Local SEO in the AI era
Where AI recommendations for nearby businesses come from, and the three things a local service provider needs to have in order.
Lees dit in het Nederlands →$ ask --region=own-city --advice=who
Local SEO in the AI era
In the AI era, local SEO is broader than a position on the map: when someone asks an AI assistant for a recommendation nearby, the answer comes from business profiles, reviews and local listings. Three things matter then: an up-to-date Google Business Profile, consistent listings of name, address and phone number, and a site that answers customers' questions literally. This piece is written for those who draw customers from their own region; if you work nationally, this isn't your first priority.
A physiotherapist, an accountant, an installer: anyone who draws customers from their own region experiences the shift to AI assistants in their own way. For years, local SEO was about ranking at the top of the map; now that same work also determines whether you're mentioned when someone asks an assistant who can help them nearby.
01 / the questionWhat happens with a recommendation question
An AI assistant asked for a local recommendation compiles its own shortlist. It combines what it can find: business profiles, reviews, listings on aggregator sites and the businesses' own websites. From those sources it distils a handful of names with a short rationale attached.
Unlike a map with ten pins, that answer is a selection. Whoever isn't on it simply doesn't exist for that question. And the selection leans heavily on what's written about you, not on how good your work is; the two only coincide if your customers have recorded that somewhere.
Also note how the question is phrased. A customer rarely types the name of your trade the way you use it. They describe a problem: who can fix my tap, where can I go with back pain, who does the bookkeeping for a sole trader. The assistant translates that problem into a field of work and searches within it. If you're only described under your trade title, you miss the questions phrased in problem language.
The AI doesn't recommend the best, but the best described.
02 / the sourcesWhere those recommendations come from
The most important source for local questions is your Google Business Profile, with the reviews attached to it. Next come review sites and trade directories, and only then your own website. That AI systems lean heavily on third-party sources is a general pattern; why that works that way is covered in why AI refers to third parties more often than to your site.
For local practice this means something reassuring: you don't have to become a content machine. The sources that weigh most heavily are profiles and reviews you largely already have. They just need to be correct, up to date and consistent with each other.
03 / the three thingsWhat you need to have in order
The checklist, with the first concrete action for each point:
- An up-to-date Google Business Profile. Opening hours, services, photos and reviews that get answered. A profile that's stood still for two years reads to an AI as a business that may no longer exist. First action: check your opening hours and services today, and reply to every unanswered review.
- Consistent listings. The same name, the same address and the same phone number everywhere your business appears: profile, site, directories, review sites. Discrepancies make it unclear to a machine whether it's the same business, and that lack of clarity costs listings. First action: make a list of every place your business is listed and align the name, address and phone number.
- A site that answers questions literally. How much does a treatment cost, do you work in my town, how quickly can you come. Put those questions as headings on your site and answer them in the first paragraph underneath. That's exactly the text an assistant can cite when someone asks that question. First action: pick your most visited page and rewrite it according to that principle.
Of those three, the first can be sorted in an afternoon, the second is a job of checking and correcting over a few evenings, and the third is ongoing work you can tackle page by page. The order is deliberate: an assistant that doesn't trust your profile doesn't even get to your site.
04 / the limitWhen this isn't your priority
If you've just started and barely have any reviews, focus first on satisfied customers willing to write that down somewhere. Without that foundation there's little for an AI to find, and no amount of optimisation changes that.
If your calendar is structurally full from word of mouth, doing nothing is a defensible choice. This work pays off for those who want to grow or notice that inflow is declining. If you want to know where you stand first, you can check it yourself with a handful of test questions.
And if you mainly sell to customers outside your region, this piece wasn't written for you: the general rules for visibility then weigh more heavily than local profile work, and you're better off starting with the basics at datadriftdigital.com and the rest of this series.
How do I test whether AI recommends my business locally?+
Why does the AI mention my competitor nearby but not me?+
Do reviews count towards AI recommendations?+
Do I need to be listed on every review site and in every directory?+
What if my business has just started?+
Does your business appear when someone in your region asks an AI for advice?
The scan
This is the baseline measurement: fixed search queries on ChatGPT, Gemini, Claude, Perplexity and Google. Current prices are on the pricing page. You keep the report even without a follow-up.
This text was produced with AI assistance and checked and approved for publication by a human.