Some selected quotes that point at the problem we’re solving:

The modern data stack is defined as everything between an architectural diagram1 and a state of mind,2 but nobody would characterize it as a coherent layer suitable for widespread application development.

That’s Meltano!

Even if Casado’s prediction is a bit too bold, it’s easy to imagine a future full of smaller, more focused data apps.4 Surely, even if we don’t all leave Salesforce.com for getsalesforce.ai (YC S22), we’ll at least have dedicated analytical tools for managing remote workforces, for connecting online marketing campaigns with offline retail sales, and for reliably computing SaaS metrics.

Matatika is an early example of such a data app built on top of Meltano as the “data platform development framework”! cc @aaron_phethean

Still, Casado’s vision is missing a key piece: The bridge between where we are today, and the platform that’s needed to make these apps possible. The modern data stack—and more specifically, the centralized data warehouse—is a good foundation; data apps are a cool idea. How do we plug the latter into the former? We have the underpants and we want the profit, but what do we do in the middle?”

That’s Meltano!

Piecemeal architectures like this are also a broken experience for customers. Business concepts like users, customers, transactions, and events are defined in a mix of dbt models, queries in reverse ETL tools, configurations in third-party apps like those above, and the application code of internal tools. This can be, as was the case for metrics, a mismatched mess.

That’s what Meltano solves!