Hoshin planning: choosing the few objectives the plant will genuinely pursue

The discipline is subtraction, and it is unpopular

A leadership team asked for its priorities will produce a page of them, because each item has a sponsor and dropping one is a visible defeat for somebody. The method's contribution is a forced constraint: a handful of breakthrough objectives, each with an owner, resource and a stated measure, with everything else explicitly parked and the reason recorded. The parked list matters as much as the chosen one, since it tells managers further down that they may decline requests which do not serve the agreed few. Without that permission, deployment adds work rather than concentrating it.

Catchball is a negotiation, not a cascade

The distinguishing mechanic is that objectives travel downwards as intent and come back upwards as commitments with the means attached. A department told to shorten lead time responds with what it can achieve, what it needs, and where the target is unrealistic given the equipment and people it has. That exchange usually happens more than once and is meant to be argumentative. Skipped, targets arrive as instructions, get accepted publicly and disowned privately, and review meetings become variance explanations. The exchange also surfaces conflicts between departments early, which is the other reason to insist on it.

Breakthrough work is not the same as running the business well

Deployment covers things requiring the organisation to become different: a new capability, a step change in delivery, entering a market. It does not cover keeping quality stable, meeting the schedule and looking after equipment, which belong to daily management and need their own routines and measures. Confusing the two yields either a plan stuffed with business-as-usual items dressed as objectives, or an improvement programme consuming the attention needed to run the plant. A useful test is whether the objective would appear again next year in the same words; if so, it is daily management.

The review rhythm is the mechanism

A plan reviewed annually is a document. One reviewed on a regular cycle, with owners present and current status visible, is a management system. The review must be capable of concluding that an objective is failing and either resourcing it properly or stopping it, otherwise everything stays amber until the year ends. Two habits protect it: examining the countermeasures and obstacles rather than the proportion of tasks completed, and holding the meeting where the work is often enough that the reported picture can be checked against the physical one.

How to tell a deployment has become a spreadsheet

The signs are consistent. Objectives are phrased so success is unmeasurable. Every department's contribution is a project it had already started. Progress is reported as tasks completed rather than movement in the measure. Owners send deputies to reviews. The matrix is maintained by an analyst nobody on the floor has met. At that point the deployment is a reporting overhead and the organisation is being steered by whatever the monthly figures demand. Recovery usually means cutting the objective list hard, in public, and rebuilding the exchange with departments that stopped believing in it.

Frequently asked questions

How do we know we have chosen too many objectives?

Ask each owner what they gave up in order to take theirs on. If nobody can name something they stopped doing, the objectives are additions to an already full workload and most will drift. A second test is resourcing: for each objective, name the people whose time is committed and check whether the same names keep appearing. Overlapping ownership across many objectives is the usual mechanism by which a plan becomes quietly unachievable.

How does this differ from the goal-setting frameworks used elsewhere?

The mechanics look similar and the emphasis differs. Policy deployment insists on very few objectives, on a two-way negotiation before commitment, on separating breakthrough work from daily management, and on reviewing obstacles rather than scores. Frameworks that cascade goals downwards without the negotiation tend to produce alignment on paper. The other distinction is an expectation that missing a target triggers analysis of why the countermeasures did not work, rather than a rating conversation about the owner.

What should happen when an objective is clearly going to be missed?

Say so early and treat it as information about the plan rather than about the person. The review should establish whether the obstacles were foreseeable, whether the promised resource actually arrived, and whether the objective is still worth pursuing at all. Legitimate outcomes include adding resource, changing the approach, extending the horizon, or stopping altogether. What corrodes the system is an objective known internally to be dead while it continues to be reported as recoverable.