Hoshin kanri is PDCA applied to strategy, and its tools are the stages of that cycle. The loop only works if every link holds. In practice one link breaks more than all the others: the step back from what was learned to the plan.
In this guide
Key Takeaways
- →Four links: matrix to bowler, bowler to countermeasure, countermeasure to A3, A3 back to matrix.
- →Links 1 to 3 hold because each has a visible trigger: a new plan, a red cell, a second red month.
- →Link 4 has no automatic trigger, and the matrix often belongs to someone who wasn’t in the A3.
- →A plan-change box on every A3, even when the answer is “no change”.
- →A standing quarterly agenda item that applies every completed A3’s plan changes.
- →Record which A3 prompted each change, so next year’s reflection can trace it.
Four links
| Link | From | To | What passes |
|---|---|---|---|
| 1 | X-matrix | Bowler | Which rows to track, with phased plan values |
| 2 | Bowler | Countermeasure | A gap, with its cause |
| 3 | Countermeasure | A3 | A gap that persists despite countermeasures |
| 4 | A3 | X-matrix | A change to the plan: target, grade or priority |
Links 1 to 3 run forward through the year and mostly hold, because each one is triggered by something visible: a new plan, a red cell, a second red month. Link 4 has no automatic trigger. Nothing on the A3 form says “now change the matrix”, and the matrix is often owned by someone who wasn’t in the A3.
The same cycle, seen at four different speeds rather than as four links, is in PDCA in Hoshin Kanri: The Cycle at Every Level of the Plan.
How each link breaks
Link 1: the bowler is rebuilt by hand
If the bowler is a separate workbook, someone copies rows from the matrix. Rows get missed, renamed or left with annual targets only. Fix: the bowler reads from the matrix.
Link 2: red without a cause
The row is red and the comment says “behind plan”. Fix: the countermeasure rule, cause, action, owner and date before the review. See Countermeasure Sheets in Hoshin Kanri: Tracking Actions on Off-Track Objectives.
Link 3: A3s never triggered
Rows stay red for months with a new one-line countermeasure each time. Fix: an escalation rule, an A3 after two consecutive red months. See When a Red Bowler Row Needs an A3 in Hoshin Kanri.
Link 4: findings don’t reach the plan
The A3 concludes the measure was wrong or the mechanism doesn’t hold. It’s presented, praised and filed. The matrix is unchanged. Fix: every A3 ends with an explicit plan-change section, even if it says “no change”, and the quarterly review acts on it.
An A3 that doesn’t change the plan when its findings say it should is a report about a problem, not a solution to one.
Closing link 4
Three practices make the step back to the plan routine rather than heroic.
- A plan-change box on every A3. Does this finding change a target, a correlation grade, a priority or a measure? Which, and why?
- A standing quarterly agenda item. Review every A3 completed in the quarter and apply its plan changes.
- A record on the matrix. The changed item notes the A3 that prompted the change, so next year’s reflection can trace it.
How that box is written, and what belongs in it, is in How to Write a Hoshin Kanri A3, Section by Section.
A worked example: one loop, start to finish
| Stage | Tool | Content |
|---|---|---|
| Plan | X-matrix | Pricing discipline priority graded primary against price realization; measured by launch of a new pricing tool |
| Do and check | Bowler | Tool launched in February: priority 100%. Price realization flat through April |
| Check deeper | A3 | Only 20% of quotes priced in the tool; sales teams use spreadsheets they trust more |
| Act | Countermeasure | Spreadsheet pricing disabled for large deals; tool training in three regions |
| Act | Catchball and matrix | Priority’s measure changed to share of quotes priced in the tool; correlation kept as primary |
| Plan, again | X-matrix | Next month’s bowler tracks adoption, not launch |
Illustrative.
The fifth row is link 4. Without it, the bowler would have shown the pricing priority at 100% for the rest of the year, and next year’s plan would have measured the next tool by launch too.
What a broken link 4 costs
The worked example prices it exactly. The pricing tool launched in February, so the priority read 100% from February onward. With link 4 broken, that row stays green for ten months while price realization never moves, the review has no reason to look at it, and the A3 that explained why sits in a folder.
Then the cost repeats. Next year’s planning team sees a priority that was delivered on time and a KPI that did not respond, with no record connecting the two, so they measure the next tool by launch as well. One broken link turns a single year’s mistake into a recurring one, which is the specific failure the reflection is supposed to prevent.
Among published alternatives for holding that link, OnStrategy lists $200 per month for unlimited users. i-nexus publishes a free Starter tier for up to five users and $28 per user per month at its Standard tier. The relevant test for either is narrow: when an A3 concludes that a measure is wrong, does anything carry that conclusion to the correlation it questions, or does it stop at the document?
Profit.co publishes two plans, Essentials and Enterprise, quoting against modules and scale rather than a public seat rate. Hoshin Kanri sits in Essentials alongside OKRs and Goals, Balanced Scorecard and Strategy Roadmap. What drives the cost of hoshin kanri software lists the rest to total, and the 30-day free trial covers one monthly cycle, enough to see links 1 to 3 running.
The loop in Profit.co
Profit.co closes all four links inside the product.
- Link 1: the bowler reads the same objectives as the X-matrix.
- Link 2: check-ins carry each month’s cause and action beside the value.
- Link 3: a countermeasure A3 opens from the exception queue, an initiative escalation or a red or amber bowler cell, anchored to the gap.
- Link 4: an Act decision on a correlation, to keep, downgrade or upgrade it, is written to that correlation’s strength history and linked to the A3. A decision that would change a committed target opens a catchball session; the original commitment stays as the baseline.
Link 4 is the one templates lose, and here it can’t be skipped: the A3’s Act step is where the revision closes. More on the Hoshin Kanri software page.
Free Product Demo
Find where your loop breaks
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
The PDCA cycle that connects the hoshin tools: the X-matrix places bets, the bowler tracks them, countermeasures and A3s address gaps, and their findings change the plan, which the next month’s bowler tracks.
At the step from A3 findings back to the X-matrix. A3s are written and presented, but the target, correlation or measure they question is rarely changed.
Add a plan-change section to every A3, review all completed A3s at the quarterly review, and record on the matrix which A3 prompted each change.
Because each of the first three is triggered by something visible: a new plan, a red cell, a second red month. Nothing on an A3 form says to change the matrix, and the matrix often belongs to someone who was not in the A3.
Someone rebuilds the bowler by hand from the matrix, so rows get missed, renamed or left with annual targets only. The fix is a bowler that reads from the matrix rather than copying it.
Yes. Stating it makes the decision deliberate rather than skipped, and leaves a record that the question was asked when next year’s reflection looks back.
