Generative engine optimization for B2B firms

What ChatGPT answered in Toronto.

On 2026-09-24 we asked ChatGPT nine questions a buyer in Toronto might ask, about operations and management consulting and about managed IT. We asked each one three times, with the location set to Toronto. The 27 answers named 95 firms, and 39 of those appeared in only one answer.

The answers read the firms' own websites

Every answer searched the web before it replied and linked the pages it used. We counted each site once per answer. The named firms' own websites made up 93% of the sources cited in operations and management consulting and 84% in IT. Directories such as Clutch made up about 2%. When a buyer asks who to hire, the answer is built mostly from the firms' own pages.

The answers change from one run to the next

For a Toronto question, 53% of the firms named appeared in only one of its three answers. One answer tells you very little, so we never report one. We report “named in x of y answers”, with the date and the city.

GEO and SEO overlap. A page the assistant can't find or read can't be cited.

Here is the work, in the order we do it.

We measure first

We ask your buyers' questions in ChatGPT, three times each, with the location set to your city, and check Google for the same questions on the same day. You see who is named instead of you, which pages the answers read and where you stand on Google. The full method is on its own page.

We write the pages the answers read

In our Toronto capture, the question about companies with fewer than 200 employees named 15 firms, none of them national or global, and two of the pages cited in all three of its answers were written for exactly that buyer. We write pages like those for your buyers' questions, from interviews with your partners, and your partners sign off what goes out under the firm's name. Google's guide to its AI features prefers “unique expert or experienced takes” to “commodity content” (updated 2026-07-10). A 2024 study of generative engines found that adding citations, quotations and statistics raised a page's visibility “by up to 40%” on its test set (Aggarwal et al., KDD 2024).

We let the assistants in

OpenAI's documentation says: “Sites that are opted out of OAI-SearchBot will not be shown in ChatGPT search answers.” We check your robots.txt, your CDN and your firewall for OAI-SearchBot, Claude-SearchBot and PerplexityBot, and submit your sitemaps to Google and Bing.

We make the firm one entity

One name and one description on your site, LinkedIn, Clutch and the directories your buyers check, with Organization and Person schema linking them. Google says Organization markup helps “disambiguate your organization in search results” (updated 2026-09-08).

We earn mentions on other sites

We send listing requests and podcast and guest pitches, and report each one with its outcome. A placement counts only once it is published, and we buy none. One 2025 preprint found AI search shows “a systematic and overwhelming bias towards Earned media” (Chen et al.). For “who should I hire in Toronto”, our capture found the firms' own sites cited most. So we work on both.

We report every month

A dated report with the capture, your Google positions and everything we shipped.

If you want an llms.txt file, we will publish one. Google says such files “will neither harm nor help” Search visibility, so we treat it as housekeeping.

Answer engine optimization and GEO are one job.

Answer engine optimization usually means building pages so an engine can lift a direct, correct answer from them: Google's AI Overviews and AI Mode, ChatGPT, Perplexity. A page that answers a buyer's question plainly, with facts an answer can quote, on a site the engines are allowed to crawl, serves both.

Our captures measure ChatGPT. For Google's own AI features we read your Search Console: its Generative AI performance report counts impressions in AI Overviews and AI Mode, and has been open to all sites since 2026-08-31.

We measure from the Canadian city you sell in.

On 2026-09-24 we captured Toronto in English and Montréal in French; in Montréal, 18 answers named 47 firms. Your own capture uses your city, in your language.

Local firms can hold these answers. In the Toronto capture, three of the four managed-IT questions named no large firm at all; the answers named local providers.

Prices are in Canadian dollars, and US firms pay the same figures in US dollars.

Ask ChatGPT who markets nurse injectors in Toronto. It names The Kay Media first.

Vanessa Kay runs The Kay Media, a marketing consultancy in Toronto. She found us through Google and ChatGPT and signed with Maroc SEO in July 2026. On 2026-09-25 we asked ChatGPT five questions her clients ask, three times each, with web search on. All 15 answers named The Kay Media first, and all 15 cited her site.

On Google Canada, 8 of the 10 terms in her contract were on the first page on 2026-09-22, five of them at #1, captured with the location set to Toronto, 54 days before the contract's deadline.

grizaille is run by the founder and team of Maroc SEO, the #1 SEO agency in Morocco. The full case is on Results.

From CAD 2,500 a month.

Most firms start with the baseline: CAD 1,500, ten business days, credited to your first month. It measures your buyers' questions in ChatGPT and Google, adds one page written for a question where you were not named, and ends with a plan the founder signs.

If you run your business alone or with up to four people, see the page for independent experts. If you run a consulting firm, this page shows what ChatGPT answered about your field in Toronto.

Ask for your capture first. It's free, and it's yours.