How AI Search Is Changing Local Business Discovery in the UK
A UK customer looking for an accountant in 2024 typed “best accountant near me” into Google. They scanned a Google Maps three-pack, read a few reviews, and called whoever looked most credible.
The same customer in 2026 might still end up calling someone from Google Maps. But the path before that call has changed. They asked ChatGPT for accountants who specialise in contractors. They asked Perplexity to compare three local options. They saw a Google AI Overview answer that named two firms by reputation. By the time they opened Google Maps, the shortlist was already in their head.
This is the shift that matters for UK local-service businesses. AI search is not replacing Google Maps. It is replacing the research stage that used to happen on Google’s blue links. Local SMBs that ignore the shift are losing the consideration phase. Local SMBs that abandon Google Maps for AI search are losing the booking. The right response is two-track, not one-track, and this post breaks down what is changing, which UK local-service categories are most exposed, and how to invest accordingly.
What Local Search Looked Like Before AI Engines
For most of the last decade, UK local discovery ran on a single dominant pattern. Customer types a “service + location” query into Google. Google returns a Maps three-pack at the top, organic blue links below. Customer picks from the three-pack roughly two-thirds of the time. Blue links served as overflow when the three-pack did not satisfy the query.
That pattern made Google Business Profile the single most valuable local marketing asset for UK SMBs. Optimisation tactics were stable and well-documented: verify the listing, build review volume, add accurate categories, post regular updates, reply to questions. The pattern rewarded businesses that ran consistent local SEO rather than businesses that produced thought leadership content. If you want a refresher on the wider search shift, a plain-English explainer on what GEO means covers the foundations.
Three things were true that are no longer fully true. The customer’s research happened entirely inside Google or via direct browsing of business websites. Peer-style recommendations happened in private channels (text messages, family conversations) that search engines could not see. And the shortlist a customer brought to Google Maps was usually empty or one name long. In 2026, all three have shifted: customers do meaningful research inside AI engines before they touch Google Maps, peer-style recommendations now appear inside AI answers because the engines weight Reddit and forum content, and shortlists are longer when customers arrive at Maps because they have already filtered three or four candidates from an AI conversation.
The Two-Stage Local Discovery Funnel in 2026
UK local discovery in 2026 has split into two distinct stages with different platforms winning each. The research stage is moving to AI engines for high-consideration services. The booking stage stays on Google Maps for almost all categories.
The research stage is where customers form a shortlist. They are not yet ready to call. They are asking comparative or qualifier questions. “Which solicitors in Birmingham handle commercial property?” “Are there accountants in Manchester who work with creative agencies?” “What should I look for in a divorce lawyer?” These queries get more useful answers from ChatGPT, Perplexity, and Gemini than from a Google Maps three-pack, because the customer wants synthesis and reasoning, not a list of pins on a map.
The booking stage is where customers convert. They have a name or two in mind. They want phone numbers, opening hours, parking, exact address, latest reviews. Google Maps wins this stage because it integrates location, contact, and reviews in one interface. ChatGPT cannot tell you whether a dentist’s car park is busy on a Tuesday afternoon. Google Maps can.
How UK SMBs should split budget between GEO and traditional SEO covers the broader budget split argument. For local-service businesses specifically, the rule of thumb is more skewed toward traditional SEO and Google Business Profile than the cross-vertical average. Most local SMBs should hold 70-80% of their search budget on Google Business Profile, local SEO, and review management. The remaining 20-30% goes to AI search visibility. The exact split depends on which category you are in, which we cover below.
The mistake we see most often is local SMBs treating the funnel as if AI search has replaced the booking stage. It has not. The head-to-head comparison of GEO and traditional SEO covers the same logic at a wider level, with the booking-stage skew even stronger for local-service businesses.
How ChatGPT, Perplexity, and Gemini Recommend UK Local Businesses
AI engines do not return a Maps three-pack. They synthesise an answer from sources they have indexed or fetched live, and that synthesis names specific businesses by reputation. Understanding how each engine arrives at those names matters because the optimisation tactics differ.
ChatGPT pulls from its training corpus plus live web search. For UK local queries, it relies heavily on review aggregators (Trustpilot, Yelp, Yell.com), professional directories (Law Society for solicitors, ICAEW for accountants), Reddit threads, and the businesses’ own websites. It tends to name businesses that appear consistently across multiple sources. A local SMB with a strong Google Business Profile but no Trustpilot presence and no Reddit mentions is largely invisible to ChatGPT.
Perplexity weights recent content more heavily than ChatGPT and pulls live from search results plus Reddit. For UK local queries, Perplexity often surfaces named recommendations from recent forum threads, specialist directories, and trade press. It is the engine most sensitive to community sentiment and the one most likely to surface a niche specialist over a large general firm if the niche specialist has stronger forum and review presence.
Gemini and Google AI Overviews share an index. They lean on Google’s existing local search infrastructure plus the wider web. For UK local queries, Gemini often surfaces businesses that already rank well in standard Google local search, which means strong local SEO doubles as Gemini optimisation. The difference is structural: Gemini favours content with clear claim-and-evidence pairs over keyword-heavy pages, even for the same underlying business.
Microsoft Copilot pulls from Bing’s index plus its own knowledge sources. For UK local queries, Copilot is the weakest of the four because Bing’s UK local data is thinner than Google’s. Most UK SMBs do not need to optimise for Copilot specifically until that changes. Across all four engines, the shared signal is consistent presence on review aggregators, professional directories, and specialist forums, plus a website with specific factual claims about what the business does and who it serves. Generic “we provide quality service” content is filtered out by all four.
Which UK Local Service Categories Are Most Exposed to AI Search
Not every UK local-service category is equally exposed to AI search. High-consideration services where customers research before they call see meaningful AI search influence. Low-consideration services where customers search and call within minutes are barely affected yet.
| Category | AI search exposure | Typical research-stage AI use | Recommended GEO investment |
|---|---|---|---|
| Solicitors and accountants | High | Specialism comparison, qualifier questions | 25-35% of search budget |
| Healthcare and wellbeing | Medium-High | Practitioner approach, treatment comparison | 20-30% of search budget |
| Professional consultancy | High | Vendor evaluation, niche specialism | 30-40% of search budget |
| Trades (plumbing, electrical) | Low | Rare research stage | 5-10% of search budget |
| Hospitality and food | Low-Medium | Cuisine match, ambiance comparison | 10-15% of search budget |
| Retail and showroom | Low | Stock check, location | 5-10% of search budget |
The pattern is consistent. The higher the consideration cost (financial stakes, expertise required, ongoing relationship), the more research happens before contact, and the more AI search matters. The tactical guide on getting your business cited by ChatGPT covers the citation mechanics that high-exposure categories should focus on.
One UK-specific nuance worth flagging. Professional services regulated by trade bodies (Law Society, ICAEW, ACCA, RICS) get a citation lift in AI engines because the engines treat regulator-directory presence as a credibility signal. If your business has not claimed and optimised its trade body listing, that is one of the highest-impact moves you can make.
What Google AI Overviews Changes for Local Queries
Google AI Overviews now appears on a meaningful share of UK local-service queries. The Overview sits above the Maps three-pack and answers the customer’s question in synthesised text with citations to sources Google considers authoritative.
For local SMBs, three things matter about Overviews. They reduce clicks to organic blue links because the answer is already on the page. They can name specific businesses by reputation, often pulling from review platforms and editorial sources rather than the businesses’ own sites. And they behave differently on commercial-intent queries (where they are more cautious about naming specific businesses) versus informational queries (where they freely synthesise recommendations).
The practical result for UK local SMBs is that Overviews steal some research-stage attention from Google’s organic results, but they have not yet reduced Google Maps three-pack traffic meaningfully. Customers who would have read three blog posts before deciding now read the Overview and one of the cited sources. Customers who would have clicked the Maps three-pack still click it.
What this means for content production is specific. Local-service businesses should produce content that answers the qualifier and comparison questions Overviews tend to surface for, with clear claim-and-evidence structure. Generic “why choose us” pages are not cited. Specific “how to choose a commercial solicitor in Manchester” or “what to expect from a first consultation with a contractor accountant” content is, and the same content also performs well in ChatGPT and Perplexity.
Where Local SMBs Should Not Invest in AI Search Yet
Some UK local-service categories should defer AI search investment in 2026. Three patterns indicate “not yet”.
Categories where customers convert within minutes of searching. Emergency trades, takeaway food, hairdressers, taxis. The customer types “plumber near me” because their boiler has failed. They will call whoever appears first with a phone number. AI search adds nothing because there is no research stage to influence.
Categories where local Google Business Profile is severely under-optimised. If your business does not yet have verified GBP, accurate categories, regular review volume, and accurate opening hours, that is the work to do first. AI search investment without strong GBP foundations is layering an upper-funnel tactic on a broken lower-funnel.
Categories where the business serves a tightly geographic catchment without national or specialist positioning. A single-location dentist serving a 10-mile radius gets little benefit from AI search visibility because the queries that surface them are predominantly proximity-based, which Google Maps already handles. The exception is if the practice has a clear niche (paediatric, orthodontics, anxiety patients) that customers actively research.
If you have decided to invest, the platform-specific guide on getting cited by Google Gemini and AI Overviews covers the tactical detail. The decision to invest should come first.
The Two-Track Tactical Plan for UK Local SMBs
For UK local-service businesses in the high-exposure categories, the right response is a two-track plan that maintains Google Business Profile dominance for the booking stage while building citable content for the research stage.
Track one is Google Business Profile and local SEO. This continues to drive most of your booked enquiries and should stay your largest investment. The work has not changed. Verified GBP. Accurate categories and service areas. Regular review volume from real customers. Photo updates. Posts about service updates. Local citations from authoritative UK directories. Schema markup using LocalBusiness.
Track two is AI search visibility for the research stage. The five highest-value tactics for UK local SMBs are claiming and optimising the relevant trade body or regulator directory listing, building consistent presence on Trustpilot and the largest review aggregator in your category, producing content that answers the comparison and qualifier questions customers ask, using claim-and-evidence sentence structure throughout that content, and earning citations in trade press, niche directories, and specialist forums where your category’s customers research.
The GEO and AI search work we run for UK local businesses usually starts with an audit to see which track is genuinely weak before recommending any investment. We see SMBs investing heavily in track two while their GBP is half-finished, and SMBs assuming track one is enough while losing research-stage attention to competitors. Both produce the same outcome: enquiry volume that does not match the marketing spend. The right balance for most high-exposure UK local SMBs is roughly 70% track one and 30% track two in 2026, likely shifting toward 60-40 over the next 18-24 months as AI search adoption deepens.
What This Costs to Implement
The cost to add AI search visibility on top of existing local SEO depends on whether you commission research-stage content piece by piece or take a packaged GEO retainer. Either approach should sit alongside continued investment in Google Business Profile and local SEO, which most UK local SMBs already budget at £500-£2,000 per month. AI search investment is incremental on top of that, not a substitute.
For piece-by-piece content, UK local-service businesses commissioning research-stage content (qualifier guides, comparison articles, niche specialism pages) typically pay £400-£800 per article. A starter set of 8-12 articles covering the main qualifier questions in your category runs £4,000-£10,000 over 3-4 months. Ongoing maintenance to keep content current adds £500-£1,500 per month.
Packaged GEO retainers for UK local businesses typically run £1,500-£4,000 per month and bundle content production, citation building, review management, and monthly reporting on AI search visibility. The combined search budget for a UK local SMB serious about both tracks usually lands between £2,500 and £6,000 per month in 2026.
The honest payback timeline for AI search work specifically is 4-9 months. Trade body directory and review aggregator presence shifts within weeks. Editorial citations and forum mentions take longer. Most local SMBs see meaningful AI engine visibility within 6 months of consistent investment, and meaningful enquiry volume influence within 9-12 months.
Not in 2026, and not soon. Google Maps owns the booking stage of local discovery for almost every category because it integrates location, hours, contact, and reviews in a way no AI engine currently matches. AI search is taking the research stage that used to happen on Google’s organic blue links. Expect a shift in where research-stage attention happens, not a replacement of the booking-stage interface.
Yes, but it is a low-priority job for most UK local SMBs. Bing Places feeds Microsoft Copilot’s local recommendations, which are still a small share of UK local search traffic. Claiming and optimising it takes around an hour per location and the upside is real. Skip it only if your time is genuinely better spent on the higher-traffic priorities.
Yes, more than before. Google reviews are one of the strongest signals AI engines use to verify a local business’s credibility. ChatGPT and Gemini both reference Google review patterns when synthesising local recommendations. Reviews remain the single highest-ROI local marketing investment for UK SMBs in 2026.
Open GA4 and segment traffic by referrer. Look for sessions from chatgpt.com, perplexity.ai, gemini.google.com, and bing.com (which captures Copilot traffic). For most UK local SMBs in mid-2026, this is between 1% and 6% of organic traffic depending on category. The number is small but growing fast. Track the trend month-on-month rather than the absolute number.
Partly yes. Strong presence on Trustpilot, Yelp, your trade body directory, and review aggregators can earn you AI engine mentions without ever publishing a blog post. Content production amplifies that base. For local SMBs with limited capacity, optimising directory and review presence first is the right starting point because it produces visibility faster than content production does.
Apple Maps now incorporates AI-generated business descriptions for some categories, and voice queries on iOS often go through ChatGPT or Apple Intelligence. The traffic is small in 2026 and not yet worth dedicated optimisation for most UK SMBs. The same content and review patterns that work for ChatGPT and Gemini work for these surfaces too. Optimise for the major engines first and the rest follow.