Do Contractor SEO Services Actually Work?

Yes, contractor SEO services work, with a caveat that matters more in 2026 than it did two years ago. Contractor searches are transactional and local, which is the category least damaged by AI Overviews. But SEO produces phone calls on a delay, not traffic on demand, and most contractors measure the wrong number and conclude it failed.

By Carl Jiang, Co-Founder · Last updated November 2026

The honest answer, including the part about AI

SEO works for contractors for a structural reason: homeowners with an urgent, local, expensive problem search for someone nearby and then call. That behaviour has not changed. What has changed is how many of them click a blue link on the way.

The numbers are worth knowing before you budget. Ahrefs analysed 300,000 keywords and found AI Overviews correlate with a 58% reduction in click-through rate for the top organic result. Pew Research, across nearly 69,000 sessions, found 8% of users click a link when an AI Overview is present against 15% when it is not.

Here is the part most agencies will not tell you, and it cuts in your favour. That damage is concentrated on informational queries, not transactional ones. Seer Interactive’s analysis found position-one CTR on informational queries fell around 34% with an AI Overview present, while transactional queries fell about 12%. “How does a roof membrane work” is the query getting answered without a click. “Roof replacement Kent WA” is not, because the searcher wants a company, not an explanation.

So the practical implication for a contractor is not that SEO stopped working. It is that informational blog content aimed at the top of the funnel has lost most of its value, while service pages, location pages and your Google Business Profile have lost very little. An agency still selling you twelve informational blog posts a month is selling you the thing that broke.

Traffic or phone calls: what SEO actually produces

Both, but only one of them pays you, and the gap between them is where most contractors lose confidence in the channel.

Traffic is an input. It becomes a lead only if the page matches what the person wanted, the site loads on their phone, and there is an obvious way to call or book. We cover that plumbing on the website development page, because a site that ranks and does not convert is the most expensive kind of success.

The sequence is: rankings produce impressions, impressions produce clicks and calls, and clicks produce leads only if the page works. A report showing traffic up and calls flat means the problem is on the site, not in the rankings.

And for contractors specifically, a large share of the result never appears as website traffic at all. Map Pack calls go straight from the Google listing to your phone. The homeowner never visits your website. If you are judging SEO by Analytics sessions, you are not seeing the channel’s best output.

How to calculate a realistic lead number for your market

Nobody can honestly tell you “SEO produces 25 leads a month” without knowing your market. But you can estimate it yourself in ten minutes, and you should before signing anything.

Work it like this:

1. Find your main service term’s monthly search volume in your metro. Say “deck builder [your city]” at 1,200 searches a month.
2. Add the realistic variants you could rank for. Service plus city, service plus suburb, “near me” variants. Call it 3,000 combined.
3. Apply a realistic click share. If you hold top-three organic and Map Pack presence, assume roughly 15 to 25% of that volume reaches you across both surfaces. That is 450 to 750 visits and calls.
4. Apply a lead conversion rate. For a contractor site that actually works, 3 to 8% of visitors become a lead. Map Pack calls convert far higher because the person is already dialling.
5. Apply your close rate to get jobs, then your average ticket to get revenue.

Run that honestly. If your metro has 300 monthly searches for your service, SEO is not going to transform your business and anyone promising otherwise has not looked. If it has 8,000, the ceiling is high enough to justify real investment.

This is also how you sanity-check a proposal. Ask the agency to show you this calculation for your market before you sign. If they cannot, they have not researched it.

Lead attribution: telling an SEO lead from a referral

This is the hardest part of measuring SEO for a contractor, and it is why so many owners genuinely cannot tell whether their agency is working.

The problem is that your lead sources overlap. A homeowner hears your name from a neighbour, then Googles you, then clicks your Google listing and calls. Is that a referral or an SEO lead? Honestly, both. The referral created the intent and search closed the gap. Attribution models fight over this constantly.

What actually works for a contractor:

Separate tracking numbers by source. A different number on your website, your Google Business Profile, and your paid ads. Same phone rings. Now you know which surface produced the call rather than guessing.

Branded versus non-branded search, split in your report. Someone searching your company name was sent by a referral, a truck, or a past job. Someone searching “deck builder near me” found you cold. Only the second group is new demand your SEO created. Any report that merges these is overstating the agency’s contribution, and the gap is large: branded position-one CTR runs more than twice non-branded.

Ask on the call, and log it. “How did you hear about us?” is imperfect because people misremember, and plenty will say “I found you on Google” when a neighbour told them your name first. But logged consistently for six months it is directional and it costs nothing.

Track first-touch and last-touch separately if you can. The referral is often first touch and search is last touch. Knowing both stops you cutting the channel that quietly assists every other one.

Watch new versus repeat customers in your CRM. SEO should be producing people who have never bought from you. If your lead count rises but they are all repeat customers, that is retention, not acquisition.

The honest position: perfect attribution does not exist in home services, and any agency claiming it does is overselling. What you can get is a defensible picture, which is enough to make a budget decision.

Seasonality: why month over month lies

Most trades swing hard by season, and comparing March to February tells you almost nothing about whether your SEO is working.

Roofing spikes after storms and in late summer. Decks and outdoor living get researched in February and built April through July. HVAC peaks twice, in the first heat and the first freeze. Plumbing is steadier but spikes in cold snaps. If your agency starts in November and you judge them in February, you are reading seasonal decline as failure.

Three rules that fix this:

• Compare year over year, not month over month. March 2026 against March 2025 is the only comparison that controls for season.
• Before year one is complete, compare against the search volume trend, not against last month. If searches in your category fell 40% and your share of them rose, you gained ground in a shrinking month.
• Judge the shoulder months. Everyone looks good in peak season. Whether SEO is working shows in whether your quiet months got less quiet.

The practical consequence for planning: content and technical work should be done three to four months ahead of your season, because pages need time to rank. Publishing deck content in April is publishing it for next year.

Month three against month twelve

Brief version, since the full timeline sits on what an SEO company does each month.

Month three should show movement, not money. Technical issues resolved, Google Business Profile properly configured, first pages published and indexed, rankings climbing on lower-competition terms, impressions clearly up in Search Console. Calls probably flat. If nothing has shipped by now, that is a different problem.

Month twelve should show a business difference. Competitive service and city terms placing, Map Pack presence across more of your radius, a measurably different call volume against the same month last year, and new-customer leads rather than repeat buyers.

The failure pattern to watch for is twelve months of rising impressions with flat calls. That means the keywords being won are the wrong keywords, usually informational terms that look good in a report and, in 2026, increasingly get answered without a click.

Frequently asked questions

Does SEO produce traffic or does it produce phone calls?

Both, but they are not the same thing and only one pays you. Rankings produce impressions, impressions produce clicks and calls, and clicks become leads only if the page matches intent and makes calling obvious. For contractors a large share of the result never appears as website traffic at all, because Map Pack calls go straight from the Google listing to your phone.

What is a realistic monthly lead number for a contractor?

It depends entirely on search volume in your metro, so calculate it rather than accepting a promise. Take your main service term’s monthly volume plus realistic variants, assume roughly 15 to 25% reaches you at top-three organic with Map Pack presence, apply a 3 to 8% site conversion rate, then your close rate and average ticket. A market with 300 monthly searches will not transform your business. One with 8,000 can.

How do you tell an SEO lead from a referral or a repeat customer?

Use separate tracking numbers for your website, Google Business Profile and paid ads. Split branded from non-branded search in reporting, since someone searching your company name was sent by a referral while someone searching “deck builder near me” found you cold. Ask on every call and log it. Watch new versus repeat customers in your CRM. Perfect attribution does not exist in home services, but this gives a defensible picture.

How much does seasonality change what good looks like?

Enough to invalidate month-over-month comparison entirely. Roofing spikes after storms and in late summer, decks are researched in February for an April build, HVAC peaks at the first heat and first freeze. Compare year over year, and before year one is complete compare your share against the category’s search volume trend. Judge the shoulder months, because everyone looks good in peak season.

What should month three look like against month twelve?

Month three shows movement without money: technical fixes done, Google Business Profile configured, first pages indexed, rankings climbing on easier terms, impressions up, calls probably flat. Month twelve shows a business difference: competitive terms placing, wider Map Pack presence, call volume measurably different against the same month last year, and new customers rather than repeat buyers.

Want this calculation run for your market before you spend anything? Book a free strategy call and we’ll work through your search volume and realistic lead ceiling with you.