the world appears to be full of people who will pay a premium for a consolidated collection of tools, packaged and bundled into a neat little suite, sold on a single bill.

In data products, modularity, flexibility, and composability are markers of our craftsmanship. But most people “don’t care about the aesthetics of software in the same way non-plumbers don’t care about the aesthetics of plumbing—they just want their information in the right place or their hot water to work.”

The same applies to other principles. I prefer open-source frameworks—unless Fivetran is just more reliable than Stitch or Airbyte. I like modularity—until it leads to “drops in signal quality, lower data transfer speeds,” and a failing user experience. As they say, a value isn’t a value until it costs you something. And if I’m honest with myself, thinking as a buyer rather than builder of data tools, I’m not willing to pay very much to hold on to them.9

Put another way, how much more would you pay for Fivetran if the product were exactly the same, but it was built on an open-source framework? How much less would you pay for dbt if it were proprietary? Yes, these changes might make vendors build better or worse software, but that’s the point: Tenets like these are a means to an end, and as a customer, I pay for the ends.