Setting up OKRs in an SMB or scale-up

OKRs are first about choosing what matters, then aligning and tracking. In an SMB or a scale-up, their value isn't to produce one more table of objectives, but to force an organization to decide, over a given period, what genuinely deserves its attention. Used well, they connect strategy to execution without adding weight. Used badly, they become a disguised task list.

So the useful question isn't "should we do OKRs," but "is our organization at the point where they bring more clarity than they cost in effort." This page goes deeper into one of the pillars laid out in Structuring your company to stay agile. For a step-by-step first cycle, the guide to your first OKR cycle is more hands-on.

The starting problem

The need for OKRs rarely shows up as a missing tool. It shows up as a surplus of priorities. As a company grows, important topics multiply, each one feels urgent, and attention scatters. Teams work hard without always knowing whether they are working on what matters most, and leadership struggles to trace the link between daily activity and strategic intent.

There's also a traceability problem. When trade-offs stay implicit, no one really knows why one topic was chosen over another, or how anyone will know whether it succeeded. OKRs answer precisely this double gap: they force you to name the small number of things that have to change, and to say in advance how that change will be measured.

What OKRs actually bring

The contribution of OKRs comes down to four linked effects. The first is focus: by limiting the number of objectives, they force you to choose rather than chase everything. The second is alignment: made visible, objectives let each team locate its work relative to the others'. The third is tracking: an objective paired with measurable key results gets steered instead of just discussed. The fourth is transparency: when everyone sees the priorities of the whole, coordination becomes a shared-objectives problem rather than a meetings problem.

John Doerr, who brought the method to Google starting in 1999, stresses one simple distinction: the objective describes what you want to accomplish, the key result describes how you will measure that you got there. It's this discipline that separates a useful OKR from a mere list of intentions. He also distinguishes committed OKRs, expected to reach 100%, from aspirational OKRs, whose target sits around 0.7, to push people to aim high without penalizing risk-taking.

The common ways they go wrong

The first failing is turning OKRs into a task inventory. You file in everything the team plans to do, and you lose what gives them their value: selection. An OKR isn't the list of planned work, it's the expression of the change you're after. The second failing is the overly vague objective, which sounds good but can't be measured, and so lends itself to every interpretation at the end of the cycle.

The third failing is inflation: multiplying objectives until prioritization disappears. Doerr recommends sticking to three to five objectives per cycle, with at most five key results each, precisely so that focus stays real. The fourth, the most subtle, is the confusion between activity and result. A key result that says "analyze," "take part in," or "support" describes an activity, not an observable outcome. As long as you only measure activity, you can be very busy without moving anything.

The conditions for success

OKRs only produce their effects if a few conditions are met. The first is genuine leadership support. If leadership doesn't own the trade-offs and hold itself to the same discipline, teams sense it quickly and the whole effort hollows out. The second is a limited set of priorities, with one clearly owned top priority. Verne Harnish insists on this point: no organization can seriously focus on too many things at once, and choosing a single number-one priority is often the hardest and most useful act.

The third condition is the quality of the key results, which must be specific, measurable, and time-bound. The fourth is making dependencies between teams explicit: an objective that is well written but poorly executed often is because it relied on a contribution from another team that was never named. The fifth, finally, is the existence of regular check-ins. Without a follow-up rhythm, OKRs die between the launch and the end-of-cycle review. Follow-up isn't an accessory of the method, it's its condition for survival.

What changes with size and maturity

OKRs don't roll out the same way at every stage. In a still-small structure, where coordination holds through proximity, imposing them too early adds ceremony with no clear benefit. They take on meaning when the informal way no longer suffices, that is, when teams no longer spontaneously see each other's work and a light framework becomes necessary to connect strategy and execution.

Management maturity matters as much as size. OKRs assume managers who can frame, follow up, and adjust. Where that relay is missing, they expose the problem rather than solve it, because an objective with no managerial follow-up stays a dead letter. Better, then, to first strengthen the management layer, and only then introduce OKRs as a shared language of priorities.

When OKRs are the right tool

OKRs are a good tool when an organization has outgrown the stage where everything is coordinated by word of mouth, when it has a management layer able to handle follow-up, and when it suffers mainly from too many priorities and too little alignment. In that case, they give a simple framework for choosing, making measurable, and tracking.

Conversely, when the real problem lies elsewhere (blurred responsibilities, absent management, an undecided strategy), OKRs won't fix it and risk masking the difficulty behind an appearance of method. They support a steering practice; they don't replace it. It's by keeping this modesty of use that you get the most out of them.

At Serendly, we help teams hold this discipline over time: clear OKRs, tied to a steady follow-up rhythm and to the managerial conversations that keep them alive, rather than a framework you write at the start of the quarter and forget about afterward.

OKRs only hold through the rhythm that follows them and the conversations that adjust them. The operating cadence and the manager-report 1:1 are the two mechanisms that keep them from slipping back into oblivion.