We start with the questions, not the rankings
At SaidTrue, an AI Visibility Audit starts with the conversation your customers are already having without you. We ask ChatGPT, Gemini, Perplexity and Claude the questions customers already ask — then we show what was said, what was sourced, and where the truth diverges. We do not open with a keyword list or a search position, because neither tells us what an engine will say when someone asks who to use or whether a business is reputable.
Then we read the answers closely
What we are looking for is how AI systems describe, recommend and sometimes misread businesses. So we work through each answer on three levels. First, the description itself: whether the engine identifies the business confidently and gets the basics right. Second, the sourcing: what the answer rests on, and whether that is the business's own material or somebody else's account of it. Third, the divergence — the distance between what was said and what is actually true. That gap is the finding. A fluent, confident answer can still be wrong, and a business can be summarised in terms it would never use for itself.
We score what came back, so the picture holds still
Impressions do not survive contact with a second question. That is why we score the answers rather than simply reading them, and why How SaidTrue Scores AI Answers is published as part of what we do. Scoring turns a vague sense that the AI "seems to like us" into a visibility report on a website's presence in AI search — something that can be compared with the next round of answers instead of dissolving into anecdote.
And we keep going, because the answers move
An answer is generated, not filed. The same question can come back differently as sources change and models are updated, which is why our work continues as monthly AI monitoring alongside a verified AI profile. The point of the ongoing cycle is that the description an engine gives of a business keeps being held against the facts, rather than being audited once and assumed settled.
Why we sequence it this way
Google isn't dying, but it's no longer the only front door. Customers increasingly ask AI who to use, whether a business is reputable, and what it actually does — and they act on the reply. Beginning with their questions, reading the answers and their sources, scoring the result and then watching it over time follows the order in which AI visibility is actually created and lost. The audit is the evidence; everything after it is the work.