We are interviewing Ioan Istrate, the creator behind Tripvento, a B2B hotel ranking API that uses geospatial intelligence and semantic AI to score properties based on traveler intent. Tripvento bypasses traditional pay to play commission models, providing an unbiased data layer for the next generation of travel platforms.
What does Tripvento do? And why is now the time for it to exist?
Tripvento is a B2B hotel ranking API that scores 35,800+ hotels across 311 destinations using geospatial intelligence, semantic AI, and 14 traveler intent signals — so travel platforms show the right hotel, not the highest commission one. Online hotel rankings are pay to play; Tripvento replaces that with transparent, data driven rankings built for OTAs, TMCs, and AI travel agents. Sub 250ms API responses, MCP server live, secure with zero public ports and multi tier api key architecture. Now’s a good time for Tripvento to exist because travel planning is rapidly shifting from traditional search towards AI agents, which require non-hallucinating, structured, and unbiased ranking layers to function effectively.
What is your traction to date? How many people does Tripvento reach?
450+ monthly unique visitors · 35,800+ hotels scored · 311 destinations · 41 countries · 2 enterprise TMCs in sandbox evaluation (Fortune 500 travel management firm + major Japanese travel operator) · API serving sub 250ms responses globally
Who does your Tripvento serve? What’s exciting about your users and customers?
OTAs, corporate TMCs, and AI travel agents that need a neutral, intent-based hotel ranking layer. Notable sandbox users include a Fortune 500 commercial real estate and travel management firm and a major Japanese logistics and travel operator — both inbound, zero outbound sales. Also hotel operators (via self ranking POST endpoint launching April 2026) who want to understand how their property scores against 14 traveler personas.
What technologies were used in the making of Tripvento? And why did you choose ones most essential to your techstack?
Tripvento relies on a high performance backend using Python, Django, and PostgreSQL, supercharged by PostGIS and pgvector to handle complex geospatial and semantic searches efficiently. To guarantee our sub-250ms API response times across the globe, we implemented Redis and Celery for seamless background processing, while keeping our infrastructure highly secure via DigitalOcean VPCs and Cloudflare Tunnels. Finally, Next.js drives our frontend, and an active MCP server ensures our product can be natively consumed by AI travel agents.
What is traction to date for Tripvento? Around the web, who’s been noticing?
Tripvento is capturing strong organic search visibility, ranking as the #1 organic Google result for highly competitive terms like "hotel ranking api" and "intent based hotel api." Beyond gaining media coverage through AP News wire releases, we are currently live in 311 destinations across 41 countries and are openly documenting our journey via a build in public blog.
Tripvento scored a 68.61 proof of usefulness score (https://proofofusefulness.com/report/tripvento) - how do you feel about that? Needs reassessed or just right?
Accurately reflects where we are. The low traction scores are honest — we're pre revenue with two sandbox users. The high utility and innovation scores reflect what we've actually built. I'd expect the score to jump significantly after first paying customer and when the POST endpoint ships in April. Re score us in 6 months.
What excites you about this Tripvento's potential usefulness? *
Hotel ranking is a $700B+ market running on commission bias — every major OTA publicly admits their rankings are pay to play. Tripvento is the first API native, intent based ranking layer built on geospatial intelligence rather than advertiser relationships. The same infrastructure that scores a boutique hotel in Savannah for a bachelorette trip scores one in Tokyo for a remote worker, in under 250ms, automatically. As travel planning shifts from search to AI agents, a non hallucinating structured ranking layer becomes critical infrastructure. That's what we're building — and the MCP server is already live for agents to query today.
Walk us through your most concrete evidence of usefulness. Not vanity metrics or projections - what's the one data point that proves people genuinely need what you've built?
Two enterprise TMCs found Tripvento independently via organic search, within 24 hours of each other, before a single outbound sales call was made. That's the signal. Not a friend, not a warm intro — cold inbound from companies with real procurement budgets, driven purely by the product ranking for the right keywords.
How do you measure genuine user adoption versus "tourists" who sign up but never return? What's your retention story?
On the B2B side, retention is binary — sandbox users either make API calls or they don't. Both current TMCs have made requests, which separates them from signups who never engage. On the pSEO side, GA4 shows 2m 12s average engagement time, which is well above content site benchmarks. People landing on intent ranked hotel pages are reading, not bouncing.
If we re-score your project in 12 months, which criterion will show the biggest improvement, and what are you doing right now to make that happen?
Evidence of traction — by a wide margin. Right now we have visibility but no revenue. The B2C POST endpoint launches April 2026, creating a self serve path to first dollar. Within 12 months I expect paying customers across both B2B and B2C channels, which directly addresses every low score on the traction side.
How Did You Hear About HackerNoon? Share With Us About Your Experience With HackerNoon.
I've heard about HackerNoon on search engines by looking for syndication for our technical #build-in-public articles.
With two enterprise TMCs—including a Fortune 500 firm—currently running sandbox evaluations, what are the key milestones required to convert these pilots into full commercial deployments?
Three milestones: (1) sandbox API call volume increases — we're monitoring this weekly, (2) a product review call where we walk through ranking methodology and answer technical questions, (3) a commercial proposal aligned to their volume tier. The constraint is procurement cycle length, not product readiness. The B2C POST endpoint launching in April creates parallel revenue while these deals move through their process.
As you scale your geographic coverage from 311 to 400 destinations, how is your infrastructure evolving to maintain those sub-250ms API response times globally?
Response times don't degrade with destination count because computation happens at ingestion, not at query time. PostGIS spatial relationships are precomputed when a city is added — the query layer just reads indexed results. New destinations deploy in 20 minutes via automated pipeline. Scaling from 311 to 400 is a data operation, not an infrastructure operation.
Given the shift toward AI travel agents, how do you continuously validate that your 14 traveler intent signals accurately predict real-world traveler needs better than legacy OTA reviews?
Three feedback loops: (1) pSEO CTR data — when travelers click intent ranked results, that's implicit validation, (2) enterprise sandbox feedback — both TMC users have engaged with specific scoring questions, which surfaces signal gaps, (3) the POST endpoint creates explicit feedback — hotel operators who disagree with their score tell us exactly why, which is training data. The signals aren't static — they're weighted and will be recalibrated as real usage data accumulates.