The browser support decision nobody wants to make (and how to make it anyway)

"We need to support all browsers" is the default answer to a question that was never actually asked out loud: which specific browsers and versions, at what usage threshold, justify the engineering cost of supporting them. Left undecided, the default becomes an unstated, unbounded commitment that quietly taxes every feature decision going forward.

The actual decision should be data-driven and specific: pull the site's real analytics for browser and version usage over the last three to six months, set an explicit threshold, commonly somewhere around one percent of traffic, below which a browser gets graceful degradation rather than full feature support, and document that threshold somewhere the whole team can see it.

This isn't about abandoning users on older browsers, it's about being explicit rather than accidentally inconsistent: a documented policy means every future feature decision has a clear answer for "does this need to work in browser X," instead of that question getting relitigated, inconsistently, on every single pull request by whoever happens to be reviewing it.

The teams stuck relitigating browser support on every feature are almost always the ones who never made the decision explicitly in the first place. An unstated policy isn't cautious, it's just undecided, and undecided policies get applied differently by whoever's making the call that day.