Sojern, internal pilot · 2026
Tidemark: an AI-visibility audit tool
Built and piloted it

For a hotel, an answer-engine result changes the question from “do we rank” to “does the model know we exist, and does it repeat us correctly”.
I built Tidemark, a front-end audit tool that scores a hotel domain on that visibility, covering how the domain is represented to answer engines and where the gaps sit, then used it to shape a GEO strategy across the hotel portfolio.
This is a pilot. I built it and ran it; the outcome so far is a repeatable way to ask the question, not a proven lift in a metric. GEO is new enough that a clean before-and-after number would probably be noise. What holds up is the method: structured data quality, entity consistency and crawlability are what an answer engine consumes, and all three are measurable today.
JavaScript · Technical SEO · Structured data · LLM APIs