We got tired of paying for four tools and still not knowing what to do Monday morning.

That was the actual starting point. Every SEO and AI-visibility tool on the market is happy to hand you a dashboard full of scores and charts. Almost none of them tell you what the numbers mean, or which of the ten problems on the page is actually worth fixing first. So the job of running a site's visibility quietly turns into a second job: hours a week spent interpreting tools instead of acting on them, split across three or four separate subscriptions that don't talk to each other.

LetsGetSurfaced started as an attempt to fix that for one team, and turned into a product once it was clear the same problem was everywhere: marketing managers, fractional CMOs and agencies all doing the same manual translation work, every week, by hand.

Along the way, the way we actually test AI visibility changed a lot from the original plan -- and it's worth being specific about why, because it's the part most tools in this space skip. It would have been far easier to just ask an AI model to “look at a website” and summarise what it finds. It would also have been close to worthless: a shallow, one-shot glance at a handful of pages, a different answer every time you ask, recommendations that sound plausible but don't point at an actual cause. That's not an audit, it's a hunch with good formatting.

So instead, every LetsGetSurfaced report is built on an actual crawl of your entire site -- sitemap discovery, full page indexing, the same starting point a search engine uses -- before any AI model ever gets asked a question. The AI testing happens on top of that real inventory, not instead of it. It's slower to build than the shortcut version. It's also the difference between a suggestion that's a hit-or-miss guess and one that's grounded in the crux of what's actually on your pages.