Strategy Review

A strategy review examines whether the diagnosis, guiding policy, assumptions, and evidence still support the chosen actions. Delivery progress matters, but does not by itself establish that the strategy is sound. Richard Rumelt emphasizes that a clear challenge makes evaluation possible. If the diagnosis is unstated, first reconstruct the goals, constraints, assumptions, and reasoning behind current choices. The review can then decide what to preserve, clarify, change, or replace; absence of a written diagnosis does not require discarding everything.

Two clocks

Use periodic checks and event-driven assessment together. For example, an organization might review its technology roadmap quarterly and its broader strategy annually, while choosing shorter intervals where risk or change requires them. Assign owners to signals such as changed support coverage, new obligations, acquisitions, competitor developments, a challenged riskiest assumption, or an important measurement. A signal does not automatically reopen the diagnosis: assess whether it affects a task, a delivery commitment, priorities, policy, or the underlying challenge. A failed test may reject one solution while leaving alternatives worth examining. Check whether the evidence and the expected value of further learning justify more work. In the fictional insurance group, a support notice was not connected to a timely decision, so an expensive extension became necessary. A functioning review route could have exposed the choice earlier; its quality does not depend on a real event having already triggered it.

What a review may change

Review authority should match the impact of the evidence. A quarterly meeting can escalate a challenge to the diagnosis, while a routine notice may only require a delivery-plan update. Record the rationale, alternatives, owner, authority, and implications for obligations and commitments, including decisions to make no change. An architecture decision record is appropriate for architectural choices; other strategic choices can use a business decision record with equivalent reasoning. Preserve the earlier decision’s context and link its replacement when it changes. The Five-Year Plan Was Rewritten Every Year shows how this review process connects to the roadmap and its uncertainties.

References

• Richard Rumelt, The perils of bad strategy, McKinsey Quarterly, June 2011