Why AI recommends national chains instead of your shop

First, the honest size of the chain problem

It is smaller than you think. Across those 21 home-services audits the engines named 218 rival businesses, and national franchises accounted for just 5.5% of them — 12 mentions. Roto-Rooter, the most-named chain in the country, took 3.2%. Six of the 21 companies had a franchise named against them at all. The other 94.5% of the time, the business AI recommended instead of you was another local shop — the one across town, with a real van and a real owner.

So when a contractor tells us “AI just recommends the big chains,” the data says: mostly it doesn't. It recommends your actual competitor. That is a more solvable problem and a more uncomfortable one.

What is remarkable is that home services is the only trade where a franchise shows up at all. We ran the same cut across med spas, dentists and personal-injury firms — 547 distinct rivals, and not one national brand between them. Zero. If you're a plumber, you are the only kind of local business whose category has a national incumbent in the room. The full cross-vertical benchmark is here.

The real finding: AI doesn't read your website

This is the part we didn't expect. We went through all 126 answers and counted which sources the engines leaned on when they described a contractor — the places they went to find out whether you're any good.

77% of the answers cited at least one third-party review source. Only 19% referenced the company's own website — and that's counting generously, across the 18 of 21 businesses that had a domain on file at all. The BBB alone outranks your website by 2.5×. Reddit — strangers arguing in a city subreddit — outranks your website by nearly 2×.

Read that again, because it inverts twenty years of contractor marketing advice. You paid for the website. You paid for the SEO. And when a homeowner asks an AI who to call, the model goes and reads a BBB profile you have never logged into and a Reddit thread you have never seen.

This is also the actual reason chains win when they win

Once you see the source list, the franchise advantage stops being mysterious and stops being about advertising. Roto-Rooter is not winning because a model was impressed by its brand. It is winning because it has a BBB profile, a Yelp page, an Angi listing and a Reddit mention in every city in America — so whatever corpus the engine reaches for, in whatever town, there is a Roto-Rooter entry sitting in it, already written, already described.

You have one city's worth of that. The chain has four hundred. It isn't out-spending you on the model — it is out-appearing you in the handful of places the model actually retrieves from. Which means the counter-move is not a bigger ad budget. It is being present, and described, in those same places.

What to do about it

1. 1Claim and complete your BBB profile. It is the single most-referenced source in our answer set (48%) and it is the one most contractors ignore because it doesn't send them clicks. It doesn't send clicks — it sends citations. Different job.
2. 2Go and look at your city's subreddit. Reddit appeared in 37% of answers. Search your trade and your city. If your competitors are named in those threads and you are not, that is a live, ongoing input into what AI tells your customers. You cannot astroturf this and shouldn't try — but you can be the company that actually shows up and answers, under your own name.
3. 3Fill in the aggregators you already half-own — Yelp (26%), Angi (20%), Google Business (13%), HomeAdvisor (7%). Not for their traffic. For the fact that the engine reads them as evidence.
4. 4Get your service area and specialisms written down in prose, somewhere retrievable. The engines describe businesses; they cannot describe a business whose pages only say “Quality Service Since 1994.” Say what you do, where you do it, and what you don't do.
5. 5Then check whether it worked. The rival list underneath your score is a literal list of who is getting the calls that were meant for you.

Methodology

21 completed AskedAbout free mini-audits of US home-services companies — plumbers, HVAC contractors, roofers and electricians — run in production between 2026-06-12 and 2026-07-10 across 12 cities (Austin, Boise, Charlotte, Chicago, Columbus, Dallas, Denver, Indianapolis, Kansas City, Nashville, Phoenix, Tampa). Each audit puts three buyer-intent questions (“what are the best {trade} in {city}?”, “who do you recommend?”, “is {business} a good choice?”) to two AI search engines — OpenAI's search model and Perplexity's Sonar — producing 126 answers. Source attribution was counted by scanning each answer's full text for named sources; “own website” means the answer referenced the business's own domain. A rival is any business named in an answer other than the audited one. Aggregate figures only; the businesses are anonymized and were never contacted.

Limits, stated plainly. n=21 is small and these are the trades and cities we happen to audit, not a random draw — read the source table as a strong signal, not a national census. Counting a source as “referenced” means the engine named it in its answer; it is not the same as a formal citation link, and Perplexity (which cites) will naturally surface more sources than a model that doesn't. The franchise list is hand-enumerated (Roto-Rooter, Mr. Rooter, Benjamin Franklin Plumbing, One Hour Heating & Air, ARS/Rescue Rooter, Aire Serv, Goettl, Mister Sparky, Service Experts, Mr. Electric), so a regional chain we failed to recognize is counted as a local independent. An earlier version of our vertical benchmark put the franchise share at 12.9% and Roto-Rooter at 10.3%; those figures were contaminated by our own synthetic smoke-test audits and were corrected on 2026-07-11. The numbers on this page are the clean ones.

Related reading: the cross-vertical benchmark (152 businesses, four verticals), why AI recommends your competitors instead of you, why AI doesn't recommend your home-services business, and our page for home-services companies.