Tools We Built and Threw Away

260606 · Not everything ships. Some tools taught us what not to build. Here are the ones that did not make it.

We have killed more tools than we have shipped. That is not failure. That is how you avoid shipping bad tools.

A tool gets killed for one of three reasons: it is too complex to stay simple, it duplicates something that already exists and does it better, or nobody — including us — actually uses it.

THE COMPLEXITY TRAP

Some tools start simple and grow. A text editor that gains plugins. A converter that gains fifteen output formats. At some point it stops being a tool and becomes a product. We are not building products. We kill it.

THE "ALREADY EXISTS" CHECK

Before shipping anything, we ask: does this already exist and is it already free? If yes, we need a real reason to build it anyway. "Ours will be simpler" is a reason. "Ours will have more features" is not.

WHAT THE DEAD TOOLS LEFT BEHIND

Every killed tool taught us something. The complexity ceiling. The scope that looks small and is not. The feature that sounds useful and never gets used.

The graveyard is where the good decisions live.