In this guide
- Two horizons
- How annual plans drift
- Four linking rules
- What happens at year end
- A short checklist for the reset
- A worked example: three years of reasonable targets
- What three points of drift actually costs
- What two horizons need from the software
- Two horizons in Profit.co
- Questions this article answers
Key Takeaways
- →Hoshin runs on two horizons at once: a three-to-five-year breakthrough and the annual objectives that serve it.
- →Separate documents cause drift. Annual targets get set from last year’s result because that is the number in front of the planning team.
- →Set annual targets by working back from the trajectory, not forward from last year. The tension that creates is the point of the exercise.
- →A plan that hits every annual target and misses the breakthrough has failed once, slowly, with no single year to point at.
- →The shortfall has a number attached, and finance already has both halves of it in the management accounts.
- →The annual reset is the test. Strategic planning software that starts a fresh file each January is where the link between horizons dies.
The breakthrough objective is the reason the annual plan exists. When the two live in separate documents, each year’s plan drifts back towards last year plus a few percent, and the breakthrough becomes a slogan.
Two horizons
Hoshin kanri works on two timescales at once. Breakthrough objectives describe a change the organization intends to make over three to five years: a new business model, a step change in cost position, a transformed customer experience. Annual objectives describe what has to happen this year for the breakthrough to stay on track.
The X-matrix is designed to hold both, with breakthrough objectives driving annual ones and annual ones driving the priorities and KPIs around the rest of the grid. In practice, the two horizons often come apart.
How annual plans drift
The breakthrough is usually set at a strategy offsite and recorded in a strategy document. The annual plan is set in a different meeting, often by different people, and recorded in the X-matrix. Each year, the annual planning team starts from last year’s results, because those are the numbers in front of them.
The result is predictable. Annual targets become last year’s figure plus a reasonable increment. Each one is defensible. Together they add up to a trajectory that falls well short of the breakthrough, and nobody notices until year three, when the gap is too large to close.
If you can’t see the breakthrough from the annual plan, the annual plan will quietly stop serving it.
Four linking rules
- Every annual objective names its breakthrough. If it doesn’t serve one, ask why it’s on the matrix.
- Every breakthrough has this year’s objectives, or an explicit decision that it’s paused this year, with a reason.
- Annual targets are set from the trajectory, working back from where the breakthrough needs to be, not forward from last year.
- Cumulative progress is visible. The review can see how far the breakthrough has moved across all years so far, not just this one.
The third rule is the hardest and the most valuable. Working back from the breakthrough often produces an annual target that looks unreasonable against last year. That tension is the point. It forces either a bolder plan or an honest conversation, through catchball, about whether the breakthrough is still realistic.
What happens at year end
The annual reset is where the two horizons meet. On paper, it usually means a fresh matrix: new annual objectives, re-entered breakthroughs, and last year archived. Much of what was learned stays in people’s heads.
When both horizons live in one system, the reset works differently. Breakthrough objectives carry over unchanged. This year’s annual objectives close with their results and their catchball history intact. Next year’s annual objectives are created against the same breakthroughs, and the team can see the cumulative trajectory before setting them.
The question shifts from “what should we do next year?” to “given where the breakthrough is after this year, what has to be true by next December?”
A short checklist for the reset
- For each breakthrough, how far has it moved since it was set, and is the trajectory on track?
- Which of this year’s annual objectives contributed most, and which correlations turned out to be wrong? (See The X-Matrix Is a Set of Hypotheses. Treat It Like One.)
- Is any breakthrough no longer the right ambition? If so, change it deliberately, through the strategy process, rather than by letting the annual plan drift.
A worked example: three years of reasonable targets
A breakthrough commits to raising services from 31% to 40% of sales over three years. Here are two ways the annual targets could be set.
| Year | Forward target (last year plus a reasonable step) | Result | Trajectory target |
|---|---|---|---|
| Year 1 | 33% | 33% | 34% |
| Year 2 | 35% | 35% | 37% |
| Year 3 | 37% | 37% | 40% |
Illustrative. The forward targets were all met.
With forward targets, every year is a success. Each review is green. And at the end of year three the business is at 37% against a 40% commitment, with no year in which anyone could point to the moment the plan fell behind.
With trajectory targets, year one is harder and might be missed by a point. That miss is visible in the first year, when there’s still time to add priorities, reshape the path or revisit the breakthrough openly.
A plan that meets every annual target and misses its breakthrough hasn’t succeeded three times. It has failed once, slowly.
What three points of drift actually costs
The gap in that example is three points of services share. That sounds like a rounding error in a strategy document, and it is not: it is a number finance can calculate exactly, because both halves of it are already in the management accounts. Nobody has to estimate anything.
| What to count | Where the number comes from |
|---|---|
| The shortfall, in the breakthrough’s own unit | The breakthrough target minus where you actually landed. Three points, in this example |
| What one point is worth | The margin differential between the two revenue mixes, which finance reports already |
| How many years it compounds | From the year the trajectory first fell behind, not from the year anyone noticed |
| The cost of finding out in year three | The priorities you would have funded in year one, when they were still cheap enough to work |
Four rows, no estimates. The last one is the largest and the only one that is genuinely a judgement.
That last row is the whole argument for linking the horizons in one place. The three points were not lost in year three; they were lost in year one, when a 33% target was accepted instead of 34% and nothing on the page made the difference visible. A tool that shows the trajectory beside the proposal turns a slow failure into a decision someone gets to make while it still costs very little. If you need to turn that into a funding request, building the business case for hoshin kanri software has the one-page structure, and what drives the cost of hoshin kanri software has the other side of the ledger.
What two horizons need from the software
Most strategic planning software is built around a single planning period, because most planning is annual. Hoshin is not, and the mismatch only shows up at the first reset, roughly a year after the purchase. Six things are worth testing before that.
| What two horizons need | Why it matters | How to test it in a demo |
|---|---|---|
| Objectives that carry a multi-year period | A breakthrough spanning 2026 to 2028 cannot live in an annual container | Ask to see an objective whose period is three years long |
| Annual objectives nested beneath it | The link has to be structural, not a naming convention people maintain by hand | Ask what happens to the link when the annual objective closes |
| A matrix readable across years | The reset question needs the cumulative position, not this year’s slice | Ask to see the X-matrix for a year that has already closed |
| Trajectory visible while setting targets | Rule three fails if the planner cannot see both numbers at once | Ask where next year’s target is entered, and what is on screen beside it |
| History that survives the reset | Check-ins and catchball from closed years are the input to the next plan | Ask to read a catchball thread from two years ago |
| No annual re-entry | Rebuilding the matrix each January is exactly where the link between horizons dies | Ask what the January reset actually requires a person to do |
The last row is the one to push hardest on. It is also the one most likely to be answered with a template.
The sixth row deserves the most skepticism in a demo, because “you just copy last year’s matrix” is a perfectly reasonable-sounding answer that quietly reintroduces the whole problem. Copying is re-entry. The breakthrough should not be something anyone retypes in January; it should simply still be there, with two years of movement already recorded under it.
This is also where annual planning software and hoshin kanri software genuinely differ, and the difference is worth naming before a shortlist forms rather than after. Annual planning tools, and most strategic planning tools generally, are built to produce a good plan for next year. Hoshin kanri software is built to keep a multi-year commitment honest across several of them, which is a different job and needs a different data model underneath. The seven pass/fail tests apply to both, and the six rows above are the ones specific to running two horizons at once.
Two horizons in Profit.co
In Profit.co, objectives can carry a multi-year strategy period, such as 2026 to 2028, while annual objectives and quarterly key results run on their own periods beneath them. The X-matrix can be viewed across several years, so this year’s plan is read against the breakthrough it serves. Each objective keeps its check-in history and catchball sessions, so the reset starts from the record. More on the hoshin kanri software overview.
Hoshin Kanri sits in the Essentials plan alongside OKRs and Goals, Balanced Scorecard and Strategy Roadmap, which matters more here than it looks: the quarterly OKRs teams set beneath an annual objective run on the same data model as the breakthrough three levels above them, so the trajectory is readable from top to bottom without an export. Running the specialist route instead means a hoshin tool, a separate OKR software subscription and balanced scorecard software for the board view, with the multi-year link maintained by hand across all three. The 30-day free trial is long enough to load one breakthrough with its annual objectives underneath and see whether the view answers the reset question.
Free Product Demo
See your breakthrough and annual plan together
A 30-minute walkthrough, tailored to your team.
Book a free demo
What happens in the call
How your teams set goals, track progress and review results.
Strategy, performance, projects and engagement on one platform, with AI.
A practical path to align teams, execute faster and drive results.
Questions this article answers
A change the organization commits to over three to five years, such as a new business model or a step change in cost position, which annual objectives are set to serve.
Because annual targets are usually set from this year’s results plus a reasonable increment. Each is defensible, but together they fall short of the breakthrough trajectory.
They carry over unchanged unless deliberately revised through the strategy process. Annual objectives close with their results, and next year’s are set against the breakthrough’s position.
Hold a breakthrough on its own multi-year period with annual objectives nested beneath it, show the X-matrix across closed years, and keep check-in and catchball history through the reset. Annual planning software that starts a fresh file each January breaks the link.
Work backwards from where the breakthrough needs to be, not forwards from last year’s result. The planner needs both numbers on one screen. Expect the first trajectory target to look unreasonable; that tension is the point, and catchball is where it gets settled.
Yes, until the annual reset. A new workbook each January closes last year’s reasoning, so the planning team starts from results rather than from the trajectory, which is exactly the drift this article describes.
