Drowning in meetings: spending months defining a data model
Elliot Rock · 4 November 2026 · 1 min read
Years ago I worked with a solution architect during the merger of two very recognisable wagering companies. He and a data scientist were modelling how to bring two customer wallet systems together. The modelling and transformations took them 18 months.
Most of that time went on working out what the data meant and getting people to agree on it.
If you’ve worked on a data model, you’ll know the pattern:
- A workshop to agree what a “customer” is
- Another one because finance and operations count “orders” differently
- Someone maps the relationships in a diagramming tool, and it’s out of date within weeks
- A change upstream, and the cycle starts again
The meetings happen because everyone starts from a blank page. Each person brings their own understanding of the source systems, and the room spends weeks reconciling them.
That's a big part of why I started building Neuroplastiq. If the work of reading the source systems and proposing a first model happens up front, with the evidence behind each mapping visible, the conversation changes. People react to a draft and correct it, which takes far less time than building from nothing.
Good data engineers have better things to do with their brains than months of workshops.
How long did your last data model take to agree, and where did most of that time go?
Let’s chat about how Neuroplastiq can reduce the time it takes to get to the good part of a data project.