The contractors appearing most often in Google's Map Pack top three had more Google reviews and more listed locations. Their websites were also more likely to offer online scheduling, even after accounting for review count. Our comparison of 1,117 US contractor websites, fetched Nov. 1, 2026, points to a practical website standard: make the business easy to understand, publish useful buying information and give customers a working path to book.
Last verified: November 1, 2026.
We believe a contractor's website should help a customer choose the business and take the next step. Clear services, prices, credentials and booking options serve that purpose. This study looks at which homepage features appeared alongside strong Google Maps performance and where the next website project can make the customer experience better.
We read the homepage code of plumbing, HVAC, electrical, roofing and other contractor sites for features a website project can change: schema, llms.txt, links to prices, specials and financing, scheduler code, word count and platform. The comparison measures those features alongside Map Pack standing; it isn't a before-and-after redesign experiment.
Map Pack leaders had more reviews and more listed locations
Leaders, in our data, held a top-three spot on at least a tenth of the points we sampled in their market across four weekly scans; they're the same leaders as in our star-rating study. We compared the leader websites we could read with two groups from our census of contractor sites in the same map results: sites that reached the top three sometimes, and sites that never did.
Start with review counts and listed locations, which differ between these groups. Neither measures company size directly. Of leaders' sites, 71.0% had 200 or more reviews on their largest Google listing, against 14.7% of sites that never reached the top three. Many leaders were multi-location brands, too: 40.6% of their sites had two or more Google listings in these markets, against 2.8% of never-top-three sites.
Taken raw, the homepages differ as well. Leaders' homepages linked to prices, specials or financing 51.6% of the time, against 20.2% for sites that never reached the top three. That share climbs with Google review count, though, across every site we checked: from 10.0% of sites with fewer than 50 reviews to 73.2% of sites with 500 or more.
Inside each review band the two groups landed close together, and leaders came out lower in two of the three bands in the chart below. Reweight the never-top-three sites to the leaders' mix of review counts and their rate comes out at 49.7%. Most of those links go to specials and financing pages, not to price lists.
The review-count pattern
Sites with more Google reviews link specials and financing more often
Share of homepages linking prices, specials or financing
Bars run from 0 to 100%.
50 to 199 reviews
3 of 18 sites
20 of 96 sites, weighted
200 to 499 reviews
5 of 14 sites
14 of 32 sites, weighted
500 or more reviews
24 of 31 sites
22 of 31 sites, weighted
Cheers comparison of US contractor websites by Google Map Pack standing, four weekly scans, weeks starting Sept. 28 to Oct. 19, 2026; homepages re-fetched Nov. 1, 2026 (UTC)
| Reviews on the largest Google listing | Map Pack leaders | Never top three |
|---|---|---|
| 50 to 199 reviews | 16.7% (3 of 18 sites) | 19.6% (20 of 96 sites, weighted) |
| 200 to 499 reviews | 35.7% (5 of 14 sites) | 42.4% (14 of 32 sites, weighted) |
| 500 or more reviews | 77.4% (24 of 31 sites) | 71.0% (22 of 31 sites, weighted) |
Compare website features within review bands to separate the raw pattern from differences in the businesses' review histories.
The website foundations common to both groups
At the same review count, none of the gaps below was big enough to tell apart from chance in a sample this size. LocalBusiness schema sat on 71.9% of leaders' homepages against 64.8% of never-top-three sites. Leaders came out lower on a valid llms.txt file (a plain-text file that gives AI tools a short summary of a site), 33.3% against 41.4%, and on a dollar amount anywhere in the homepage text, 29.7% against 36.9%.
Financing links were a dead heat at 40.6% and 40.3%. More of leaders' homepages carried 1,000 or more words (71.9% against 62.2%), and more were built on WordPress.
Contractor homepages at the same review count
Matched for review count, only the scheduler row pulls apart
Share of sites with each feature
LocalBusiness schema
llms.txt file
Link to prices, specials or financing
Financing link
Dollar amount in homepage text
1,000 or more words of text
Online scheduler code
Cheers comparison of 1,117 US contractor websites by Google Map Pack standing, four weekly scans, weeks starting Sept. 28 to Oct. 19, 2026; homepages fetched Nov. 1, 2026 (UTC). Leaders 64 readable homepages (69 sites for llms.txt)
| Homepage feature | Map Pack leaders | Sometimes top three | Never top three |
|---|---|---|---|
| LocalBusiness schema | 71.9% | 62.9% | 64.8% |
| llms.txt file | 33.3% | 40.1% | 41.4% |
| Link to prices, specials or financing | 51.6% | 52.4% | 49.7% |
| Financing link | 40.6% | 40.0% | 40.3% |
| Dollar amount in homepage text | 29.7% | 38.1% | 36.9% |
| 1,000 or more words of text | 71.9% | 68.6% | 62.2% |
| Online scheduler code | 39.1% | 23.4% | 23.2% |
Leaders' pages were heavier and no faster, with a median of 258 KB of HTML against 150 KB and a server response of 665 ms against 508 ms. That's one fetch of the full HTML from one location, not Core Web Vitals, so it says nothing about speed work for organic search.
Schema spells out each location's phone, address and hours in a form machines can parse, and the multi-location schema guide shows how to mark it up. A concise llms.txt can point AI tools to useful pages. Keep both accurate alongside the visible content, service pages and booking path. These are parts of a well-built site; their presence alone did not distinguish the Maps leaders in this sample.
The one clear lean toward leaders: online scheduling
One feature pulled apart. At the same review count, 39.1% of leaders' homepages loaded online scheduler code, against 23.2% of never-top-three sites, and the lean showed up in each review band with enough leaders to compare.
The one clear lean toward leaders
Leaders loaded an online scheduler more often in each review band
Share of homepages loading online scheduler code
Bars run from 0 to 60%.
50 to 199 reviews
4 of 18 sites
8 of 96 sites, weighted
200 to 499 reviews
5 of 14 sites
7 of 32 sites, weighted
500 or more reviews
16 of 31 sites
10 of 31 sites, weighted
Cheers comparison of US contractor websites by Google Map Pack standing, four weekly scans, weeks starting Sept. 28 to Oct. 19, 2026; homepages re-fetched Nov. 1, 2026 (UTC)
| Reviews on the largest Google listing | Map Pack leaders | Never top three |
|---|---|---|
| 50 to 199 reviews | 22.2% (4 of 18 sites) | 8.0% (8 of 96 sites, weighted) |
| 200 to 499 reviews | 35.7% (5 of 14 sites) | 23.2% (7 of 32 sites, weighted) |
| 500 or more reviews | 51.6% (16 of 31 sites) | 32.3% (10 of 31 sites, weighted) |
Online scheduling deserves a close look in the next website review. The association gives us a reason to test the customer path and measure completed bookings; the study design and 23-feature comparison are detailed in Method.
It does fit what our star-rating study found on the Business Profile side: listings with a booking link reached the top three more often at matched review counts, for example 52.8% against 41.1% at 100 to 249 reviews. That booking link is a separate field on your profile, so check where it sends people as well as the scheduler on your site.
On the website, add a scheduler that fits how customers book and the slots your team can honor. Connect it to the right branch and measure completed bookings.
Price pages are rare, and worth it for customers
Lots of sites link something with a price word in the address, which isn't the same as a price page. Of the 960 contractor homepages we checked, 59 linked a page whose address names prices, costs, rates or fees, and we opened every one. Only 15 (1.6%) led to a page with prices on it, and leaders were no more likely to have one: 2 of 64. The rest were service cost pages, cost guides, blog posts, quote forms, estimators and broken links.
That leaves a clear opportunity to answer a question customers and AI assistants already ask. Google began offering searchers in some industries an AI agent that calls local businesses for prices, and in a study published Dec. 10, 2025, Invoca found that 26% of those calls from July to November 2025 went unanswered. In 49% of the answered ones, the business gave no price, most often in pest control and plumbing, where whoever answered said they need a home visit to price the job. Invoca didn't give a sample size.
A price page answers the question before anyone picks up the phone. Publish a diagnostic fee, a trip charge or a typical installed range your crews actually charge, and say what moves the final price.
Any fee that lands on every job goes inside the price or range, not on its own line. California's hidden-fee law has required that since July 1, 2024; Minnesota passed a similar law in 2024, and other states have rules of their own.
If "every job is different" is the objection you hear from your own team, our pricing playbook shows how to publish ranges you can stand behind. Keep your specials page and date every offer, but don't count specials or financing as a price page.
Keep the full pricing explanation on your website. Google also provides a dedicated price field in the Business Profile services editor, so use the appropriate field when adding a service price and keep it consistent with the page.
What a redesign still has to get right
You can run the first check this week, before anyone sends a proposal. On a computer, open your homepage in Chrome and press Ctrl+U (Option+Command+U on a Mac) to see the raw HTML. Search it with Ctrl+F (Command+F) for four things: your main services and the towns you serve as plain text, a "tel:" link with the number on your Business Profile, a link to a page that shows real prices, and your scheduler or request path. Then click through that price link and read what it actually tells a customer.
Page source is what a crawler that doesn't run JavaScript gets, and in Vercel's late-2024 test, "none of the major AI crawlers currently render JavaScript" (Gemini uses Googlebot's rendering). So the phone path has to survive a script failing. In our census, 86.0% of the 976 readable contractor homepages had a clickable phone link, and 8.6% had no phone link, form or booking prompt in the raw HTML at all. If call tracking swaps the number with a script, make sure the raw HTML still carries a real tel: link that matches your Google Business Profile.
Then build the pages AI engines cite, which a redesign often neglects. Of 74,969 AI citations to contractor websites in the Cheers Market Baseline, 43.1% went to homepages and about 48% to service, location or service-plus-city pages, while blogs and resource pages got 1.5%. That argues for one page per service and per location you actually serve, each with its own local facts, which our location-pages study lays out.
AI agents may already be reaching your booking pages. In Microsoft Clarity's Oct. 29, 2026 study of more than 5.2 million agentic sessions, 8.42% included a book-appointment event, though sessions can include a human and an event doesn't show the task was finished.
Last, test the whole path on cutover day. Place a test call and send a service request through the new site, then confirm the right person got both and your tracking recorded them.
What to ask a web designer before you sign
If you're shopping for contractor website design, or HVAC, plumbing or roofing website design in particular, get answers to these first:
- What evidence backs any ranking promise, and what will you measure to show the work helped? Google says "there's no way to request or pay for a better local ranking on Google," and our comparison found strong differences in review counts and listed locations, with online scheduling the clearest homepage association.
- Will there be a price page with real numbers, and will mandatory fees sit inside the price?
- Will services, towns and the phone number be in the raw HTML, with a real tel: link?
- Is a scheduler in scope, and who answers the slots?
- How will you check calls, forms and tracking at cutover?
- If there's an AI chat or booking assistant, does it say it's a bot? California's bot disclosure law covers bots used to sell to people in California.
Operating partners can turn those answers into the site standard for every acquired brand: a price page with mandatory fees inside the price, services and a real tel: link in the raw HTML, one page per service and per location the brand actually serves, and calls and tracking verified at cutover. Judge plugins and widgets by the booked jobs they bring in, and leave the choice of plugin to each brand's web vendor.
What Google says the Map Pack runs on
Google's own page on local ranking says local results are "mainly based on relevance, distance, and popularity," and that prominence is "also based on info like how many websites link to your business and how many reviews you have." Whitespark's 2026 Local Search Ranking Factors survey of 47 invited local search experts (Nov. 6, 2025) leans the same way: "On-page signals and link signals appear to have dropped a bit," while review signals rose. That's expert opinion, not Google's formula, and it fits the stronger review histories we saw among leaders.
Build the website, profiles and reputation together. Invite honest reviews from real customers and earn mentions from local sites, such as your equipment brand's dealer locator or a local news story about your crew. Give those sources a useful service or location page to link to. Don't buy reviews or links, though; Google's policies bar both, and the FTC's consumer review rule, in force since Oct. 21, 2024, bars buying fake reviews or paying for reviews that must be positive. Our review collection guide covers asking inside Google's rules, with neutral requests and no staff review quotas.
Method
In a Cheers study of 15,149 competitor listings in Google's local map results across four weekly scans (Sept. 28 to Oct. 25, 2026), 348 were leaders, holding a top-three spot on at least a tenth of the points we sampled in their market. Cheers samples those results each week on a grid of points around 66 customer locations, with each customer's own locations removed; the scans are the weeks starting Sept. 28 to Oct. 19, 2026. "Sometimes top three" means at least one top-three appearance short of the leader bar, "never top three" means none, and each website takes its best listing's group. This measures Google's local map results, not rankings in AI answers.
The 1,117 sites are the 69 US contractor websites behind leader listings whose website address we could read, once a few platform and non-US sites came out, plus a random sample of the rest from our Nov. 1 census (594 sometimes, 454 never), restricted to US home-service contractor sites and weighted back to all the sites in those results. For most leader listings, the website link in our data was a Google redirect that hides the address. Those leaders had more reviews than the ones we could read, so our 69 sites lean toward the smaller leaders and the 40.6% multi-listing share is a floor.
We fetched raw homepage HTML on Nov. 1, 2026 (UTC), with no JavaScript; 998 homepages were readable, 64 of them leaders', and 982 answered a second fetch later that day. On that second fetch we read every price-word link by hand, counting a price page only if its raw HTML showed at least two dollar amounts. The 1.6% rate's base is the 960 census contractor homepages that answered it. Of the 59 homepages with one, 15 led to a price page, 7 to a service cost page with dollar figures, 14 to cost guides or blog posts, 18 to a price-titled page with at most one dollar figure, and 5 elsewhere.
For "at the same review count," we reweighted the other groups to leaders' mix of four review bands (under 50, 50 to 199, 200 to 499, 500 or more), using the largest count on any of a site's Google listings; we didn't measure revenue. Only one leader sat under 50 reviews, so the band charts leave that band out. We compared 23 features, with bootstrap 95% intervals from 1,000 resamples and Wilson intervals for leaders' own rates.
For leaders minus never-top-three sites, the scheduler gap's 95% interval ran from +0.4 to +31.0 points, and every other row in the matched chart included zero (LocalBusiness schema, for example, ran -7.3 to +21.2). A broader test for booking software code of any kind didn't clear the bar, and no detectable difference isn't proof that two groups are the same. Re-grouped without the markets of the company with the most tracked locations, and separately with no company's markets above a fifth of any group's weight, the scheduler lean held in direction and no other feature favored leaders. Two noisier markers we don't report, any form tag and any CAPTCHA name in the code, ran lower on leaders' homepages.
It's observational and homepage only, so every rate is a floor, and review counts are the latest value we saw rather than October's. The listings come from Cheers' weekly scans of Google's local map results in markets where 10 Cheers home-service customers operate, so they aren't a national random sample. The 74,969 AI citations come from a separate study, the Cheers Market Baseline (100 US and Canadian metros, Aug. 31 to Sept. 20, 2026). We name no site, and we'll re-run this with the next census.
Cite this: Cheers, Contractor Website Design and the Map Pack: 1,117 US contractor websites compared by Google Map Pack standing, crawled Nov. 1, 2026 (UTC). Dylan Allen-Arnegard.
If you'd rather have someone run the site, this is what our team does.
Sources
- AI features and your website, Google Search Central, Dec. 10, 2025. No special markup or AI text files needed for Google's AI features.
- Google's AI Is Calling (pricing request calls study), Invoca, Dec. 10, 2025. Unanswered AI price calls and answered calls with no price.
- Hidden fees FAQ, California Attorney General, undated. Mandatory fees belong inside the advertised price.
- Price Transparency Law FAQ, Minnesota Attorney General, undated. Minnesota's similar hidden-fee law.
- Maps user-generated content policy, Google Maps contribution policy, undated. The "For merchants" clause on pricing in listings.
- The rise of the AI crawler, Vercel, December 2024. Major AI crawlers don't render JavaScript.
- What AI Agents Do on Websites: Insights From 5.2 Million Sessions, Microsoft Clarity, Oct. 29, 2026. Book-appointment events in agentic sessions.
- Tips to improve your local ranking on Google, Google Business Profile Help, undated. Relevance, distance and prominence, and no paying for a better local rank.
- Bots: disclosure, California Legislative Information, undated. California's bot disclosure rule for sales.
- Local Search Ranking Factors 2026, Whitespark, Nov. 6, 2025. Expert panel view that on-page and link signals dipped while review signals rose.
- Trade Regulation Rule on the Use of Consumer Reviews and Testimonials, Federal Trade Commission, Oct. 21, 2024. The ban on buying fake or positive-only reviews.
Sources read November 1, 2026.
Dylan Allen-Arnegård is the CEO & Co-Founder of Cheers, the done-for-you local visibility partner for established home-service companies and PE-backed portfolios.