What strategists can learn from architecture

Managers routinely claim that their strategic planning process creates large, detailed documents, but often little else. It’s as if the process serves no purpose other than to create the plan, and execution is somehow separate.

An approach that we think might work better would be to treat strategy making as if it were a design process. We’re not the first to propose that strategy borrows from design; in HBR articles, Henry Mintzberg drew the analogy with the potter throwing a bowl and Roger Martin has made an explicit connection with design. But the aspect of design we want to focus on here is a bit different.

The key feature of the design process that interests us is the concept of “levels of design”, a notion that the creation of a design goes through a series of levels of increasing complexity and detail.

When an architect designs a new house, for example, he does it in stages of increasing detail. At the first level, he sets out a few basic principles that he and the client agree on: the kitchen should face east to catch the morning sun, the car port should be close to the kitchen to help unload groceries, the patio should face the South, there should be three reception areas on the ground floor, etc.

At Level 2 he draws a rough sketch of the building. It might include a basic plan for each floor, a plan of how the building might sit in its plot and a view from each side.

Level 3 is a scale blueprint with accurate measurements, detailing invisible but essential aspects, such as which walls are load bearing, and how the plumbing and wiring will be routed. It might include suggested positioning for the main bits of furniture and some important material choices, such as clay roof tiles or a wood floor in the living area.

Level 4 is the quantity surveyor’s list of materials and quantities: the number of clay tiles needed or the yards of copper wire. Final decisions are made about power points, about the taps in the bathroom and the colour of the kitchen wall, often delegated to lots of different specialists.

Level 5 involves the many issues that come up as the new house is being built. An extra power point is needed for the wi- fi router. The chosen wood-fired burner requires a bigger alcove than planned.

If we used the concept of “five levels of strategy”, we might be less likely to find ourselves with fancy plans and little action. We would still need strategic planning documents, but we would recognize them for what they are: rough sketches at Level 2. If we decide not to follow the strategy, we can leave it at this level. But, if we want to execute the strategy, we will realise that we need to develop plans at Level 3 and Level 4.

A Level 3 plan identifies the main organizational units responsible for different parts of the strategy and the operating model that links these organisational units together. For each unit, the plan specifies the outcomes expected, the timeframes, the way the unit will work with other units, any constraints on the unit and the resources available. The sum of outcomes at Level 3 will achieve the objectives defined at Level 2.

A Level 4 plan specifies the people, the money and the time needed for each sub-task within each unit. It also explains how units will continue with existing activity and the sequence to take on the extra tasks required by the strategy.

Just as the work of the quantity surveyor often throws up issues for the architect to resolve, so too will Level 4 plans raise issues to be resolved at Levels 2 and 3. Strategic planning should, therefore, include an iterative process for dialogue between levels. This requires a concept of levels of strategy and clarity about the work at each level. In the military, the dialogue process is called back-briefing. Interestingly, in strategy work, we do not have a good label for this activity.

Finally, Level 5 planning is about the inevitable adjustments that need to be made as events unfold. The money is not available when expected. Critical people leave with little prior notice. The competitor reacts aggressively to the plan. A supplier is late.

Sometimes events cannot be handled by adjustments at Level 5. Changes are needed at Levels 4 or 3 or 2. This is when clarity about levels is again helpful. Knowing when events require a Level 5 or a Level 2 response ensures that strategies do not wander off course unnecessarily or fail to correct course when circumstances demand it.

Maybe, just maybe, an awareness of levels of strategy work will help organisations do better strategic planning. It can’t very well do a lot worse.

Written by Andrew Campbell and Mark Lancelott. Published in Harvard Business Review.

Comments are closed.