Frequently asked questions
Short answers to the things people worry about before they upload data.
What formats do you support?
We currently support CSV and Excel files for direct upload. For databases or APIs, we usually start with exports and then discuss more direct integrations once we understand your setup.
Who owns the data and models?
You own your data and the custom outputs we build for you. We don’t resell or share your raw datasets. Any internal improvements we make use anonymised learnings only.
How long do you keep our data?
By default, only for the duration of the engagement. We can agree on shorter retention windows for sensitive projects or longer-term storage if you want recurring analysis.
Can you sign an NDA or DPA?
Yes. For most client work we’re happy to use your standard NDA / data processing agreement or provide a template you can review.
How long does a typical project take?
For a single dataset and a clear question, initial results usually take days, not months. Larger or messier setups can take longer, but we’ll give you a realistic timeline before you upload anything.
What if our data is messy or incomplete?
That’s normal. Part of the job is profiling your data, highlighting issues and showing what’s still possible. We’ll be clear about limitations instead of pretending the numbers are perfect.
Do you use our data to train generic models for other clients?
No. Any models trained for your project are scoped to your data and your problem. We don’t build a giant shared model out of everyone’s raw data.
What’s the best way to get started?
Start small: one dataset, one key question. Use the contact form to describe what you have and what you want to know, and we’ll suggest a minimal first project.