Annual planning means agreeing three to five outcomes that must be true by December, attaching one metric and a recorded starting value to each, naming a single owner, and scheduling the mid-year point where you are allowed to change your mind. The year sets direction; the quarters do the work.
Most annual plans fail in the same undramatic way. They are written in December, they sound right, and by April nobody can say whether any of them is on track — because the goals were never written in a form that can be scored. This guide is about avoiding that specific failure. If you are setting structured goals for the first time, start with goal setting for small teams →
What annual planning is for — and what it isn’t
Annual planning exists to answer one question: what must be true about this business twelve months from now? It is not the place to decide what anyone works on in February. Mixing those two jobs is why annual plans become long, stale documents — the direction is still valid in June but the task list stopped being valid in week three.
The year decides direction
Three to five outcomes, each with a metric and a baseline. Deliberately coarse. It should still read as correct in month eight, or you chose activities rather than outcomes.
The quarter decides work
Owned, measurable goals for thirteen weeks, re-cut four times a year against the same direction. This is where reality gets a vote, and where plans are allowed to change without ceremony.
That split is also the honest answer to a question small teams ask a lot: if a quarter is the right planning unit, why plan a year at all? Because a sequence of four independently chosen quarters is not a strategy — it is four unrelated pushes. The year is what makes the quarters add up. For the quarter itself, see quarterly planning for small teams →
When to run it
Start six to eight weeks before the year begins. That is early enough to gather real inputs and late enough that the numbers still mean something. Compressing the whole thing into one December afternoon is the most common scheduling mistake, and it is why so many annual goals arrive without baselines.
Gather the inputs
Pull this year’s scores, the numbers you actually run on, and whatever you have learned about your customers. No meeting yet — just make sure the room won’t be arguing from memory.
Set direction
A half-day with the people accountable for outcomes. Agree the three to five things that must be true by year end. Direction only — no quarterly detail yet.
Write and baseline
Turn each one into a goal with a metric, a starting value and a target. This is the step teams skip, and it is the step that makes the year scoreable.
Cut the first quarter
Run normal quarterly planning against the annual direction. The year becomes work here, not in the annual session.
How to set company goals for next year
Seven moves take a small company from “what should next year look like?” to a set of goals that can still be scored in month nine.
Close the year you are leaving
Score every goal from the year ending — hit, partial, missed — before writing a single new one. Planning forward without closing backward means you plan from memory, and memory flatters.
Name what must be true by December
Not activities, outcomes. Three to five statements about the state of the business a year from now. If a statement could be true whether or not you did anything, it is a forecast, not a goal.
Attach one metric to each
One number per goal, chosen because it moves when the goal is achieved and stays flat when it isn’t. Two metrics on one goal means you have two goals and haven’t admitted it.
Record the starting value on the day you write it
Write down where the metric stands right now, with the date. Without a baseline you cannot compute progress in March — you can only compute distance from zero, which is a different and usually meaningless number.
Put a single name on each goal
One owner per goal, and it is the person accountable for the outcome, not the person who does the most work on it. Shared ownership reads as care and behaves as no ownership at all.
Decide what you are not doing
Write the short list of things you deliberately dropped to make room. This is the half of planning that gets skipped, and it is the half people quote back to you in June when everything is somehow still a priority.
Schedule the reforecast before you leave
Put a mid-year review in the calendar now, while the plan still feels obviously right. Nobody schedules it later, and a year is long enough that at least one of your goals will be overtaken by events.
Step four is the one that separates a plan from a wish, and it gets its own section below. Step six is the one people remember you for. For turning each of these into goals your teams can act on, see how to align team goals with company goals; for what these look like written out, see company goals examples →
Write a goal you can still score in month nine
A yearly goal is only scoreable if you recorded where you started. This sounds obvious and is skipped constantly, because in December the starting value feels like something everybody knows. By September nobody remembers it, the system of record has moved on, and the honest answer to “are we on track?” becomes a matter of opinion.
Grow revenue significantly this year.
Grow monthly recurring revenue from $41,000 on 1 January to $70,000 by 31 December. Owner: Maria.
With a start, a target and a date, progress is the distance you have covered between the two: (now − start) ÷ (target − start). At $53,000 in June that is 41% of the way, which is a fact rather than an argument. The alternative most spreadsheets use — current divided by target — would report 76% and quietly congratulate a team that is behind.
The same formula is what saves goals that are supposed to go down. “Reduce churn from 8% to 2%” computed as current over target gives 400%, capped to a triumphant 100% on the day you write it. Computed as distance covered, churn at 6.5% is 25% of the way — which is both correct and considerably less flattering. You can check any of your own numbers against this in the OKR score calculator, and the full scoring rules live in OKR grading and scoring →
Turning the year into quarters
The annual set should never be worked on directly. Each quarter, take the three to five annual outcomes into your quarterly session and ask one question per goal: what has to move in the next thirteen weeks for this to still be reachable in December? That produces the quarter’s goals, and it keeps the year present without turning it into a task list.
Two rules keep this from degenerating. First, not every annual goal needs quarterly work every quarter — a goal that is genuinely parked this quarter should be visibly parked, not silently ignored. Second, the quarterly goals are re-decided each time; they are not the second instalment of a plan written in December, because December did not know what March would look like.
The mid-year point, where you are allowed to change your mind
Halfway through, sit down with the annual set and give each goal one of three labels: on track, reforecast, or retired. The point is not to rescue the plan — it is to make sure the plan still describes what you are actually doing, because a plan that has quietly diverged from reality teaches everyone to ignore plans.
Retiring a goal on purpose, with a written reason, costs you nothing and buys you credibility. Leaving a dead goal on the list until December costs you the next planning session, because everybody arrives knowing the exercise is theatre. For running the close itself, see how to run a goal review meeting →
Celorly records the starting value the moment you create a goal, so the baseline you need in September exists whether or not anyone thought to write it down in December. Progress is computed as distance covered, so goals that go down score correctly instead of showing 100% on day one. Set the year as a long cycle, cut quarterly cycles beneath it, and the annual numbers roll up from the quarters rather than being retyped.
Common annual planning mistakes
Nearly all of these are versions of the same thing: writing a goal in a form that cannot be checked, then discovering it too late to act.
Goals with no starting value
“Grow revenue” with no recorded baseline can never be scored. In month nine you will be arguing about whether it counts, and the argument will be unwinnable because the data to settle it was never captured.
Twelve annual goals
Three to five is a plan. Twelve is a wish list with a deadline. A small team can carry roughly one meaningful company goal per quarter per priority — count backwards from that.
Cascading the year straight to individuals
Annual company goals belong to the company. Push them onto individual objectives and they turn into a performance-review instrument, which reliably produces sandbagged targets rather than ambitious ones.
Set in December, opened in December
A goal nobody looks at between kickoff and close is decoration. The year needs the same weekly and quarterly rhythm as anything else — the annual horizon changes the scope, not the cadence.
Copying last year with bigger numbers
Rolling last year’s goals forward with a 20% uplift skips the only question worth asking: is this still the thing that matters? Re-decide, don’t index.
Treating the plan as fixed
Refusing to reforecast when reality moves doesn’t protect the plan, it just makes everyone quietly stop believing it. Adjust openly in the middle; that is what the mid-year review is for.
Annual planning — quick answers
How many company goals should you set for a year?+
Three to five. That is enough to describe a direction and few enough that every one of them can get real attention. Beyond five, a small team cannot give any single goal sustained focus for twelve months, and the list stops functioning as a set of priorities.
When should you start annual planning?+
Begin gathering inputs six to eight weeks before the year starts, hold the direction-setting session about four weeks out, and write the goals with their baselines two weeks out. The first quarter is then planned normally in week one, against the annual direction.
What is the difference between annual planning and quarterly planning?+
Annual planning sets direction: three to five outcomes that should be true by year end. Quarterly planning turns that direction into owned, measurable work for the next thirteen weeks. The year is the backdrop; the quarter is what you execute against.
How do you measure progress on a yearly goal?+
Record the metric’s starting value on the day you set the goal, then measure progress as the distance covered between that start and the target — not as the current value divided by the target. Without a recorded baseline, a yearly goal cannot be scored honestly at any point during the year.
Should annual goals be assigned to individual employees?+
No. Annual company goals should stay at company level, with a single accountable owner each. Cascading them directly onto individual objectives turns them into performance-review criteria, which pushes people to negotiate easier targets rather than ambitious ones.
What should you do if an annual goal stops making sense mid-year?+
Reforecast it or drop it openly at the mid-year review, and write down why. A goal that everyone knows is dead but nobody has retired teaches the team that goals are decoration — retiring it deliberately teaches the opposite.
Can you use OKRs for annual goals?+
Yes, with one adjustment: keep the annual set as objectives with a small number of key results, and re-score the key results quarterly rather than once at year end. An objective scored only in December gives you no chance to act on the number.