GEO for local businesses

This guide is for owners of shops, studios, restaurants, clinics and other businesses that serve people nearby. It explains how to improve the chance that an AI assistant names you when someone asks for a good place near them. Most of the work is a few days of tidying up listings, then about an hour a week of upkeep. It does not show how to become the first recommendation, because no documented method can guarantee that.

The short version

1. Claim and complete your listings on the map and review platforms your customers use.
2. Use one business name, address, phone number and set of opening hours everywhere.
3. Earn genuine reviews and reply to them. Never buy or filter them.
4. Find and fix the other pages AI reads about you: classifieds, business registries and complaint sites.
5. Put the facts customers ask about on your own website, in plain text.
6. Measure how often assistants mention you, with real customer questions, every month.

1. Claim your listings where customers look

AI assistants answer local questions from what they can find about you. For Google, that starts with your Business Profile. In China, Dianping, Meituan and Amap play the same role, and each assistant draws on different platforms (Which sources Chinese AI assistants cite).

Google says local results are mainly based on relevance, distance and popularity. It also states: “There’s no way to request or pay for a better local ranking on Google.”

Officially documented · Google Business Profile Help

Google advises keeping business information up to date: “Businesses with complete and accurate info are more likely to show up in local search results.”

Officially documented · Google Business Profile Help

Google documents how its own local results work. It does not document how Gemini, AI Overviews or other assistants pick local businesses. Chinese map and review platforms publish no such documentation either. Treat complete, accurate listings as the base that every assistant can read, not as a ranking lever.

Fill in every field the platform offers:

• category, address, service area, phone, opening hours and holiday hours;
• services or menu, price range and what is included;
• recent photos of the place, staff and work;
• qualifications or licences that customers check (for example a coach’s certification or a clinic’s licence).

2. One name, one set of details

Assistants have to work out that several pages describe the same business. Different spellings of your name make that harder.

Google requires that a Business Profile name “should reflect your business’s real-world name”, as used on the storefront, website and stationery. “Including unnecessary information in your business name isn’t permitted, and could result in the suspension of your Business Profile.”

Officially documented · Google Business Profile Help

A vendor study of Pilates studios in Hefei found one studio recommended under three names: its full name, a short name and a name with a business-type suffix (81, 60 and 17 times). Two other leading studios were also split into separate names. In one answer, Qwen openly said it was guessing that two names meant the same studio. (Translated from Chinese.)

Observed in practice · observed 2026-08 · 合肥图灵象人工智能科技有限公司 (unpublished report, not publicly available; readers can't check it)

Pick the exact name on your sign and use it everywhere: map platforms, review sites, social accounts, your website and your business registration where possible. Keep address and phone formats identical. Don’t add keywords or the neighbourhood to the name.

An assistant can only recommend a business it is sure exists. One name, spelled one way everywhere, is the cheapest GEO work a shop can do.

3. Earn genuine reviews, and answer them

Reviews feed both platform rankings and what assistants say about you. That makes them a target for manipulation, and the rules are clear.

Prominence is partly based on how many websites link to a business and how many reviews it has. Google says: “More reviews and positive ratings can help your business’s local ranking.”

Officially documented · Google Business Profile Help

Google Maps removes fake engagement. This includes reviews paid for directly or in kind, and offering incentives such as discounts or free services in exchange for a review. Merchants must not discourage negative reviews or “selectively solicit positive reviews from customers”.

Officially documented · Google Maps Help

China’s revised Anti-Unfair Competition Law took effect on 15 October 2025. Article 9 forbids false or misleading commercial claims about matters including user reviews. It also forbids helping others make such claims by organising fake transactions or fake reviews.

Officially documented · 上海市市场监督管理局

What you can do:

• Ask every customer for a review in the same way, after the visit, with a link or QR code.
• Reply to reviews, especially critical ones, with facts and what you changed.
• Never pay for reviews, offer a discount for them, or ask only happy customers.

4. Look after the pages you don’t control

Many of the pages AI reads about a local business are not on review platforms.

In the Hefei study (3,469 citations across consumer AI apps), the most cited source type was local classifieds sites, at 28% of citations. Few studios manage these. Local lifestyle review platforms contributed 324 citations and short-video platforms 244. Business registry lookups and complaint sites together contributed 201. (Translated from Chinese.)

Observed in practice · observed 2026-08 · 合肥图灵象人工智能科技有限公司 (unpublished report, not publicly available; readers can't check it)

In the same study, DeepSeek cited complaint and consumer-rights platforms 75 times, 23.9% of all its citations.

Observed in practice · observed 2026-08 · 腾讯新闻（报道合肥图灵象普拉提行业GEO白皮书）

Search for your business name on these sites. Correct old prices and addresses on classifieds listings, and claim the listing where you can. Resolve open complaints and record the outcome on the platform. Make sure your business registration shows the current name and address. If an assistant still repeats something wrong, follow When AI gets your brand wrong: find it, fix it, prevent it.

5. Put the facts on your own website

Your website is the one source you fully control. Customers ask assistants specific questions, such as whether a studio handles postnatal recovery or whether a clinic has evening slots. A page that answers them gives assistants something accurate to cite.

• One page per location, with the name, address, phone and hours in plain text, matching your listings.
• What you offer, for whom, at what price, and what makes you different.
• Staff qualifications and real photos.
• Make sure search engines can index the site (AI search access audit). Organization markup helps keep the facts unambiguous (Organization structured data).

In the Hefei study, the studio appeared in only 34% of answers to niche pre-purchase questions, such as who is good at postnatal recovery. Broad category questions returned it 77% of the time.

Observed in practice · observed 2026-08 · 腾讯新闻（报道合肥图灵象普拉提行业GEO白皮书）

Niche questions are where a clear page about your specialities can make a difference.

What not to do

• Don’t buy a guaranteed “No. 1 in AI”. No platform offers paid placement in organic local results. Google says outright that you can’t pay for local ranking (section 1). Treat a guarantee as a warning sign (Before you hire a GEO agency).
• Don’t mass-publish fake recommendation articles. China’s 2026 Qinglang campaign treats this as AI data poisoning.

The Cyberspace Administration of China’s 2026 campaign lists AI data poisoning among its targets, including the use of GEO for malicious marketing. It also covers mass-produced near-identical posts and fake engagement.

Officially documented · 新华网（中央网信办通知）

• Don’t buy or filter reviews (section 3). It breaks platform policy and, in China, the Anti-Unfair Competition Law.
• Don’t stuff keywords into your name. It confuses assistants and can get a Google profile suspended (section 2).
• More on why seeded mentions don’t work: Claim: seeding brand mentions across the web boosts AI visibility and GEO regulation in China.

Anyone who promises your shop first place in AI answers is promising something no platform offers. Pay for work you can check, not for a rank.

How you’ll know it’s working

AI answers vary from run to run, so one screenshot proves nothing.

The vendor tracked one studio’s recommendation rate over two months. It started at about 40% and fell to 10% during a month with no new content. It then climbed to 68% within three weeks and fell back to 40%. The report concludes that claims of a “stable first place” do not match what it observed. (Translated from Chinese.)

Observed in practice · observed 2026-08 · 合肥图灵象人工智能科技有限公司 (unpublished report, not publicly available; readers can't check it)

A simple monthly routine:

1. Write 10–20 questions real customers ask, in their words, including your area (“best Pilates studio near [district] for back pain”).
2. Ask each in the apps your customers use, from your city, logged out. Repeat each one many times; Measuring AI visibility by sampling prompts explains how many.
3. Record the share of answers that mention you, and whether the details are correct.
4. Ask new customers at the front desk how they found you, and note “AI assistant” as an option.
5. Track Google Business Profile calls and direction requests alongside.

Compare month to month, not day to day. For reporting upward, see For marketing leads: measuring GEO and reporting it upward.

Evidence at a glance

Recommendation	Evidence
Complete, accurate listings	Official (Google, for its own local results)
One consistent name	Official (Google rules); observed (Hefei study)
Genuine reviews, no incentives	Official (Google Maps policy, Chinese law)
Fix classifieds, registry and complaint pages	Observed (one vendor study, one city)
Answer niche questions on your website	Observed; benefit for AI not documented
Measure shares over many runs	Observed
Avoid guaranteed rankings and seeded articles	Official (Google, CAC)