## When I don't build my own tool I talk a lot about building purpose-built tools. SparkDrop for...

When I don't build my own tool

I talk a lot about building purpose-built tools. SparkDrop for my content pipeline. Braindown for markdown. A custom job board when I needed to hire a video editor.

So people naturally ask: do you build everything yourself? Where's the line?

Here's my honest answer: I don't always build. There are times when buying makes more sense — even now, when AI makes building faster than ever.

The integration test

My heuristic comes down to one thing: third-party integration complexity.

When the tool I need depends heavily on external APIs, authentication flows, webhooks, and data syncing between multiple services — that's where I reach for an existing product. Not because I can't build the initial version, but because those integrations are fragile. They break when the third party changes something. They require maintenance that has nothing to do with the actual problem I'm solving.

Payment processing, email delivery infrastructure, social media publishing APIs — these are areas where the ongoing maintenance cost of building your own far outweighs the benefits. Someone else should be dealing with Stripe's webhook edge cases and Twitter's API rate limits. That's their full-time job.

The good news

The vast majority of tool needs don't meet that criteria.

Most internal tools — dashboards, content pipelines, review workflows, admin panels — are straightforward data-in, data-out systems. They don't need complex third-party plumbing. They need to fit your exact process, and that's precisely where a custom build wins.

An off-the-shelf tool gives you someone else's workflow. A custom build gives you yours. And now that AI has compressed the build time from months to days, the calculus has shifted hard toward building.

So my framework is simple: if the tool lives mostly in your own world — your data, your process, your team — build it. If it lives mostly in the world of third-party dependencies and ongoing integration maintenance — buy it.

Most of the time, you'll end up building. And you'll end up with something that fits.

Keep building,