Pricing
Every model runs in your browser, so there is no compute for us to meter and no per-seat maths to do. The free plan is the whole product on our data; Pro is the whole product on yours.
Free
£0forever
The complete workflow, on our sample datasets.
- Four sample datasets, fully modellable
- Feature engineering with adstock and saturation
- OLS and conjugate Bayesian regression
- Health scorecard and diagnostics
- Budget optimiser and forecasting
- Save and reopen sample projects
Questions people actually ask
- Where does my data actually go?
- Nowhere. The Python runtime is compiled to WebAssembly and runs in your tab, so the file you open is read by JavaScript on your own machine. Closing the tab is what deletes it — we never hold a copy.
- How does saving work if there is no server?
- You download a .beacon file containing your data and your model, and drop it back in to carry on. It is a real file you own, so backing it up or sharing it with a colleague is up to you.
- Why is there no MCMC?
- Full posterior sampling is too slow to run in a browser tab. Beacon Light uses ordinary least squares and conjugate Bayesian updating, which give you priors, uncertainty intervals and coefficient bounds in milliseconds rather than minutes.
- Is my model any good?
- The scorecard is there to tell you when it is not. It grades fit, residual behaviour, collinearity, coefficient signs, contribution plausibility, transform shape and whether you have enough data — and names the variables causing each problem.
- Can I cancel?
- Any time, from your account page. You keep access until the end of the period you have paid for, and your saved files keep working on the free plan for anything built from sample data.