Reliable data platforms enable reliable data products

To build reliable data products we need a data platform that provides us capabilities to do so.

For example, if one of the causes of your unreliability is that your upstream schemas change regularly, then you need a data platform that allows you build in change management, through data contracts.

As another example, if one of the cases of your unreliability is the inability to deploy your ETL with confidence, then the platform needs to provide capabilities for automated testing, blue/green deployments, easy rollbacks, and so on.

It’s through the use of these capabilities that allow consumers to rely on that data product, and use it to build something valuable or make key decisions.

I spoke a lot more about this, and also about the contract-driven data platform, on the most recent Straight Data Talk podcast. Check it out on Apple Podcasts or Zencastr.