Customers · Backers · Partners
Forecasts that explain themselves
Every TUBR prediction comes with the causes that drove it. We call it CauseCasting: no black box, no "trust the model" — your customers see why the forecast says what it says, and they act on it.
A prediction layer for your platform
Embed our engine to give your customers prediction superpowers — demand, footfall, occupancy, staffing — without standing up a data science team. You keep the relationship; we deliver the forecasts. Powered by our proprietary Physics-Informed AI, which reasons about cause and effect from a fraction of the data conventional AI needs.
Scalability with one API, a couple of clicks
One engine, many outlets. TUBR is architected to scale across hundreds of locations from a single integration — each outlet getting predictions tuned to its own context, without bespoke setup per site. Onboarding a new outlet is an "on switch", not a project.
Industries
Proven across the hardest forecasting problems we could find.
Our engine has predicted train delays for Eurostar, rail operations for Network Rail, product demand in salons, and occupancy in hotels. Radically different problem shapes, one engine. The work generalises. Right now, we are going deep on hospitality. Two ways:
Tubr Live is our direct product for independent restaurants, cafés and bars — the engine in production, serving operators who would never have a data science team of their own. See tubr.live.
Platform partners carry the engine to the multi-site operators and chains they already serve. If that is you, let us talk.
Case studies
What we built, where it ran, what it changed.
In their words
“We worked with TUBR on an exciting new project: could we predict the arrival of our trains at their destination? We chose TUBR for many reasons. We wanted an innovative start-up. We wanted a partner who could match our drive and ambition. We wanted a company who could clearly add value to Eurostar. TUBR delivered all of this. The team were able to understand our needs quickly. They delivered a great product, really quickly. They were great at communicating, and at reassuring our frontline. They're a great partner. They absolutely delivered what we needed!”
Neil Powling, Eurostar Data Operations Manager
23 October 2025
Why it's different
Three claims you can stand behind.
Causation over correlation.
Our proprietary methodology reasons about cause and effect rather than matching surface patterns. That is what lets it generalise to new conditions instead of confidently repeating old mistakes. We call the output CauseCasting.
Works on sparse data.
Our engine needs roughly 20% of the training data conventional AI approaches require. New outlets and new sites get useful predictions in weeks, not after a year of data collection.
Built to be embedded.
It is an engine, not a dashboard you have to live in. It is designed to feed predictions into the tools your users already work in.
For platforms
A predictive plug-in for the platforms your customers already use.
Most operators don't want another login. So TUBR is built to drop into the platforms they already use. You bring the platform and the customer relationship. We bring the prediction layer — demand, footfall and operational forecasts delivered through our API and surfaced natively inside your product. Your customers get predictive intelligence without changing how they work; you get a differentiated feature without building a forecasting team.