LinkedIn·Monday, 17 August 2026·9d ago

Data analysts should not be the last line of defense for data quality. Yet that's where many teams end up. A data issue enters a pipeline…

Maarten Masschelein

CEO & Co-Founder @ Soda | Data and Context Management for AI

Data analysts should not be the last line of defense for data quality. Yet that's where many teams end up. A data issue enters a pipeline upstream but isn't discovered until someone notices a broken dashboard, an unexpected KPI, or inconsistent results in an analysis. By then, the work has shifted to analysts and stewards. They investigate the problem, trace it back to the source, and clean up data that should never have reached them in the first place. A better approach is to prevent bad data from moving downstream. Data contracts define what a dataset must look like before it can move through a pipeline. With Contract Autopilot, you can generate a baseline set of contracts from your existing datasets instead of writing every check by hand. As your data evolves, Contract Copilot lets you update those contracts in plain English, shows exactly what changed, and lets you approve the proposal before it is applied. That changes the work. Instead of spending time cleaning data after problems appear, teams spend their time reviewing proposed quality rules and investigating the exceptions that actually matter. The goal is not to automate analysts. The goal is to automate the repetitive work that keeps them from doing analysis.