In this guide
Key Takeaways
- →Digital is not live. A matrix in a spreadsheet, slide or whiteboard tool is easier to edit and still has to be updated by someone.
- →Three conditions: values come from the work, correlations are records you can open, history is kept. Miss one and you have a digital poster.
- →Five things change: reviews move to exceptions, catchball continues all year, the matrix reaches below the top table, the reset starts from evidence, accountability gets specific.
- →Four new risks arrive with it: noise, status gaming, stale inputs and clutter. Each needs a deliberate countermeasure.
- →A live matrix that is quietly out of date is worse than a paper one, because people trust it.
- →The three conditions are a buying specification. Most products meet one or two, and the gap only shows in month three.
Putting an X-matrix into software doesn’t make it live. Three conditions do. When they hold, the review, catchball and the annual reset all work differently, and a few new problems appear that are worth planning for.
The life of a paper X-matrix
A typical X-matrix has a short working life. It’s drafted over a few days of planning, refined over a few weeks of catchball, then printed, pinned up and photographed. From that point it’s a reference document. Progress is reported elsewhere: priority owners in their own slides, KPI owners in the monthly business review, finance in its own scorecard.
Within a quarter the matrix and the business have drifted apart. A priority has been rescoped; a KPI definition has changed; an owner has moved on. None of this reaches the matrix, because updating it means reopening the file, reprinting and redistributing it, and nobody owns that job.
At year end, the next offsite starts from a fresh template. Some of last year’s thinking survives in people’s memories. Most of it, including which correlations worked and why targets were set where they were, is gone.
Digital isn’t the same as live
Moving that matrix into a spreadsheet, a slide or a whiteboard tool makes it easier to edit. It doesn’t change its life cycle. It’s still a picture that somebody has to keep up to date. A live X-matrix meets three conditions.
- Values come from the work. Progress on priorities and KPIs flows from the check-ins of the people who own them, not from an administrator copying numbers in before a review.
- Correlations are records. Each dot is something you can open: its grade, the owners on both ends, and the current state of each.
- History is kept. Plan against actual over time, who changed what and when, and the catchball discussions that set each target.
If any one of the three is missing, you have a digital poster. It will drift the same way the paper one did, only more quietly.
Five things that change
The review moves from rows to exceptions
When the matrix is current before the meeting starts, the review no longer needs to walk every row to find out where things stand. It can open with the exceptions: objectives in outcome lag, primary correlations whose lag windows have closed, bowler rows that have been red for two months. The time saved goes into the conversations that change decisions.
Catchball continues after planning
On paper, catchball is a phase. It happens between the offsite and the final matrix, and then stops. On a live matrix, the same negotiation can reopen any time the evidence calls for it: a target that needs revisiting, a priority that should be retired, a grade that no longer fits. Because the discussion is attached to the objective, the reasoning travels with it.
The matrix reaches below the top table
A printed matrix is read by the people who built it. A live one can be opened by a site leader who wants to see how their priority connects to a breakthrough objective, or by a team lead whose key result feeds a KPI on the matrix. That visibility is a large part of what hoshin kanri promises, and it is hard to deliver with a document that lives in a leadership folder.
The annual reset starts from evidence
When breakthrough objectives span several years and the matrix keeps its history, next year’s planning starts from this year’s record: which priorities delivered, which correlations held, which targets were renegotiated and why. The offsite stops being a blank-page exercise and becomes a revision of a living plan.
Accountability gets specific
On paper, a dot has no owner. On a live matrix, every correlation connects two named people: the owner of the priority and the owner of the KPI. When a link fails, the question isn’t whose row is red, but what the two of them believed about the mechanism and whether it held. That is a more useful conversation, and a fairer one.
What gets harder
A live matrix removes the problem of drift. It introduces four others. None is a reason to stay on paper, but each needs a deliberate countermeasure.
| New risk | What it looks like | Countermeasure |
|---|---|---|
| Noise | Monthly fluctuations trigger reactions to effects that haven’t had time to appear | Store lag windows on primary correlations; review against trend, not a single month |
| Status gaming | Owners learn to keep their row green by measuring completion, not change | Require at least one adoption-type key result on every priority |
| Stale inputs | Late check-ins mean the matrix looks current but isn’t | Run a fixed check-in cadence with reminders, and show when each value was last updated |
| Clutter | Adding a priority or a dot is so easy that the matrix grows all year | Keep the planning gate: new correlations pass the same checks as the original ones |
The stale-input risk deserves emphasis. A paper matrix is obviously out of date, and everyone treats it that way. A live matrix that is quietly out of date is worse, because people trust it. The check-in cadence is what keeps it honest, which is why a hoshin tool that only displays data, and doesn’t run the cadence, falls short of live.
Getting from poster to live
The move doesn’t have to be all at once, and it shouldn’t be. A sequence that works well:
- Load the current matrix as it stands, mid-year if that’s where you are. Don’t wait for the next planning cycle.
- Wire up the primary correlations first, so their two ends report live progress. Leave secondary and partial links as records for now.
- Run three monthly reviews from the live matrix, opening with exceptions, before cascading to sites or functions.
- At the next planning cycle, use the year’s record as the starting point and apply the full correlation gate.
The detailed version is in Migrating Your X-Matrix From Excel or PowerPoint and Implementing Hoshin Kanri Software: The First 90 Days.
A worked example: the same April review, twice
Here is what an illustrative leadership team knows walking into its April review, with the same plan run as a poster and as a live matrix.
| Question | Paper or spreadsheet matrix | Live matrix |
|---|---|---|
| How old is the data? | Collected 8 days ago by the program lead | Owners checked in over the last week; each value dated |
| Is any priority finished while its KPI is flat? | Possibly; the numbers are on different slides | Two objectives sit in the outcome-lag quadrant |
| Why was the services target lowered in February? | The regional lead remembers; it was in an email | The catchball session on the objective records it |
| How are the two pilot sites doing? | Their own files; not in this pack | Visible as contributing key results on the group objective |
| What will we decide? | Whether to ask red rows to accelerate | Whether one outcome-lag case is adoption or mechanism |
Illustrative.
The plan is identical in both columns. What differs is what the room can see without anyone preparing it, and therefore which decisions are available. On the left, the review can only react to colors. On the right, it can diagnose.
How to tell, before buying, whether a matrix will be live
The three conditions above are not a philosophy. They are a buying specification, and the useful thing about them is that every vendor’s demo will appear to satisfy all three. Each one has a question that separates a live matrix from a well-rendered picture of one.
| Condition | The question that tests it | What a picture sounds like |
|---|---|---|
| Values come from the work | Who typed the numbers I am looking at in this demo? | “We import from your spreadsheet” or “an admin keeps it updated” |
| Correlations are records | Can I click a dot, and what opens when I do? | A tooltip with a grade, or nothing at all |
| History is kept | Show me this matrix as it stood last September | “You can export a snapshot each month” |
| All three together | What does a reviewer see that nobody prepared? | A dashboard, which is the same answer as a deck |
The third row is the one most products fail and the one buyers notice last.
Row three is worth pressing hard. Export-a-snapshot is a real answer to a different question: it preserves what the matrix looked like, not why it changed. At the annual reset the useful record is the second one, and a product that offers monthly PDFs has quietly moved the reset back to a memory exercise. Ask to see a target that moved and the reason it moved, from nine months ago, in the product.
Row one is where a live matrix and good OKR software turn out to want the same thing. Both stand or fall on owners updating their own rows, which is why a tool with a heavy check-in is a tool that will end up with an administrator feeding it, and an administrator feeding it is condition one failing with the lights still on. Why hoshin software rollouts fail covers what that looks like three months in.
One scoping question belongs here too, because it is cheap now and expensive later: when a team key result or a scorecard perspective needs to feed this matrix, is that the same subscription or a second one? A matrix that is live only for the rows one department owns is a poster for everyone else. Purpose-built hoshin kanri tools vs strategy execution platforms works through that choice in full.
The live X-matrix in Profit.co
Profit.co’s X-matrix reads from the same check-ins that run its OKR and Balanced Scorecard programs, so priority and KPI progress update as owners report. Each objective opens into its check-in trend against plan, its contributing key results with weightings, and the KPI correlations mapped to it with their grades, expected lags and owners.
Profit.co’s exception queue flags each correlation whose hypothesis didn’t hold, and initiative escalation flags delivery problems, so outcome lag reaches the review as a named item rather than something someone has to spot. Each correlation keeps a strength history, and countermeasure A3s stay linked to the gaps they address, so the matrix carries its own record of what was learned. See how it fits together on the hoshin kanri software overview.
On the scoping question above, the answer here is one subscription: Hoshin Kanri, OKRs and Goals, Balanced Scorecard and Strategy Roadmap all sit in the Essentials plan, so a team key result and a board scorecard perspective feed the same matrix without a second purchase or a second check-in habit. Running balanced scorecard software alongside on its own subscription is where condition three usually breaks first, because the two systems keep separate histories of the same number. The 30-day free trial covers a full monthly cycle, which is the shortest honest test of condition one.
Free Product Demo
See your own matrix running live
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
An X-matrix whose progress values come from the people who own each row, whose correlations are records with owners and grades, and which keeps its history, so it stays current without anyone maintaining it by hand.
No. A matrix in a spreadsheet or slide is easier to edit but still has to be updated by someone. If any of the three conditions is missing, it drifts like a paper matrix, only less visibly.
Overreacting to monthly noise, owners gaming status by measuring completion, check-ins arriving late so the matrix looks current but isn’t, and the matrix growing because adding items is easy.
Ask who typed the numbers on screen, click a dot and see what opens, and ask to see the matrix as it stood nine months ago with the reasons things changed. Then ask what a reviewer sees that nobody prepared.
No. A snapshot preserves what the matrix looked like, not why it changed. The annual reset needs the second one, and a product offering monthly PDFs has moved the reset back to a memory exercise.
It will be, if team goals or scorecard perspectives sit in a separate subscription. A matrix that is live only for the rows one department owns is still a poster for everyone else.
