paid search rents buyer intent at auction — you pay to reach whoever is searching, regardless of whether their roof is due. RoofPredict targets the roofs themselves: an estimated roof-age range per address from aerial imagery and per-roof storm-likelihood modeling, layered onto your own CRM or mailing list to rank doors and routes. It is a probability signal, not a guarantee, and it does not replace inspection and estimating. Used together, targeted outreach to due and storm-worn roofs gives you a cheaper base, with paid search amplifying it to catch active searchers.