Rocks (EOS)

In the Entrepreneurial Operating System, known as EOS and popularized by Gino Wickman's book Traction, Rocks are the small set of priorities a team commits to finishing in the coming quarter. The recommended range is three to seven Rocks, with many teams aiming for the lower end, because the whole point is focus. A Rock is not a routine task or an ongoing responsibility; it is a specific, meaningful objective that must be completed within the ninety-day quarter and that moves the company toward its longer-term plan. Each Rock is owned by a single person, so accountability is clear, and Rocks are set both at the company level and for each leader individually. The name comes from a familiar image, popularized in Stephen Covey's work, of filling a jar with big rocks first before the pebbles and sand, meaning the most important things get placed before the smaller ones crowd them out.

Rocks work because they convert a vague ambition into a short, concrete list with a deadline. A company might have a one-year plan and a longer vision, but a quarter is a horizon a team can actually see and execute against, so the Rocks translate the bigger goals into what must be done in the next ninety days. Every quarter the team resets, reviewing whether the last set of Rocks was completed and choosing the next set, which keeps priorities current as conditions change. Rocks are reviewed in the team's weekly meeting, so progress is visible and problems surface early rather than at the end of the quarter. The discipline of committing to only a handful of priorities, and of naming an owner for each, is what turns strategy into steady, accountable execution rather than a wish list that never gets done.

Rocks are easy to confuse with ordinary goals or tasks, but the distinctions are deliberate. A task is a small unit of work; a Rock is a substantial quarterly priority that might contain many tasks. A goal can be vague or open-ended; a Rock is specific, owned by one person, and bounded by the ninety-day quarter, so it is either done or not done by quarter's end. The number is capped on purpose. Keeping Rocks to three to seven per person or team forces the hard choice of what matters most now, which a long list of goals never does. In this sense Rocks are less a new idea than a disciplined format: a few named, owned, time-boxed priorities, reset every quarter.

Rocks also invite comparison with OKRs, the objectives-and-key-results method used at many technology companies, and while both push teams to focus on a short list of priorities each quarter, they are not the same. OKRs pair an ambitious objective with measurable key results and often deliberately set stretch targets that teams do not expect to fully hit. Rocks are framed as commitments meant to be completed, with a clear owner, and they sit inside the wider EOS framework of vision, meetings, scorecards, and issue-solving rather than standing alone. A team can run OKRs without EOS, but Rocks are a component of EOS specifically. The practical difference is tone and system: OKRs emphasize measurable stretch within their own method, while Rocks emphasize a small number of owned, completable priorities inside the EOS operating rhythm.

Setting Rocks well begins with restraint. The temptation is to name everything important as a Rock, but a list of a dozen priorities is no priority at all, so the discipline is to choose the three to seven that will most move the business this quarter and let the rest wait. Each Rock should be specific and finishable inside the ninety days, with a clear definition of done, so there is no ambiguity at quarter's end about whether it was completed. Give every Rock a single owner, because shared ownership dilutes accountability. Tie the Rocks to the company's one-year plan so the quarter's work ladders up to something larger. Then review them every week, so a Rock that is slipping gets attention while there is still time to recover it rather than at the final review when it is too late.

The failures are familiar to anyone who has watched priorities dissolve. Setting too many Rocks spreads a team thin and guarantees several go unfinished, defeating the purpose of a focused list. Vague Rocks with no clear finish line let a quarter end in argument about whether the work was really done. Rocks with no single owner drift, because everyone assumes someone else has them. Ignoring Rocks between quarterly planning sessions, rather than reviewing them weekly, lets slippage hide until it is irreversible. And treating routine work as Rocks clutters the list with things that would happen anyway, crowding out the true priorities. The remedy is the discipline the format was built for: few Rocks, each specific, owned, tied to the bigger plan, and reviewed often enough to keep them on track.

Rocks — from the Entrepreneurial Operating System and the book Traction — are the three to seven owned priorities a team commits to completing each 90-day quarter.

Etymology: source.

Search interest for this term over the last five years:

View interest-over-time on Google Trends →

Areas of marketing where rocks (eos) is a core concern: