Hoshin Kanri

Thinking Tools6.2 Creating an Action PlanAlso known as Policy Deployment

From postwar Japanese quality management, set out by Yoji Akao (Hoshin Kanri: Policy Deployment for Successful TQM, 1991 English edition).

Turns a few breakthrough objectives set at the top into each level's own contribution through back-and-forth negotiation, then reviews progress monthly over a yearly cycle.

One breakthrough objective deployed through three departments, and two targets that crept in

The year's objective sits at the root; each department's contribution was negotiated up and down until the parts added up to the whole. Select a box to read how its number was agreed. Two of the leaf targets do nothing for the objective: mark them, the check every hoshin review makes. Illustrative.

0 of 2 marked1 of 2 marked2 of 2 marked

Select a box to read how its target was agreed, then mark the two targets that do not serve the objective.Delivery time from 10 days to 5. The breakthrough objective for the year, set by the leadership team. The three departments' agreed cuts add up to it: 1.5 plus 1.5 plus 2 days.Order entry: 2 days to 0.5. Leadership first asked for zero. Sales operations said it could reach a day and a half without help from IT; with one developer for two months they agreed to half a day.Web orders feed the system. Ends rekeying emailed orders, which caused most of the wait.Auto-approve accounts in good standing. Nine in ten orders skip the credit queue.Redesign the commission plan. A real project, and it does nothing for delivery time. It crept in because sales had it planned anyway; it belongs in a different plan.Pick and pack: 3 days to 1.5. The warehouse offered one day off; the plant manager pushed back for two. They settled on a day and a half, with a second packing shift funded.Release orders every hour. Ends the every-other-day batch.Add a second packing shift. Clears the afternoon backlog the same day.Cut warehouse headcount 10%. A cost target from a different plan, and it works against the objective: fewer people means slower picking. Raise the conflict at the review.Transit: 5 days to 3. Logistics proposed the full two-day cut itself, once the cross-dock was in the capital plan.Daily carrier pickups. Up from twice a week.Cross-dock in the southern region. Takes two days off the longest routes.

An example

An industrial supplier's leadership team set one breakthrough objective for the year: cut order-to-delivery time from ten days to five. In earlier years such goals had gone out as a memo, and every department had added it to a list of fourteen priorities.

This time each department was asked what it could contribute, and the numbers went back and forth. Sales operations first said order entry could drop from two days to a day and a half; with one developer from IT for two months, it agreed to half a day. The warehouse offered one day off picking and packing, the plant manager asked for two, and they settled on a day and a half with a second shift funded. Logistics took two days off transit with a regional cross-dock. The parts added up to the five days. At the first monthly review, two targets on the plans were found to do nothing for delivery time, and one of them, a headcount cut, worked against it. Both were moved out of the plan. By December the average was 5.4 days.

Why it works

Strategy usually reaches the front line as a slogan, and each level fills it with its own priorities. Deployment works the other way: a few objectives at the top, and each level says what it can contribute and what it needs to do it. The negotiation, often called catchball, is the core of the method. It turns a demand into a commitment the people doing the work believe they can keep, and it surfaces the resources the top must supply.

The check that every target traces to an objective keeps the plan short and exposes pet projects and conflicting goals. Monthly reviews against the same tree show early which branch is off track, while there is still time in the year to fix it.

The thinking behind it

Yoji Akao Yoji Akao, a Japanese planning specialist, is recognized as the developer of hoshin kanri, the policy deployment method that cascades a few strategic goals through every level of an organization; his book on it reached English readers as Hoshin Kanri: Policy Deployment for Successful TQM (Productivity Press, 1991, translated by Glenn H. Mazur). The practice grew out of Japanese quality management in the 1960s, where the name first appeared at Bridgestone Tire. With Shigeru Mizuno, Akao also developed quality function deployment.

How to run it

  1. Set three to five breakthrough objectives. Choose the few changes for the year that would matter most, each with a number and a date.
  2. Ask each level for its contribution. Have every department propose what it can deliver toward each objective, and what it would need.
  3. Negotiate up and down. Pass the numbers back and forth until each level's targets add up to the one above and its owners believe them.
  4. Record the links. Draw how each target ties to an objective, in a tree or an X-matrix, so every target has a parent.
  5. Strike what does not tie. Move any target with no link to an objective into a separate plan, and raise any that conflicts.
  6. Review monthly against the tree. Check each branch's actual against its target, and agree on countermeasures where it is behind.
  7. Reflect at year end. Ask what the plan got wrong before setting next year's objectives.

Where it goes wrong

Too many objectives. Twelve breakthrough goals are a to-do list, and every department picks the ones it likes. Hold to three to five, and say which earlier priorities stop.

The negotiation is theater. If the top's first number always stands, people stop offering honest ones and pad their estimates. Change numbers when the lower level makes a case, and pay for what it needs.

The reviews become reports. A monthly meeting where each department reads its status changes nothing. Spend the time on the branches that are behind and on what will be done differently next month.

When to use something else

  • OKRs. When teams set their own quarterly objectives with less negotiation, and ambition matters more than exact roll-up.
  • Layered Actions Breakdown. When one goal needs breaking into actions for a single team, without a yearly cycle.
  • KPI Design. When the trouble is choosing the measures themselves, before any cascade.

Where it sits in SOLVE eX

Go to the source

  • Hoshin Kanri: Policy Deployment for Successful TQM, Productivity Press; English edition, a translation of his Hoshin kanri katsuyo no jissai, 1991

Back to the catalog