Synthetic audience research · Travel first
See your project through the eyes of a points-savvy frequent flyer, a cautious first-time traveler, or a vacation planner skimming on their phone—and discover what clicks, what raises doubts, and where you might lose them.
First Points Question
Budget Newcomer
Weekend Escape Planner
Delta Apologist
Upgrade Snob
Two Point One Club
Surcharge Accountant
Bonvoyed Survivor
Fine Print Auditor
Airport Queue Reader
Keyboard Navigator
Large Text Reader
Google Snippet Arrival
AI Citation Checker
Contrarian Veteran
Hotel Shortlist Maker
Seat Selection Worrier
Avios Collector
KrisFlyer Premium Planner
First Card Applicant“Wait — does booking here still earn my status nights?”
Bonvoyed Survivor · Skeptical readers
Who pays for this ranking? I can't see a disclosure anywhere near it.
If I book through this link, do I still get my nights? That decides it for me.
The rate looked fine until the resort fee. What else am I not being told?
Will applying hurt my score? Nothing here tells me if I'd even get approved.
The audience
Fifty traveler personas in twelve behavioral families, each with a persistent identity, a designed profile and a research history. A loyalty expert, a first-time booker and someone burned by hotel promises read the same page very differently — that disagreement is the point.
Airline loyalty defender
Loyalty basics explorer
Benefit-exception investigator
First-time visitor to the US
Use-it-now redeemer
Cautious first-time applicant
Trait meters show designed persona DNA, not observed behavior.
How a project works
Start from what you want to learn — “Can a first-time visitor trust this verdict?” — not from a test script.
Pick personas, or let Gate83 assemble a contrasting panel — one from each of the twelve families.
Reader modes add the situation: a Google drop-in, a daily regular, a commenter.
Each persona works through your real pages in a real browser. Every step and screenshot is kept.
Where they agreed, where they split, and fixes — each linked to the evidence it rests on.
Gate83 Research Report
Reports lead with the finding and the disagreement behind it, and keep three things apart: what the browser recorded, how each persona read it, and inferences that still need checking. No score appears without verified evidence.
Read the sample reportFinding 01 · High severity
The verdict persuades experts and loses the people it's meant for.
Who trusted the hotel verdict?
What's ready today
Audience library, profiles, panels and reader modes
Projects and recorded journeys today; live browser execution is being connected
Consolidated reports from recorded journeys
Follow-up projects and comparisons; trend tracking in development
Concept projects can be planned; automated evaluation is in development
Matched comparisons of your own projects are available
Travel first, built to expand
Gate83 starts in travel, loyalty and travel publishing — where expertise, price sensitivity and trust vary the most.
Can first-time visitors find a hotel they would actually book, and understand why it was recommended?
Do members understand what they gain or lose in elite benefits by booking this way?
Can travelers find award availability and understand the true cost including surcharges?
Does our redemption guidance feel credible to people who already optimize points?
Do ads and affiliate modules interfere with reading and trust on our articles?
Can readers distinguish an independent card review from a paid recommendation?
Can a traveler quickly confirm whether they can get into a specific lounge?
Where do families planning a trip lose confidence before committing?
Does our content work for readers outside the US, and for visitors coming to the US?
In short
Gate83 is synthetic audience research for travel websites. Fifty traveler personas, each with a stable identity and a designed behavioral profile, work through your real website in a browser. Every step is recorded with screenshots, and the report separates what was observed from what we infer — so you see what clicks, what raises doubts and where you lose people. A project is $500 per site, with up to 5 research questions, 12 personas and 3 runs each.