The mechanic behind it
Model training absorbs brand names in proportion to how often they appear. A franchise appears in hundreds of city-specific pages, national press, and its own location directory. That produces a strong, general association: this brand does this trade. The assistant can then attach it to any city with very little retrieval work.
Your business, if it is well covered, has a strong association in exactly one place. That is a better answer for the customer, and it is more expensive for the model to arrive at, because it requires actually retrieving local sources rather than recalling a name.
How much of this we can actually see
The table below is drawn from our own leaderboard corpus, which runs the same buyer questions across many metros and trades and records who gets named. Where the same brand name is returned across ten or more separate metro leaderboards, it is doing exactly what this page describes: travelling on national recognition rather than local evidence.
The good news in this
A national association is broad and shallow. It has no answer for the specific question. When the customer asks about a particular job, a particular neighborhood, or a constraint like same-day or after-hours, the chain's generic coverage runs out, and the assistant has to retrieve. That is the moment a well-covered local business wins, and it is where the higher-intent customers are anyway.
What to do about it
- Stop competing on the bare category question. "Best plumber in [city]" is the question chains are strongest on and the one with the least buying intent.
- Own the job-level questions. One page per specific job, in the customer's words, with a checkable fact in the first two sentences.
- Own the constraint questions. Same-day, after-hours, emergency, warranty, financing, specific brands you service. Chains rarely publish these at city level.
- Get named on the local roundups. A single credible local best-of page is often enough to break a chain's monopoly on a question.
The job-level question inventory
Chains win the generic question. This exercise finds the twenty questions where they do not, which is where your next four service pages come from.
Build the list
- Write down the ten jobs you most want more of, in the words a customer would use, not your invoice codes.
- For each, add the constraint that customers actually ask about: same day, after hours, weekend, emergency, warranty, financing, or a specific brand you service.
- Add your three biggest neighborhoods or suburbs by name, not just the metro.
- You now have a list of specific questions. Ask five of them in an assistant and note who gets named.
Turn it into pages
- Give each of the top four its own page, with the question as the heading.
- Answer it in the first two sentences, before any preamble about your company history.
- Include one number nobody else publishes: a typical response time, a price band, a warranty length.
- Name the neighborhoods in plain text on the page.
Then check
- Re-ask the same five questions a month later, with the wording unchanged.
- Record whether you are now named, and by which assistant.
- If nothing moved, the gap is third-party coverage rather than your own pages, and the next step is a roundup rather than a fifth page.
23,926
citations recorded in AI answers
798
distinct source domains cited
85%
of them point at a business's own website
Names that travel across metros
From our leaderboard corpus of 40 metros across 21 trades: businesses returned by assistants in ten or more separate city leaderboards.
- 01TerminixPest Control Companies40 metros
- 02Two Men and a TruckMoving Companies40 metros
- 03The Cleaning AuthorityHouse Cleaning Services39 metros
- 04A1 Garage Door ServiceGarage Door Companies38 metros
- 05OrkinPest Control Companies38 metros
- 06H&R BlockAccountants & CPAs38 metros
- 07Planet FitnessGyms & Fitness Studios34 metros
- 08Valvoline Instant Oil ChangeAuto Repair Shops34 metros
Computed live from 182 stored scans covering 23,926 citations across 798 distinct domains. Assistants were asked real buyer questions unprimed; every source domain they cited or linked was recorded and classified. Full method and sample.
You are looking for the term
Brand authority
The advantage a chain has here is what the field calls brand authority: the strength of the association between a name and a category in the model's own weights. You are unlikely to out-authority a franchise nationally, which is why the winning move is specificity rather than scale.
- Do chains pay to be recommended?
- No. There is no paid placement in an assistant's recommendation. Their advantage is coverage volume, which is earned by having hundreds of locations written about.
- Can a single-location business ever beat a franchise here?
- Routinely, on specific questions. The generic category question is the franchise's strongest ground and the customer's weakest intent.
- 01Top ways to ensure your content performs well in Google's AI experiencesGoogle Search Central
- 02AI features and your websiteGoogle Search Central
- 03Tips to improve your local ranking on GoogleGoogle Business Profile Help
- 04GEO: Generative Engine OptimizationAggarwal et al., arXiv
Every link above was opened and confirmed reachable on September 1, 2026.
The automated version
See what ChatGPT, Claude, Gemini and Perplexity say about your business
The free scan runs the buyer questions for your trade and city across all four assistants, shows who gets named instead of you, and lists the sources those answers came from. No signup, about thirty seconds.
Run my free scan