Site Readability
Can AI assistants read, parse, and cite your destination's web presence? Structure, schema, crawlability, and freshness across the properties your team stewards.
The platform
When AI recommends a place, make sure it's yours.
DAVI™, the Destination AI Visibility Index, scores your destination 0 to 100 every month across two layers, then reads that score against your brand pillars, your themes, and the sources AI leans on.
Two layers
AI Perception
What the assistants actually say when a visitor asks. Sentiment, accuracy, specificity, and whether your destination is named at all.
Why one number
Simple enough for a board. Deep enough for scrutiny.
A board meets for two hours and has a dozen items on the agenda. One number, moving in one direction, against a published rubric, is what earns AI a place on that agenda. It is a headline, not the whole story. Underneath, DAVI™ opens all the way down to the verbatim answers, so anyone who wants to challenge the number can, line by line.
0
DAVI™ composite
Layer 171 / 100
Can AI read you
Site structure, crawlability, schema, page speed, content depth, freshness, and the technical signals that decide whether an assistant can use your site at all.
- Technical readability
- Schema and structured data
- Content depth by topic
- Freshness and update cadence
Five levels under the headline
0 to 100. The number the board sees, tracked month over month against the same rubric.
78 this month, up from 74. One line on a board deck, one direction, one rubric that does not move.
What you see at this level
- DAVI 78
- Prior month 74
- Band: Established
Sample figures shown. Your board sees your own.
How we measure
Asked until the number holds.
The same rubric, the same six assistants, every month. We publish the method because a score you cannot interrogate is a score no board should spend against.
20,000+
Prompt library
Traveler-based queries written in the language visitors actually use, built for destinations and drawn from every stage of the trip decision.
6
Assistants, every run
ChatGPT, Claude, Gemini, Perplexity, Grok, and Google AI Overviews. The same question, put to all six.
1,800+
Queries per destination
The prompt set localized to your markets, your comp set, and your pillars, then run against each model.
95%
Confidence level
Each question is asked over and over until the variance tightens and the result holds at 95 percent confidence.
LLMs are non-deterministic. That is the whole reason for the rigor.
Ask an assistant the same question twice and you can get two different answers. A single screenshot proves nothing. So we ask each question repeatedly across all six models and keep sampling until the result reaches a 95 percent confidence level, then report the margin alongside the score. Every figure links back to the verbatim answers behind it, so your team can check the work rather than take our word for it.
DAVI™ Score, Destinationville
Destination AI Visibility Index
786 pts this monthEstablished
Needs attentionDevelopingEstablishedLeading
Built from captured visitor answers each month across six AI assistants.
Assistants now name Destinationville in most trip-planning answers for the region, and cite your own pages more often than third-party listicles. Partner coverage is the thinnest part of the picture this month.
Where the score comes from
AI perception81
Website readiness74
Partner ecosystem68
Sample destination. Your board runs monthly in the Zertura Platform.
See it live in a demoFour modules, one story
Signal is what AI says. Verify is where that came from. Track is what moved. Direction is the counsel your team acts on from there.
Signal
What AI says about youBrand pillars
Your pillars go into the platform, then we measure whether AI associates your destination with each one. Where a pillar is claimed but absent from the answers, you see it.
Perception and visibility
Sentiment, accuracy, and how often you are named across ChatGPT, Claude, Gemini, and Perplexity, each question asked repeatedly so variance is visible.
The answers, verbatim
Read the raw responses behind every figure. Nothing is a black box, so you can defend the number line by line.
Reviews and sentiment
What visitors and residents are saying in the places AI reads back, alongside the visitor journey from first question to booking.
Verify
Where the answer comes fromCitations and sources
Which publications and pages the assistants pull from when they answer about your destination, and where you are absent from the sources that get cited.
Themes
What is most talked about in your category, what your destination is absent from, and which themes are worth a season of content.
Pillar alignment
Whether the sources being cited actually carry your brand pillars, or someone else's version of your story.
External and earned
Backlink authority, referring publications, top anchors, and link sources. The earned footprint AI reads before it describes you.
Track
What moves, and what heldVideo search
YouTube and TikTok results for the questions visitors ask. Whether your destination is named in the title, description, hashtags, or transcript, and which creators own the subject.
Campaign lift
What a seasonal campaign or event did to how AI answers about you, and what held once the campaign ended.
Traditional search
Rank tracking, local pack and geo-grid, the website score, and GSC and GA4 outcomes with the AI referral split called out.
Markets and queries
How the answer changes by feeder market, season, and visitor intent, query by query.
Direction
What to do with what we foundContent direction
What to write about, the angles, working titles, and a brief per theme. Included, no separate charge. Your team writes and publishes, so the voice stays yours.
Partner scorecards
Each enrolled partner scored on the same rubric across lodging, attractions, food and beverage, retail, and events, with priorities ranked for them.
Board reporting
A written monthly readout you can hand to a board, against a published rubric, including the months the number holds steady.
Ask DAVI
Ask questions about your score in plain language and get the next move, with the sources behind it.
Go deeper
See your own answers before you decide anything.
Thirty minutes, your destination run through the panel, and an honest read on where your team's effort belongs first.
Book a demo