How to set company goals in 2027
Why Q4 is the time to do it, why SMART goals and cascades get in the way, and what to do instead
Every January, a great many leadership teams sit down with fresh coffee and good intentions to work out what the year is for, and by the time they've agreed it, written it up and explained it to everyone, it's somewhere around the middle of February and a sixth of the year has already gone by without a clear direction. I've watched plenty of founders and CEOs go through exactly that, which is why the first piece of advice in this article is also the simplest one: if you want to set good goals for 2027, the time to do it is now, in the last quarter of 2026.
What follows is how I'd go about it. It's opinionated, because I've come to believe that most of the received wisdom about company goals was designed for a different kind of company in a different era, and I'd rather tell you what I actually think than give you another balanced list of frameworks to choose between.
Why Q4 is the right time to set 2027 goals
The fourth quarter is the one point in the year where you have almost a full year of evidence about what worked and what didn't, while still having enough room before January to think clearly rather than react. Your numbers for the year are mostly in, your team has a fairly good sense of what's been getting in the way, and the conversations you have now can be about the year ahead rather than the fires of the current week.
Setting goals in Q4 also means that the first working day of January is the first day of progress rather than the first day of planning, which sounds like a small thing but compounds over the year. A team that starts on the 2nd of January knowing what matters will be six weeks ahead of a team that starts by booking the planning day, and in a growing company six weeks is often the difference between hitting a goal and explaining why you didn't.
What 50 CEOs told us about why goals fail
We interviewed 50 CEOs about the goals they'd set and why so many of them hadn't been met. I expected a long list of reasons, and there were certainly plenty of individual stories, but when we looked across all of the conversations, the same two reasons came up again and again, and they accounted for the overwhelming majority of the goals that had been missed.
The conditions changed
The market moved, a customer left, a competitor launched, a new AI tool rewrote the economics of a department, and the goal no longer fitted the world it was set in.
The team wasn’t aligned
Everyone was working hard, but not in the same direction, so the effort of one department quietly cancelled out the effort of another.
What struck me most is that neither of these is really about ambition, talent or effort. Almost every one of those CEOs had a capable team that was working hard, and most of the goals were perfectly sensible when they were set. The goals failed because the world moved and the goal didn't move with it, or because the team drifted apart without anybody noticing until it was too late, and I'd argue that any approach to goal setting for 2027 has to be designed around those two problems before it's designed around anything else.
Why SMART goals are part of the problem
I know this will annoy some people, because SMART goals have been taught on management courses for decades, but I think they make both of those problems worse rather than better. The idea of making a goal specific, measurable, achievable, relevant and time-bound sounds entirely reasonable, and the intention behind it is good, but in practice I see three things go wrong with it in growing companies.
- They take a huge amount of effort to get right. Teams spend hours arguing over the exact wording and the exact number, and by the time everyone has signed off, the energy that should have gone into the work has gone into the definition of the work.
- They're hard to pivot on. Once a goal has been made that precise and that public, changing it feels like admitting failure, so when the market moves, teams tend to keep chasing a target they privately know no longer makes sense.
- They're hard to remember. A goal that needs a sentence, a number, a baseline, a date and three caveats to be accurate is a goal that nobody outside the leadership team can repeat back to you, and a goal people can't remember is a goal they can't act on.
I'd much rather a company had a goal that everyone can say from memory in one line, with a rough sense of when, and a clear description of what would be true when it's done, and then kept the precision for the evidence of how it's going rather than the wording of the goal itself.
Long-term goals and short-term tactical goals
None of this means abandoning the long view. I think every company needs a long-term goal that gives the year a direction, something like doubling revenue, becoming the obvious choice in a particular market or getting to profitability, and that goal should be stable enough that people aren't confused about where the company is heading from one month to the next.
Underneath that long-term goal, though, I'd keep the shorter-term goals deliberately tactical and deliberately flexible, so that they can change as you learn without anyone feeling that the whole strategy has been thrown out. The long-term goal tells you where you're going, the short-term goals tell you what you're focusing on to get there right now, and when the conditions change, which they will, it's usually the short-term goals that should move while the long-term one holds. That's how you deal with the first reason those CEOs gave without losing your sense of direction every time the market shifts.
Why cascading goals cause conflict
Cascading goals make a lot of sense on paper, because the logic is so tidy: the company sets its goals, each department sets goals that support them, each team sets goals that support their department, and in theory everyone ends up pulling towards the same thing. In practice, I've seen cascades create more conflict between departments than almost anything else in a growing company.
The problem is that each department ends up optimising for its own goal, and those goals often pull against each other in ways nobody intended. Finance, quite rightly, is asked to improve efficiency and keep costs under control, while marketing is asked to find new channels for growth, which means spending money on experiments, and a good number of those experiments will fail and look entirely wasteful from where finance is sitting. Both departments are doing exactly what their goals asked of them, and the result is a running argument about spend that slows the whole company down and leaves both teams feeling that the other one doesn't understand the business.
Fewer goals, with the right people behind them
The alternative I'd suggest is to have far fewer goals and to put the right people from across the company behind each one, rather than giving every department a set of goals to optimise for in isolation. If the company's goal is to grow revenue from new customers, then finance, marketing, sales and product all have a part to play in that one goal, and the conversation about experimental spend becomes a shared conversation about what's most likely to move the goal, rather than a tug of war between two budgets.
For most growing companies I'd aim for one goal at a time for each team working towards something distinct, and rarely more than a handful across the whole company, because every additional goal splits attention and creates another place for the team to drift apart. I wrote more about why this matters in Alignment beats activity, but the short version is that a team that agrees on what matters will reliably outrun a busier team that doesn't.
Where EOS and OKRs fit in 2027
I have a lot of respect for EOS and OKRs, and I've seen both of them help companies get focused when they were struggling, so I don't want this to read as a takedown. Both of them did something genuinely valuable, which was to force leadership teams to commit to a small number of priorities and look at them regularly, and that discipline is just as important in 2027 as it ever was.
Where I think they're starting to show their age is in the machinery around that discipline. They were designed to collect information from across a large organisation at a time when doing so was slow and expensive, so they come with scorecards, cascades, quarterly scoring and weekly meetings whose main job is to move data from the people who have it to the people who need it. I went into the history of this in Why are you doing that?, and my view now is that when AI can gather and summarise that information continuously, most of the ritual is overhead, which is the argument behind Information is cheap now. Decisions aren't. Keep the focus they gave you, and question the machinery they wrapped around it.
If you’re weighing up what to use instead, I’ve compared the alternatives to EOS and the alternatives to OKRs as honestly as I can.
Initiatives or constraints: what's holding you back?
Once a goal is set, most companies do the same thing next, which is to build a list of initiatives that they hope will get them there. More outreach, a new campaign, a new hire, a new process, each of them reasonable on its own, and together they quickly become a long list that competes for the attention of the same small group of people.
I'd ask a different question first, which is simply what's holding you back from the goal. In 1984 Eliyahu Goldratt published The Goal, which set out what became known as the Theory of Constraints: any system is held back by a small number of constraints at any one time, and improving anything other than the constraint doesn't make the system go any faster. Growing companies work in exactly the same way, so if your goal is held back by the fact that prospects can't understand the product in the first conversation, then a new partnerships initiative will just send more prospects into the same confusing first conversation.
The practical difference is enormous, and it's why I'd always rather see a team with three or four clearly stated constraints in order of priority than a team with fifteen initiatives running in parallel. There's a fuller argument for this in Remove constraints, don't add initiatives.
| The habit | What I'd do in 2027 |
|---|---|
| A SMART goal for every department | One or two goals for the company that everyone can say from memory |
| Goals cascaded down the org chart | A few goals, with the right people from across the company behind each one |
| A list of initiatives to hit the target | A short, ordered list of what’s holding the goal back |
| Progress reported monthly or quarterly | Progress visible to everyone, all the time, as the work moves |
| A plan that’s fixed until next year | A goal that’s checked every quarter and changed when the world does |
Why invisible progress kills momentum
The last thing I'd want every leadership team to take into 2027 is that a goal nobody can see progress against is a goal that slowly stops mattering. In most of the companies I talk to, progress towards the goal is visible to the leadership team once a month at best, and to everyone else once a quarter at an all-hands, which means that for most of the year most of the team has no idea whether what they did last week made any difference at all.
Momentum comes from seeing the needle move, and when people can't see it move, they understandably drift back towards whatever is loudest in their inbox, which is how a well-set goal in January becomes a forgotten slide by May. I think visibility has to be continuous and shared by everyone, so that the person who cleared something on a Tuesday can see on the Tuesday that it mattered, and so that the leadership team spots a goal drifting while there's still time to do something about it. That's what our vision of a world where every company is aligned in real time means in practice.
How I'd set company goals for 2027
If you've skimmed to this point, which I'd entirely understand, here's the whole approach in one list that you could take into a leadership meeting this week:
- Do it now, in Q4, so that January is the first month of progress rather than the first month of planning.
- Design for the two real reasons goals fail, which are changing conditions and a team that drifts out of alignment.
- Write goals people can remember, in one line with a rough when, rather than wrestling them into a SMART formula.
- Keep a stable long-term goal and let the shorter-term tactical goals change as you learn.
- Set fewer goals and put the right people behind each one, instead of cascading targets that set departments against each other.
- Keep the focus that EOS and OKRs gave you, and question the meetings and scorecards built to move information around.
- Ask what's holding you back before you build a list of initiatives, and put your constraints in order.
- Make progress visible to everyone, all the time, so that momentum builds rather than fades.
Where Clearahead fits
This is the way of working we built Clearahead around, and it's set out in full in The operating model for the AI era. You set your goal with AJ, the AI inside Clearahead, in about an hour, and AJ asks what's holding you back until you're past the symptoms and down to the handful of constraints that really matter. Your team picks up the moves under each one and says what's changed in a sentence when the work moves, and every Monday AJ writes a short brief for everyone on where you stand, what moved and what needs a decision, so the whole team can see how the goal is progressing without anybody preparing a report.
It's free for up to three people with no card needed, so if you're setting your 2027 goals over the next few weeks, you could have them set up and visible to your whole team this afternoon rather than at the end of January. If you'd like to see how it works first, Meet AJ walks through the conversation, and the five-question fit check will tell you whether it's right for your team.