When an assistant gives false information about your business
There is no button to correct an assistant's answer. Nothing equivalent to editing a business listing. The only route that works is to correct the sources these systems read, starting with the ones they favour, then wait for the next pass.
The problem is more common than anyone admits. An assistant states stale opening hours, a price from two years ago, a service you dropped, sometimes the address of a former location. The customer does not verify. They do not call either, because they already have their answer.
The cause lies in how these systems are built. They restate what the web says about you, and a large share of what they know was absorbed during earlier passes rather than fetched at the moment of the question. Correcting your site today does not rewrite what has already been learned. It only changes what will be read again.
Google itself lists keeping your Business Profile and your merchant information up to date among the good practices that apply to its generative answers. It is the only correction lever the official documentation names, and it is free.
The order of steps that works
Your own pages first, because they are the only ones you fully control. Information contradicted elsewhere has to be stated without ambiguity on your site, in text, not in an image or a table loaded by a script.
Then your official listings, on the major directories and marketplaces. Then sector directories and third-party pages still carrying the old information, in order of their visibility.
Finally, patience. The delay depends on how often each source is read again, and that frequency is not yours. Count in weeks, not days.
The two-minute check
- Ask two different assistants the five questions your customers ask before buying, and note what is wrong.
- For each error, find the page still carrying it, correct it, and run the test again a month later.
An error that persists after your site is corrected comes from elsewhere. Look for the third-party source before concluding it is an algorithm problem.
That audit, page by page and source by source, is part of what we cover in a paid audit.
An answer varies from one session to another and from one account to another. A single screenshot proves nothing, in either direction. Test several times, over several days, before drawing a conclusion or paying for a correction.
What each platform makes measurable is covered in our article on optimizing for answer engines, and the B2B buyer case in AI-assisted shortlisting.
- Google Search Central, AI features and your website, official documentation, updated December 2025, accessed October 2026. Mention of keeping Business Profile and merchant information current.

Geneviève puts the strategy for your engagement into action. She leads all our web development projects: Shopify, WordPress and the new ways of building a site with AI. She manages our team of developers and translates your business needs into technical language. She runs your organic search (SEO), your visibility in AI answers (GEO) and your site's conversion rate optimization (CRO). Her work is at the heart of three goals: Attract customers with SEO and AI, Improve your site's conversion, and Strengthen your visibility in AI answers. With Gabriel, she also builds the landing pages for your advertising campaigns. She writes mainly about SEO, AI visibility and web design.
About Falia →