How Many OKRs Should a Team Have?

How many OKRs should a team have? It’s a common question that comes up for teams that are about to create their OKRs. Either it’s a team lead asking themselves, or it’s an organisation that wants to give some internal recommendation on how many you should use.

The tricky thing is that you don’t really know what a good number is until it’s too late. You find out once you’re in the middle of a cycle and realise you picked too many, or you start feeling the consequences of picking too few. So it shows up in two moments: before you create your OKRs, and then later, once you’ve started following up and you’re wondering why it takes so long, why we can’t seem to find focus with our OKRs, or why some people in the meeting zone out the moment we start talking about them.

The short answer is a range. The more useful answer is that it depends on your team, your mandate, and how you want to spread ownership.

The right number of OKRs: a working range

There are some basic principles here. We want, let’s say, one to three objectives per team, and each objective should have two to four key results, so we end up with a manageable total. That’s what the theory says. And there are people who think you should work with just one objective, to really create focus.

The other thing from the theory: we can’t really have just one key result, or just one objective. Then we don’t have an OKR, we’ve got something we want to do and one way we measure it. We’re not capturing several sides of a strategy, we’ve basically put a label on top of a metric.

At the organisation or leadership level it’s similar, with a bit more room. There we want the strategic OKRs to cover a lot of different aspects, sales, marketing, people and culture, and so on, so we land somewhere around one to four objectives. You don’t need to fight tooth and nail over whether one objective has four key results or five. We can live with it being a few more there.

Why it depends on your team

But if we look at what’s actually right for one specific organisation, there’s value in thinking a bit broader. You can start from how big the team is. And then how big the team’s mandate is, and how much of that mandate is actually going to be spent working with OKRs. Are the OKRs there to drive change in some small, limited part of what the team does?

Take a support team. Maybe they spend a couple of percent of their time on process improvement, and the rest goes to solving support tickets. If that process improvement is what the OKR work is about, well, there isn’t much time behind it, so it can’t be a huge number of OKRs either.

The other piece is the number of people in the team. If it’s a team of two, you can picture wanting to give both of them clear visibility, that they share the OKR and own a couple of key results each. We could imagine just one OKR if we’re only two people. But take the other end of it. If we’re ten people and we still only have one OKR, that’s a problem. The goal is simple and clear for the team, sure, but a lot of people won’t feel any ownership, or really get why all this focus is on a target that maybe doesn’t touch their workday all that much. It’s a big strength to be able to hand out ownership of key results as broadly as possible in a team, to anchor the whole goal-setting process and create engagement in the follow-ups. So a bigger team is a good reason to let ourselves have another OKR.

Too few, too many

An easy example here: say we have organisation-level strategic OKRs, and we have one OKR, and the focus is on profitable growth. Turning from a focus on growth to a focus on long-term, sustainable growth, let’s say, and all the metrics are sales-related. It gives the organisation focus. But then it has this issue, that you’ve got a lot of teams that can’t really connect their day-to-day work to these high-level lagging metrics. You get low engagement, and it’s hard for those teams to connect to it.

So what’s missing here? We’d want to connect a couple more OKRs on the strategic level. Maybe one perspective on people and culture, or the workplace, and we’d need one to cover product or delivery, something like that, so we get bigger commitment to the strategy and more direct impact on those parts of the work too. Of course people and culture can contribute to profitable growth, but with only one OKR we leave a lot up to the team to decide how they’ll actually contribute. If they have a separate OKR, we can be a bit more precise, bring a bit more strategy, give a bit more direction on the how.

Same issue if we have a large team. Let’s put ourselves in the shoes of a marketing team with a lot of people, and we have only one OKR, and it’s all about generating leads. We’ll have people working on brand awareness, long-term events, strategies that might not have anything being executed this particular cycle. For them, the OKR process will do nothing. It won’t create engagement or drive progress or focus. It’ll just be for the few team members who are really attached to day-to-day lead generation, they’ll follow it closely and care about it. We lose the power of the OKR to actually align the entire team towards strategy. We’ve picked too few. We get focus, but only for some in the team. A second OKR that reaches a couple more areas through its key results is what brings everyone back in.

Picking too many is however the more common mistake by far. Is it easy to have too many? Everything feels important when you create your OKRs, and you’re scared to leave something out, because you think nobody will focus on it, or they’ll lose attention. It’s incredibly hard for a sales team not to have sales numbers in the key results, even though they might be more of a KPI and not that important for the OKRs. We want to drive some kind of change in the team, and it’s scary to leave them out, so you want them in. There’s always some negotiation about making sure everything’s represented.

And it’s the same with the key results. People often want to push some activities in too, to make sure those actually get done, so it turns into a kind of task list, or a project management list, where everybody wants their specific projects included. It’s really hard to limit the number of both objectives and key results. One habit that helps when you’re trimming is to ask, for each key result, is this the change we’re trying to drive, or just a number we already track?

In my experience, I’ve barely ever seen a team with the opposite problem, that they create too few. They can struggle to actually formulate or find their OKRs, but a team that has laser-sharp focus and only needs one OKR, where everybody knows that’s the most important thing, that’s very, very uncommon. It’s closely related to the common OKR writing mistakes that keep coming up.

What it costs your check-ins

One core thing here is that the more OKRs, the more key results. The bigger and slower the whole OKR process gets, the more time you need for each check-in, for each update. The harder the alignment sessions get across the organisation, when teams compare their OKRs. The whole framework just gets more demanding the more you put into it.

If you have too few instead, there’s a risk the OKRs become a kind of special-interest project that has very little relevance for parts of the organisation. That’s not a great spot either. What you want is a decent number, somewhere around this: the team has maybe two or three OKRs, with maybe three key results each. That’s roughly where organisations find a sweet spot, enough focus, and the time to check in maybe once a week for thirty minutes. You can still have a strong discussion with that kind of amount. Go well past it and you lose the focus and clarity on what’s actually important, and the alignment suffers too, you can’t be sure all teams read the OKRs the same way, or put the same weight on them.

A litmus test you can use today

This is the tough part, because the discussions that show up on each key result, you only really see them the first time you do a check-in. It’s really hard to anticipate which key results you’ll just fly through and which ones the team will stop and have long discussions on. It’s hard to just look at them and say, will this be thirty minutes or not? That’s something the team has to learn over time, what kind of key results, and how many, you can actually handle in an efficient way.

A core thing here is to trust the advice. Don’t think you can manage with a few more key results than recommended and that it’ll somehow work, it very seldom does. The counterintuitive bit is that it’s more about your team and your specific situation as an organisation than it is about chasing the exact numbers from the theory. And you can think of this as a tool. An organisation can put a demand on every team that they can’t have more than one or two OKRs. If focus is a core reason you’re implementing OKRs in the first place, it’s a really strong move to limit how many are allowed.

So, the honest answer to how many OKRs a team should have: start in that one to three objectives, two to four key results range, let your team’s size and mandate move you within it, and lean towards fewer whenever you’re unsure.