Strategy Refresh or Re-Plan? A Decision Guide

Six points behind plan, and a board member asks the question

Picture a leadership team three quarters into a three-year strategy. Revenue is six points behind plan. A new board member, two months into the job, asks the question out loud: is it time for a new strategy? Half the room hears an insult. The other half hears permission.

Both halves are skipping a step. A shortfall can mean the strategy is wrong, or that it was never executed, or that one supporting bet failed while the core choice is fine. Each calls for a different response, and a full re-plan is the most expensive of them. The thesis of this page: most calls for a new strategy are really calls for better execution or a targeted refresh. Reopen the core choices only when an assumption they rest on has broken.

The guide below asks three to six questions about evidence, execution and stakes, and routes you to one of eight outcomes, from “stay the course” to “re-plan on a clock.” Each outcome comes with the first steps to take. It takes about three minutes, and you leave with a named call you can defend in the room.

A strategy that was never executed has not failed. It has not been tried.

Review, refresh and re-plan are different jobs

Robert Kaplan and David Norton draw the line clearly. Strategy review meetings ask whether execution is on track, where it is failing and who will fix it. “These meetings take the strategy as a given” ¹. Testing the strategy is a separate stage: when managers find that assumptions underlying the strategy are flawed or obsolete, they reexamine it and decide “whether incremental improvements will suffice or whether they need a new, transformational strategy” ¹. That usually happens at the annual strategy offsite, but it could happen during the year after a major disruption or a new strategic opportunity ¹.

The effort row is not decoration. Kaplan and Norton report that developing an entirely new strategy may take two sets of meetings, each lasting two to three days, with research in between, and that in their experience strategies generally have three to five years of useful life ¹. A team that re-plans every time a quarter disappoints never gets to the end of one.

The method: follow the broken assumption

The guide starts with assumptions, not results, because results mix two signals. A shortfall where the work never happened is an execution problem. A shortfall where the work happened and the customers did not respond is evidence about the strategy. The guide asks which one you have before it lets you near a re-plan.

Kaplan and Norton's account of Store 24, a New England convenience chain, shows the cost of not asking. In 1998 it launched a strategy called “Ban Boredom,” built on the belief that an entertaining shopping atmosphere would set it apart. Within two years the executive team learned the strategy was not working and went back to an updated version of its earlier strategy of fast, efficient service. A Harvard Business School team later found that first-year store data already showed better execution of Ban Boredom going with worse store performance, and concluded that management could have learned a year earlier that the strategy was not working ¹. The assumption was testable a year before anyone acted on it.

What to do with your outcome

• Stay the course or clarify. The strategy is not the problem. If the leadership team tells it differently, the fix is to finish choosing, not to start again.
• Fix execution. Move money and people to the few initiatives that matter, then give the strategy a fair trial before judging it.
• Refresh. Handle the broken assumption as a single issue with its own decision date. Michael Mankins and Richard Steele found that companies taking issues one at a time, continuously, made more major strategic decisions a year than those running annual unit-by-unit reviews: 6.1 against 2.5 on average in their survey ².
• Test or stage. Buy evidence before you buy a new strategy, and keep hard-to-reverse money waiting for the signal.
• Re-plan. Write the diagnosis, then put real choices on the table. Mankins argues that management needs at least three alternatives, not minor variations on one theme, before a strategy is discussed or approved ³. Our offsite agenda builder structures the sessions.

Where a flowchart falls short

The strongest objection is that the question is rarely analytical. A new CEO wants a new strategy because it is theirs. A struggling unit wants a re-plan because a new plan resets its targets. A decision guide cannot outvote either.

True, and that is the case for using it in the room rather than alone. Answering six questions in front of the leadership team forces the politics into the open: someone has to say which assumption broke and show the evidence. If nobody can, the argument for a re-plan is about something else, and it is better named.

The second limit is harder. The guide can only test assumptions you have written down. Most strategies never list theirs, so the first “yes, we have evidence” arrives as a surprise. Whatever outcome you get, write the two or three core assumptions down now, each with the signal that would show it breaking. Next time, this question will take ten minutes instead of a quarter.

Questions

What is a strategy refresh?

A strategy refresh adjusts the targets, initiatives and budgets that sit beneath a strategy when a supporting assumption changes, while keeping the core choices of where to compete and how to win. A re-plan reopens those core choices. Kaplan and Norton describe the same split as incremental improvement versus a new, transformational strategy.

How often should a company re-plan its strategy?

There is no fixed interval. Kaplan and Norton report that, in their experience, strategies generally have three to five years of useful life, and that the strategy is usually tested at an annual offsite or during the year after a major disruption. Re-plan when a core assumption breaks, not because the calendar says so or a quarter disappoints.

How do I know if the problem is strategy or execution?

Check whether the planned work actually happened with the money and people promised. If it did not, the strategy has not been tested and the problem is execution. If the work happened and customers or costs did not respond as expected, you have evidence that an assumption behind the strategy is wrong.

Are the outcomes in this guide based on research?

The distinction between reviewing, improving and replacing a strategy draws on Kaplan and Norton's management system, and the cited figures come from the sources listed. The questions, routes and outcomes themselves are our rubric. They have not been validated, so use them to structure the conversation, not to settle it.

Where to go next

• The Annual Planning Cycle: Strategy, Plan and Operating Plan
• Strategy Execution: Where It Breaks Between the Offsite and Monday
• Strategy Offsite Agenda Builder
• Assumption Mapping | Strategy Lexicon