CleanDims.

What we are.

CleanDims exists to make dimensional data management a discipline that organisations can actually adopt. The category has been called many things and managed reactively for thirty years. Our argument, developed at length in the manifesto, is that the cost of leaving it reactive is changing now that agents are starting to produce dimensional values at machine speed, and that the practice needs to be written down and built into infrastructure before the cost becomes intolerable.

The company is in its early stages. The team is small. The substance of what we believe is in the four foundational documents, and the substance of what we build is in the product. Everything else is downstream of those two.

How we work.

A few things are true about how we approach the work, which the website should make visible without our having to say them.

We build the discipline in the open. The manifesto, the primer, the target state, and the four-layer taxonomy of what goes wrong are all published. The library and the learning curriculum are open. The interactive diagnostic is open. The product is the implementation of the discipline these documents describe, and the discipline belongs to the category, not to any one product.

We write before we build. Every meaningful piece of work at CleanDims starts with a written argument that has to hold up to reading. The argument might be wrong; what we will not do is build something whose case cannot be made in writing.

We are interested in the work, not the noise. We are not chasing valuations, not chasing trends, not chasing the news cycle of the AI era. The dimensional data problem has existed for thirty years and will continue to exist for as long as organisations have categorical data in their systems. The work is durable; the work is what we are here to do.

Who we are.

CleanDims is a small team of practitioners with backgrounds in data engineering, data governance, and enterprise software. We met working on the dimensional data problem from different angles inside larger companies, and started CleanDims because the work we wanted to do required a company-shaped commitment rather than a project inside another organisation.

The team is distributed across timezones, working primarily async, with one shared writing repository and a handful of weekly synchronous meetings. We are not hiring at the moment. When that changes, openings will be listed on careers.

Where we are going.

We are building toward a world in which the categorical surface of every business system is governed at runtime: a canonical reference for every dimension that matters, named stewards holding the authority to decide, agents reading from synchronised caches and producing confidence-weighted output, consumers acting on the confidence signal with their own policies. That world is the subject of the target state document.

Most organisations are several stages away from that world today. The work of CleanDims is to make the path from here to there specific and reachable, both as a discipline that practitioners can adopt and as a product that does the load-bearing work.

We are early. The argument is clear. The product is being built. If any of this resonates with the work you are doing or the problem you are facing, get in touch or see the product.