Why does a hidden buffer stop working?

Because it gets found: a team lands early twice, or finance compares headcount to commitments. After that the stakeholder subtracts their own guess at your padding from every date, and the two of you are negotiating against numbers neither can see.

What do you publish instead?

Three things. The capacity number, said once with the last quarter as evidence. The measured variance from the last 30 days on every date, with its trend. And the burn: what consumed the buffer when it was consumed.

What does the Roadmap Progress Agent check?

Every weekday at 9 AM it cross-references roadmap status, engineering tickets, commits, and PM updates. It flags anything in progress with no commits in five days and reports ETA accuracy: on-time percentage, average variance in days, and velocity trend over four weeks.

Is planning at 70 percent of capacity wrong?

No. Bart's reasoning holds: the buffer gets consumed reliably by things nobody could predict, and a team that delivers what it said is trusted more than one that commits to everything and misses. The only change is that the 70 percent is stated openly.

Should a Product Manager Hide Schedule Buffer?

Should a product manager hide schedule buffer?

THE SHORT ANSWER

No. Plan below theoretical capacity and say so. Dr Bart Jaworski's advice (2026-10-02) is to plan at roughly 70 percent of capacity and hold the buffer without advertising its existence. The 70 percent is sound. The hidden part fails: once a stakeholder finds the padding, every later date gets discounted, and it does nothing for the slip nobody reported. At a company I advised, a customer feature was de-prioritized for technical debt and the PM learned a week before launch. Replace the hidden buffer with a published one: measure ETA accuracy (on-time percentage and average variance in days), attach the variance to every date, and report what consumed the buffer. My Roadmap Progress Agent produces those numbers every weekday morning.

Dr Bart Jaworski's essay lists eight hard realities of the PM job, and the first, nothing is ever on time, comes with the advice to plan at about 70 percent of capacity and keep the buffer quiet. Kill the Hidden Buffer keeps the 70 percent and retires the quiet, for three reasons: a hidden buffer works once per stakeholder, it models upward the withholding his own seventh truth warns about, and it does not cover a feature that was silently de-prioritized.

The instrument for the replacement is in Build Your Roadmap Progress Agent, which reports ETA accuracy daily, and the reason padding is more visible now than it used to be is in The Cost of Being Wrong Is the Only Number That Matters Now. Both belong to the Product Leadership argument.

Last reviewed 2026-10-02 · 1 min read

Should a product manager hide schedule buffer?

SOURCES

THE LONG VERSION

RELATED ANSWERS