Local search has been the most stable corner of SEO for a decade: get the listing right, get reviews, get local citations, appear in the map pack. That machinery still works. But a growing share of local intent now goes to an assistant that returns two or three recommendations in a sentence, and that surface follows different rules.

What stays the same

The fundamentals remain load-bearing, because assistants draw from the same underlying data.

Accurate, complete business listings. Consistent name, address and phone across every platform. Genuine review volume and recency. Categories chosen precisely. Hours that are actually correct, including holidays.

If these are wrong, nothing downstream works. An assistant that finds three conflicting addresses for you treats you as an unreliable entity, and the fastest way to lose a recommendation is ambiguity about basic facts.

What changes

The result set shrinks dramatically. A map pack shows three and a list below. An assistant names two. Position eight on a results page still received traffic; the equivalent position in a conversational answer receives none. The competition is for a much smaller set of slots.

Reviews are read, not counted. Traditional local ranking weighs star average and volume. An assistant reads the text and synthesizes what customers actually say. A business with 60 detailed reviews describing specific strengths can outperform one with 400 reviews saying "great service" — because the first gives the machine something to say about you.

Specificity beats proximity more often. Traditional local search weights distance heavily. Conversational queries carry qualifiers — "who can do a same-day repair," "who handles commercial systems," "who works with older homes" — and an assistant matches on the qualifier. A business that has clearly stated what it specializes in wins queries that pure proximity would have lost.

Service area content matters again. Not the old thin doorway pages, which fail on every surface. Genuine content about serving a specific area: the conditions there, the typical problems, what is different about the work.

The practical checklist

State your specialties explicitly, in plain sentences. "We specialize in commercial refrigeration for restaurants" is retrievable. A photo gallery implying the same thing is not.

Ask for specific reviews. Instead of "please leave us a review," ask customers to mention what work was done. This produces review text with the detail assistants need.

Answer local questions on your site. The real ones: what a typical job costs in this area, how fast you can get there, what the permit situation is, what you will not take on. Each is a query someone is asking an assistant right now.

Fix your listing inconsistencies. Audit every platform. This is tedious, unglamorous, and it is the single highest-return action for most local businesses.

Be explicit about your service area. Name the places. A radius on a map is not text, and text is what gets read.

Measuring it

Ask the assistants. Run your real local queries — the phrasings customers actually use, with the qualifiers they include — through the major assistants monthly, from the geographies that matter. Record whether you appear and what is said about you.

It takes twenty minutes a month and it is the only direct measurement of a surface that traditional local rank tracking cannot see at all.