Most cascades start too late. Sites receive the group plan after it’s locked and are asked to build theirs in January. The better sequence starts site planning during group catchball, so the two plans are negotiated together and lock together.
In this guide
- The late-cascade problem
- What sites receive, and what they decide
- A sequence that works
- The roll-up check
- Sites at different levels of maturity
- A worked example: the roll-up check
- What the cascade needs from the software
- What a late cascade costs
- Cascading in Profit.co
- Questions this article answers
Key Takeaways
- →A cascade that starts after lock is a one-way handover. Catchball between group and sites never happens.
- →Sites draft their plans in weeks six and seven, during group catchball, and use them as evidence in round two.
- →The group’s “how” becomes the site’s “what”. How the site gets there is its own plan.
- →Check before lock that site targets add up. A gap is a finding, not an inconvenience.
- →Pilot sites run the full sequence; others take objectives and track them without a full matrix.
- →A site that can show what its plan delivers negotiates better than one that says a target feels high.
The late-cascade problem
In a typical planning season, the group plan goes through its offsite and catchball, locks in December or January, and is then sent to sites with a request to produce site plans. Sites spend January and February doing so. By the time site plans are ready, a quarter has passed, and the site plans have had no chance to influence the group targets they are meant to deliver.
The cascade becomes a one-way handover. Catchball between group and sites, which is meant to test group targets against site reality, never really happens.
How site matrices should be structured once deployed is covered in Deploying hoshin kanri across multiple sites. This article covers the planning-season sequence.
What sites receive, and what they decide
| Sites receive from group | Sites decide for themselves |
|---|---|
| The group priorities that become site objectives | The site priorities that will deliver them |
| A proposed target or target range for each | Their commitment, through catchball |
| Standard KPI definitions | Any additional local KPIs |
| The review calendar and check-in cadence | Phasing of plan values to local seasonality |
The principle is the classic hoshin one: the group’s “how” becomes the site’s “what”. A group priority to bring eight sites to maturity level 2 becomes, at each site, the objective of reaching level 2. How the site gets there is its own plan.
A sequence that works
| Calendar week | Group | Sites |
|---|---|---|
| 4 to 5 | Offsite produces the draft group matrix | Site leaders briefed on emerging priorities |
| 6 to 7 | Catchball round 1 with divisions | Sites draft their plans against proposed targets |
| 8 to 9 | Catchball round 2 | Site catchball with group, using their draft plans as evidence |
| 10 | Final adjustments | Site plans finalized |
| 11 | Lock group and site plans together | Lock |
Within the twelve-week planning calendar.
The key change is that sites draft their plans during group catchball, not after it. Their drafts become the evidence in round 2. A site that can show what its plan delivers, and what it would take to deliver more, negotiates far better than one that can only say a target feels too high.
The roll-up check
Before locking, check that the site plans add up to the group plan.
- Targets sum. For each group KPI that is the total of site results, the site targets add up to at least the group target.
- Every site objective traces up. Each one serves a group priority.
- Every relevant group priority has sites carrying it. No priority depends on sites that haven’t taken it on.
- Definitions match. Every site calculates the shared KPIs the same way.
When site targets don’t add up, the gap is a finding. Either some sites need to commit more, or the group target is unrealistic. Resolve it before lock, openly, rather than locking a group target the sites have already said they won’t reach.
If the site targets don’t add up to the group target, you don’t have a plan. You have two plans and a gap between them.
Sites at different levels of maturity
Not every site can run full hoshin planning in its first year. A practical approach: pilot sites run the full sequence, with their own priorities, correlations and catchball. Other sites receive objectives and targets, agree commitments in a single catchball round, and track their objectives through check-ins without a full site matrix. They move to the full sequence in a later cycle.
A worked example: the roll-up check
The group priority is to bring eight sites to maturity level 2. Ten sites are eligible. Their round-two commitments:
| Sites | Commitment | Count towards target |
|---|---|---|
| Sites A, B, C, D, E | Level 2 by Q3 | 5 |
| Sites F, G | Level 2 by Q4 | 2 |
| Sites H, I | Level 2 next year; need a maintenance lead first | 0 |
| Site J | Already at level 2; aiming for level 3 | — |
Illustrative.
The commitments add up to seven, not eight. The planning team has three options: fund the maintenance lead so site H can commit, lower the group target to seven, or accept the gap as a recorded risk. It funds the hire, site H commits to Q4, and the plan locks at eight with every site’s commitment visible on the group objective. Without the check, the group would have locked at eight with seven committed, and found out in October.
What the cascade needs from the software
The roll-up check is arithmetic, and it has to run before lock rather than after. That is the whole product requirement. Where site commitments live in site documents, totalling them is a manual exercise someone does once, in week eleven, under time pressure. It gets skipped, and the gap in the worked example surfaces in October instead.
The sequence has a second requirement that is easy to miss. Sites draft in weeks six and seven, while group catchball is still open. That means two levels editing related plans at the same time, which only works if a site objective and the group priority it serves are the same object seen from two directions.
| What the cascade needs | What the tool has to do | How to test it in a demo |
|---|---|---|
| Group priority becomes a site objective | One link, not a copied row | Show the same objective at group and at a site |
| Parallel drafting | Sites editing while group catchball is open | Edit a site plan with the group round still running |
| Targets sum | Site commitments totalled against the group target, live | Change one site’s number and show the group total move |
| Every objective traces up | A filter for site objectives serving no group priority | Show me anything at site level that serves nothing |
| Shared definitions | A group KPI definition the sites inherit | Show this KPI at group and two sites |
| Lock together | Group and site plans locking in one step | Lock the plan and show me what the sites see |
| Mixed maturity | A site tracking objectives without a full matrix | Set up a site with objectives but no priorities |
Row 3 is the difference between finding the gap in week ten and finding it in October.
Row 3 is worth insisting on. A roll-up that updates as commitments change turns the check from an audit into a constraint: the room can see the shortfall while it still has options. Row 7 matters in any group where some sites are new to the method, which is most of them in the first two cycles.
What a late cascade costs
Count what the worked example avoided. The group locks at eight sites with seven committed; the shortfall surfaces in October, eleven months later, by which point the year is spent. One site has not started, one hire was never funded, and the breakthrough that priority served is a year behind. None of that is a performance failure. It is a check that happened too late.
On licensing, the shape matters more than the number because a cascade is defined by how many people are inside it. 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. Ten sites drafting plans in weeks six and seven is ten site teams, not ten seats, and per-seat pricing quietly argues for cascading to fewer of them.
Profit.co publishes two plans, Essentials and Enterprise, quoting against modules and scale rather than a public seat rate. Site objectives can contribute to group objectives as key results, which is what makes row 3 of the table above a live total rather than a spreadsheet. Hoshin Kanri sits in Essentials with OKRs and Goals, Balanced Scorecard and Strategy Roadmap, so a site carrying a group priority runs its quarters on the same data model rather than reporting into it. What drives the cost of hoshin kanri software covers what else to total.
The 30-day free trial is long enough to connect one site plan to a group matrix and watch the roll-up total move, which is the check this article turns on.
Cascading in Profit.co
In Profit.co, site objectives can contribute to group objectives as key results, so each site’s commitment appears on the group objective with its owner, weighting and status.
The roll-up check is built into catchball. Sessions run with a balance constraint on approval, so cascaded site targets have to add up to the group target before the session can close, and each site can counter only its own number. A gap like the one in the worked example surfaces as a session that can’t be approved, rather than in October. More on the planning software page.
That also supports mixed maturity. A pilot site can run its own priorities and correlations on a full site matrix, while a site that is not ready takes objectives and targets and tracks them through the same review cadence, with no matrix of its own until a later cycle.
Free Product Demo
Plan group and sites together this year
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
During group catchball, around weeks six and seven of a twelve-week calendar, so their draft plans can inform the final group targets.
The group priorities that become site objectives, proposed targets, standard KPI definitions and the review calendar. Sites decide their own priorities and phasing.
A check before lock that site targets add up to group targets, every site objective traces to a group priority, and every site calculates shared KPIs the same way.
Treat it as a finding before lock. Either some sites commit more, the group target comes down, or the gap is recorded as a risk. Locking over it just moves the discovery to October.
Run pilot sites through the full sequence. Other sites take objectives and targets, agree commitments in one catchball round, and track through check-ins without a site matrix until a later cycle.
Link a site objective to the group priority it serves rather than copying it, let sites draft while group catchball is open, and total site commitments against the group target live, before lock.
