Company · FAQ
Straight answers to the questions operators actually ask
What we sell, what happens to your data, how we keep AI from inventing a trail, what an engagement looks like, and how much of your team's time it takes. If your question isn't here, ask us directly.
Questions
The platform
What does Nova Adventure actually sell?
Two things that work together. A modular platform of six modules (Itinerary Studio, Route Engine, Booking Flow, Signals, Concierge and Agent Gateway), and the services needed to make them work in a real operation: strategy, data engineering, custom development, disruption response, managed AI operations and governance.
Do we have to use all six modules?
No. Each module works on its own, and most clients start with one. Because they share a single data layer, adding a second later builds on the first instead of starting over.
Does Nova replace our reservation system or PMS?
No. We sit alongside your systems of record and connect to them through standard REST APIs, webhooks, CSV/SFTP and the Model Context Protocol (MCP). Your reservation system stays the source of truth for bookings.
Which systems can you integrate with?
We work by category: property management systems, channel managers, reservation systems, GDS and bedbank feeds, payment providers, CRMs, messaging channels, mapping and transit data, weather data and data warehouses. The integrations page explains the patterns and how we scope each connection.
What is Agent Gateway, in plain terms?
It's your own front door for AI assistants. Agent Gateway is an MCP server that lets assistants discover your inventory, check live availability and book through an agent-safe checkout, under your policies, at your margins, and with attribution you can see.
Which channels does Concierge work on?
Web chat, WhatsApp, SMS and email. It answers from your verified content, re-plans in-trip when something changes, and hands off to your staff with the full conversation when a person should take over.
Is it for our staff or for travelers?
Both, depending on the module. Itinerary Studio and Booking Flow are mostly staff-facing. Concierge and Agent Gateway face travelers and their assistants. Route Engine and Signals work underneath everything.
Questions
Data and privacy
Who owns our data?
You do. Your business data and your travelers' data remain yours under the customer agreement. We process it to deliver the service you've engaged us for, and for nothing else.
Do you use our data to train AI models?
We don't use your data to train models that serve other customers. We configure the model providers we use so your data isn't used to train their general models, and we document that configuration for your review.
How do you handle sensitive traveler details like mobility, diet or health?
We collect only what a trip actually needs, ask for consent where it's required, tag sensitive fields so they're handled separately, and limit who and what can see them. Our Responsible AI & Traveler Data Governance service covers this in depth.
How is our data secured?
We follow least-privilege access, encrypt data in transit and at rest, require strong authentication for staff, log access to sensitive data, and review every integration for what it exposes. The security and trust page describes our practices.
Where is our data stored?
Storage regions depend on the deployment and are agreed in your contract. If you have residency requirements, raise them on the discovery call and we'll design around them.
Can you complete our security questionnaire?
Yes. We'll answer your questionnaire in detail and walk your security team through our controls, our subprocessors and our incident-response process.
What happens to our data if we leave?
You can export it in standard formats. After the agreed transition period, we delete it from our systems and confirm in writing, except where the law requires us to keep specific records.
Questions
AI safety
How do you stop the AI from inventing venues or trails?
Grounding. Our modules recommend only from your verified catalog and approved data sources, check those against live hours and closures, and decline rather than guess when they can't verify something. We wrote about the details in Phantom Trails and Closed Cafés.
Can the AI book or spend money on its own?
Only within limits you set. By default, anything that spends money, changes a traveler's plans or makes a new kind of commitment is drafted for a person to approve. You can relax that for specific low-risk workflows once the system has earned it.
What does Concierge do when it doesn't know the answer?
It says so and hands off to your staff with the conversation and context attached, instead of improvising. Knowing when a concierge should stay quiet is part of the design.
How do you test an itinerary generator before travelers see it?
With evaluation sets drawn from your real trips, including the awkward ones, reviewed by people from travel operations. Nothing ships until it passes. Our approach is laid out in How We Evaluate an Itinerary Generator.
How do you check for bias or unfair outcomes?
We review outputs across traveler types, such as travelers with mobility needs, families, solo travelers and different budgets, and look for patterns like systematically worse routes or fewer options. Findings and fixes are documented and shared with you.
Can we switch an AI feature off quickly?
Yes. Every AI feature has an off switch and a documented fallback, so your team can return to the manual process in minutes if something looks wrong.
Questions
Engagements
Why don't you publish prices?
Because scope depends heavily on your systems, data and channels, and a published number would be wrong for most readers. Every engagement is scoped and quoted in writing after a discovery call. The engagement models page explains what drives scope.
How long does a Discovery Sprint take?
Three to four weeks. It ends with a baseline, a ranked opportunity map, one pilot design and a written quote for next steps.
What's the difference between a Pilot and a subscription?
A Pilot usually runs one or two modules on live trips for 8 to 12 weeks and ends in a go/no-go review. A platform subscription is the annual, modular license you move to once a pilot has earned it.
Are you a good fit for a small operator?
Often, yes. Many of the businesses we work with are small teams. We'll tell you on the call if a lighter starting point, or no project at all, makes more sense.
Will you sign an NDA before we share details?
Yes. We're happy to sign a mutual NDA before you share anything sensitive.
Questions
Implementation
How long until something is live?
Most clients see a module in live use during the pilot, which runs 8 to 12 weeks. Timelines depend mostly on integrations and the state of your data, and the written scope will say which.
What do you need from our team?
An operations owner with a few hours a week, access to the relevant systems or data samples, and people willing to review drafts and tell us bluntly what's wrong.
Our supplier data is a mess. Is that a blocker?
No. It's where nearly everyone starts. Cleaning and mapping supplier data is the first phase of our approach and the core of our data engineering work.
Do you support languages other than English?
Yes. Concierge and Itinerary Studio can work in multiple languages. We scope each language explicitly and test it with native reviewers before launch.
What happens after go-live?
Platform customers get support with severity-based response targets, described on the support page. If you want us to keep evaluating and tuning the system in production, that's Managed AI Operations.
Can our own developers build on Nova?
Yes. We offer APIs, webhooks and MCP interfaces, with sandbox access by request. The developers page covers authentication and how to get started.
Related
Keep exploring
- CompanyEngagement modelsHow we engage: Discovery Sprint, Pilot, Platform subscription, Build partnership and Managed AI operations, each scoped and quoted after a discovery call.
- PlatformSecurity & trustHow Nova protects traveler data and keeps AI grounded: encryption, least-privilege access, prompt-injection defenses, PII minimization and incident response.
- CompanyOur approachHow Nova Adventure implements travel AI: operational truth first, then routing, prediction and automation, with people approving every decision that matters.
- CompanySupportSupport for Nova Adventure platform customers: email support@novadventure.com or call (505) 303-9060. Severity levels, response targets and business hours.
Didn't see your question?
Ask us directly. We answer every message within one business day, and the discovery call is a good place for the harder questions.