I'm always nervous about perverse incentives that can come out of measuring software engineering too closely.
Response times and bundle size seem reasonable, but I'm also not sure how you measure improvement apples-to-apples against theoretical alternatives there.
I've always worked with small teams, so engineering KPIs haven't always offered more good than bad. I can imagine in a larger org, it could help.
Now that we're open source I could see us working community PRs merged as a KPI. Seems to generally incentivize smaller PRs, more contributors, and feature/bug-fix velocity.
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.