The short version
A business intelligence platform built around answers
FactSmith lets the people who run a business ask questions of their own data, in plain language, and get answers they can trust, without building a dashboard or waiting for an analyst.
Who it is for
Distribution, multi-store retail, manufacturing, and franchise networks with real operational data -- and no dedicated analytics team.
The people who know the business best currently spend their week producing numbers rather than acting on them. FactSmith is for them. It is not for organisations that already run specialist BI teams.
Typical questions: which customers are buying less; which stores are underperforming and why; which products are most profitable; which sites need attention this week.
If those questions today mean spreadsheets, waiting for an analyst, or reconciling two reports that should have matched, you are in the right place.
Two halves of one job
Early development
Easy Data
Gets your data in, and gives it agreed business meaning.
Available now
Ask
Gets answers back out, in plain language, for the people who need them.
The platform
Holds it together: who may see what, what things cost, what has been confirmed true.
Ask is what is available now. Easy Data is being built. The two are halves of one job, and they are not the same stage.
A minute in the product
A spoken question, a real tile, then a clarification when the date is ambiguous. Wide World Importers demo.
What you do with it
Ask a question the way you would say it. FactSmith returns an answer as a tile -- a chart, the numbers, and a record of how it was produced. A question that asks for more than one view lands more than one tile.
You can ask why it looks the way it does. You can save the answers worth keeping, group them into a dashboard, and have an administrator vouch for the ones that must be true. That voucher is what turns an answer into a fact.
Where data becomes fact
Between your systems and your answer sits a layer that knows what your business means by revenue, customer and on time. That is what makes an answer trustworthy rather than merely fast.
Why you can trust it
Permissions are enforced beneath the AI, not by it. The model writes the question; it never decides what you are allowed to see.
A regional manager asking about revenue gets their region, because the rest of the data never reaches the model. A figure you may not see says so plainly. It does not invent a number to fill the hole.
It runs where your data already lives
One company, one source database, one deployment. Not a shared cloud service holding everybody's numbers together. An evaluation copy is a folder and an installer. It needs Docker Desktop and nothing else, and it works before you give it an AI key.
Connecting a warehouse is guided
The first session is not a blank page.
FactSmith looks at which parts of your warehouse have never been asked about, proposes questions in ordinary business language, and checks each one actually runs and comes back with rows. An administrator then accepts or rejects them one at a time, with the query shown. Only the accepted ones count when we score whether the product is getting the numbers right.
What it is not
It is not a dashboard builder that happens to have a chat box. It is not a general chatbot pointed at a database. And it is not a shared platform. One company, one deployment.
See it on your own data
A short session on your data says more than any page can.