Understanding and improving your visibility on AI assistants - practical guides, methods and benchmarks.
Asked for the best businesses in Woodlands, Singapore, ChatGPT and Gemini described The Woodlands, Texas, in 51 of 72 answers. Four towns, same error.
Every morning an article comes out: one measurement of what ChatGPT and Gemini recommend, and why. Figures checked, limits stated.
At most one email a day. One-click unsubscribe. Your address is used for this and nothing else. Privacy
6 minmin read
In 26 answers about dentists in 13 British cities, ChatGPT mentions the NHS 4 times and insurance plans 10 times. Gemini names the NHS 21 times.
6 minmin read
In 18,662 recommendations from ChatGPT and Gemini, one recruiter in two is a national network, one accountant in five, and no dentist or restaurant.
6 minmin read
In 1,364 answers about British, French and Brazilian businesses, ChatGPT tells customers to check Yelp 266 times. Gemini names local sites instead.
6 minmin read
Asked for businesses in Newcastle, Gemini answered for Newcastle, Australia 13 times in 36, and for Birmingham, Alabama 4 times in 40. ChatGPT never did.
6 minmin read
We visited all 665 links ChatGPT and Gemini gave in their business recommendations. Of 474 to a business's own site, 141 lead nowhere.
6 minmin read
ChatGPT recommended businesses 10,105 times and warned about one: the lines at Kogi BBQ. Gemini warned 65 times, mostly about price and waits.
6 minmin read
ChatGPT credits 30 named dentists with their technology and names the tool for none. Across 256 such passages, AI gives only an adjective 140 times.
6 minmin read
In 16,022 passages recommending a business, ChatGPT and Gemini give a practical fact 97 times. Wheelchair access: never. Parking: once.
5 minmin read
In 3,600 ChatGPT and Gemini answers, 182 mention a free consultation or estimate. Only 14 passages credit one to the business they name. Lawyers: none.
6 minmin read
In 16,022 passages recommending a business, ChatGPT and Gemini name the person behind it 318 times. For dentists, 33 of 162 names are certainly wrong.
6 minmin read
In 16,022 passages recommending a business, ChatGPT and Gemini give an environmental reason 135 times. One names a certification; 57 are bare adjectives.
7 minmin read
In 16,022 passages recommending a business, ChatGPT and Gemini cite a guarantee 63 times. 41 belong to national networks such as Midas or Bosch.
5 minmin read
In 16,022 passages recommending a business, 2.6% cite hours, speed or emergencies. For British locksmiths it is 35.6%, for US accountants 0.
5 minmin read
In 16,022 passages where ChatGPT or Gemini recommend a business, 191 cite a prize or a guide: 1.2%. Michelin is 76 of them. For plumbers, almost none.
5 minmin read
In 16,022 passages recommending a business, Gemini says who it suits in 48% of its own, ChatGPT in under 5%: "best for families", "ideal if you need".
5 minmin read
In 16,022 passages where ChatGPT or Gemini recommend a business, 6% mention its age. Gemini does it 3.6 times more, bakeries far more than plumbers.
5 minmin read
In 3,600 AI answers, one gave a business's rating. Gemini gave customers a cut-off 67 times: usually 4.5 stars, often 50 reviews.
6 minmin read
In 3,600 ChatGPT and Gemini answers across four countries, a clear price comes up in 32% and a low one in 23%. And AI quotes the going rate, not yours.
5 minmin read
Across 3,600 ChatGPT and Gemini answers in four countries, AI named 19,561 businesses and gave a phone number twice. The next search decides.
5 minmin read
Across 872 US answers, ChatGPT opens 57% of its own by refusing to recommend - then names eight businesses right underneath. Gemini wrote that once in 436.
5 minmin read
The guide our fifty measured editions allow us to write. Five jobs in the order they pay, each with its figure - and what nobody can promise you.
4 minmin read
New York has twenty times Miami’s population. ChatGPT and Gemini seat the same shelf in both: a median of 20 names versus 20.5. Shelf size is a property of the answer format, not of the city.
4 minmin read
For a city and a trade, ChatGPT and Gemini seat a median of 21 names, and only 3 of them appear on the map of actual local businesses. The shelf is not the map; it is fame.
4 minmin read
Across four countries, 78.7% of answers about building and care trades send the customer to a directory. For restaurants, hotels and salons: 3.7%.
3 minmin read
We put the same 1,320 questions to ChatGPT and Gemini in four countries. Four times out of five, the two answers name no business in common.
5 minmin read
In one year, the traffic AI assistants send to US retail sites went from less valuable to more valuable than the rest. Meanwhile Adobe bought Semrush for 1.9 billion dollars to sell visibility inside AI answers.
5 minmin read
A fifty-member consortium is assembling an open base of places to ground AI. Reddit and Yelp book revenue from licensing what sits on their platforms. Your business is not paid for any of it - but entering those corpora costs nothing.
4 minmin read
AI assistants name 2,599 different businesses across 13 cities. 88% of them appear in a single city, only 21 are cited in eight cities or more, and two in all thirteen. The national layer of the answers is thin; the rest is local.
5 minmin read
Across 780 ChatGPT and Gemini answers, 77% talk about customer reviews, yet only 2 quote a star rating. For plumbers, electricians and HVAC contractors the answers read like a checklist: licence, insurance, emergency service, warranty. And Gemini names the state licensing board where ChatGPT almost never does.
6 minmin read
For twenty years a local business bought clicks and leads. AI assistants sell neither: they say names. What the Angi and Yelp filings show, what Pew and BrightLocal measured, and what our 872 US answers add.
5 minmin read
For one trade in one US city, four answers from ChatGPT and Gemini name a median of 17 businesses. Only two come back in more than one answer. Nearly nine names in ten appear once, and for dentists, 10 cities out of 13 have no name that repeats at all.
6 minmin read
Two ordinary ways of asking one engine the same thing, across 15 trades and 13 US cities. Only 11% of the names are shared, and 47% of comparisons share nothing at all. Yet at market level, 7 of the 10 most cited businesses stay the same.
5 minmin read
Same question, same city, two engines. In the US, ChatGPT cites 6.1 businesses per answer and 27% of its citations go to names found in three cities or more. Gemini cites 3.6, and 16%. In 9 trades out of 12, the gap points the same way.
6 minmin read
872 answers from ChatGPT and Gemini across 13 US cities. For staffing agencies, 74% of citations go to names that come up in three cities or more. For attorneys, dentists and HVAC contractors: 0%. What that changes for a local business.
7 minmin read
We crossed two measurements on 1,475 businesses: what their website tells a machine, and what AI assistants answer. Those that pass our three checks are named 2.3% of the time, the others 2.2%. The trade matters; the page title barely does.
7 minmin read
3,809 business websites read across twelve countries: only 37 shut the door on answer engines. The gap is elsewhere. Eight home pages in ten say who you are, three in ten say where.
5 minmin read
6,136 businesses audited across 18 markets: 5,948 were never named by any AI assistant, not once. In Japan and Saudi Arabia, none at all. And the name that comes back for "a dentist in France" is not a dentist.
4 minmin read
We wrote to 2,122 businesses across sixteen markets, at the address published on their own website. 104 mailboxes no longer existed. In Japan, one in eight.
5 minmin read
We read 1,004 small-business websites across six countries. Eight in ten publish an address a machine can read. The other two are not offline: they never write their city anywhere.
6 minmin read
300 measurements, 16 markets: two trades in the same city share almost no sources, while one trade across two cities shares five times more. What it changes for your listings.
6 minmin read
Two major studies measured AI engine citations. More than 60% are incorrect, 31% have a sourcing problem, and the confidence never drops. What that changes for a brand.
5 minmin read
We asked two AI assistants the same question, for four trades across five US cities. Both answer with confidence. They almost never name the same businesses.
6 minmin read
AI crawlers take thousands of pages for every visitor they return, and the file meant to speak to them is read by none of them. What the published measurements show.
6 minmin read
Two AI engines barely consult the same pages, yet agree far more on the companies they recommend. What the published research shows, and what it changes for your visibility.
4 minmin read
Same trade, different country: one French web agency in four is named by AI. In the United States, three out of eighty-seven. A big market erases.
3 minmin read
544 websites visited across twelve countries. 278 publish a readable address, 96 do not respond. In Japan, four businesses in five publish none.
4 minmin read
Three identical runs, five sectors, two engines: two cited names out of three appear only once. And a citation's value depends on the engine - one in two holds on Perplexity, one in twenty on Gemini.
4 minmin read
24 August: assistants do not read the same pages in every language. Today: they do not read the same NUMBER of them. In Busan, one domain is cited for all eleven trades.
6 minmin read
Five runs of the same question: the brand named first stays the same 81 % of the time. The problem is not stability - it is that there is no single first place.
5 minmin read
We sell repeated measurement. So we tested whether the list drifts from one day to the next. It barely does - and for one engine, not at all.
6 minmin read
« The best X » and « which one for my 10-person company » overlap by only 31 %, against 77 % when the exact same question is asked twice.
6 minmin read
We read the pages the AI says it consulted and looked for the recommended brand in them. 128 brands out of 128, with a median of six pages citing each one.
7 minmin read
Yesterday: the recommended brands change with the language. Today: the pages read to pick them. 90 % of the 652 cited domains appear in ONE language only.
6 minmin read
180 answers, 10 languages, 3 AI assistants. Overlap with English falls to 43% for online payments and 52% for web hosting - and stays intact for running shoes.
4 minmin read
48 answers, 8 communes, from Paris to a sub-prefecture of 10,201 people. Gemini and Perplexity name real bakeries everywhere - with the address. ChatGPT answers without a question only in Paris.
5 minmin read
3,780 real answers, 18 trades, three countries. ChatGPT declines to name a business in 43% of cases - 92% for solicitors, 91% for plumbers, 0% for bakers. Gemini: 1%. Perplexity: 0%.
5 minmin read
18 conversations, three countries, two turns each. At the moment the assistant has to pick one, a third of the time it names a business that was never on its own list.
4 minmin read
4,092 answers, three countries, 39 cities. ChatGPT names businesses in almost every answer - and in 100% of them it gives no link, no phone number and no address.
4 minmin read
Measured 19 August 2026 across five markets. Of 22 businesses that Google's first page surfaces, Gemini names 8, Perplexity 3, ChatGPT none. And in Manchester, none of the three.
3 minmin read
Yesterday we published a fix we refused to ship. One distinction was missing. Measured on the same 72 answers: 26 businesses found out of 28, against 23 - and none lost.
3 minmin read
Measured across 72 answers: 454 extracted names for 424 real businesses. The obvious fix absorbed 61 duplicates - and made 72 real shops disappear. We did not ship it.
4 minmin read
Measured 17 August 2026 across six markets. In Japanese and Korean, 100% of the names cited are in the local script - and our extractor required a Latin capital.
4 minmin read
Read from our own Search Console on 16 August 2026. Impressions went up tenfold in six days - and almost all of them came from people looking for something other than us.
6 minmin read
Measured on 15 August 2026 across six cities, in French, Portuguese and English. Re-asking in the same language returns 61% of the names; asking in another language, 20%.
3 minmin read
Measured on 13 and 14 August 2026 across six French cities. Perplexity keeps half its names from one day to the next, Gemini a third - and ChatGPT still names no local business.
3 minmin read
Measured on 13 August 2026 across six French cities. Of 42 businesses cited, none is named by all three assistants - and one of them names none.
4 minmin read
Measured on 12 August 2026 across four trades and four cities. The engine reads thirteen pages per answer - and almost never the site of the business it recommends.
3 minmin read
Measured on 11 August 2026 in two French cities. No business appears across all five phrasings - and on one of them, the AI names nobody at all.
3 minmin read
Measured this morning across four trades. The result forced us to fix our own score - an engine that did not search no longer votes « absent ».
3 minmin read
Measured 9 August 2026 on three very common French business names. With no city in the question, one engine elects a single establishment - and gives its address.
3 minmin read
Re-measured this morning: of 81 names AI assistants cited on 6 August, only 15 reappear on the 8th. What holds for 48 hours, what moves, and why.
4 minmin read
On 10 July 2025 Meta opened Instagram to search engines. Thirteen months on, it is the top source AI assistants cite in Brazil. A three-market investigation.
3 minmin read
Measured 6 August 2026 on plumbers in two French cities: one refusal, one from-memory list, one directory roundup. Three AIs, three answer regimes.
3 minmin read
We measured 24 Brazilian businesses in August 2026: 22 scored 0/100. One of them had won a client through AI the day before. Both things are true.
2 minmin read
More customers now ask ChatGPT for “the best X near me.” Here's how to become part of the answer - and why it isn't SEO.
1 minmin read
AEO (Answer Engine Optimization) isn't a new name for SEO. Here are the concrete differences and what they mean for your visibility.
1 minmin read
Eight concrete actions, ranked by effort, to get ChatGPT, Gemini and Perplexity recommending you. Do them this month.
