What types of strategy are there? Defining scope and focus

What should the strategy apply to?

This is the first substantive decision in the PLAN step, and it determines who sits at the table in the first place. Four levels of scope are possible.

On top of the organizational dimension comes the geographic one. Companies operating internationally usually run a group strategy, with local strategies for individual markets nested underneath. For mid-sized companies, this rarely amounts to more than one question: what applies everywhere, and what is the foreign subsidiary allowed to do differently? Leave it open and you get both at once: a headquarters that feels bypassed and a subsidiary that feels micromanaged.

A functional strategy cannot be developed in isolation. It has to be aligned with the levels above it, otherwise one area optimizes something the bigger picture doesn't even need. Digital transformation shows this again and again: IT gets a program and a budget, while at the company level it's still unclear what the whole thing is supposed to achieve.

What happens when the top level is missing?

This is the norm in mid-sized companies, and it's rarely named as such. Unit and function heads are asked to develop a strategy while there is none at the company level for them to align with.

The consequence is predictable. Every area sets its own priorities, every priority makes sense on its own, and together they pull the company in four directions. The friction then shows up as a budget conflict and gets treated as one, even though it's a strategy problem.

If that's where you are, there are two honest ways forward. Either you do the company level first, even if that costs three months. Or you write down explicitly, for your area, which assumption about the company level you're working from. That way it's clear later what the unit strategy was built on.

What kind of challenge are you actually facing?

Before you decide on focus, it's worth a minute on the question of how difficult your situation actually is. Richard Rumelt distinguishes three types, and each calls for different work.

• The choice problem. You understand how things connect and you know the alternatives. The only hard part is picking the best one under uncertainty. Deciding on a location belongs here.
• The design problem. There are no ready-made alternatives, but there are models and patterns from your own experience or from watching others. You can test a new approach against those patterns.
• The tangled problem. Neither alternatives nor transferable patterns. Cause and effect are unclear, and the data can be read in more than one way. Here the work starts with figuring out what the knot actually is.

The most common mistake is treating a tangled problem like a choice problem. What you get is a scoring matrix with five options that all rest on the same untested assumption.

What does your strategy aim at?

The type of challenge determines the focus. In practice, it sits in one of four fields, and each field has its own tool.

The tools behind the four fields are well known, and that's not what makes them interesting. What's interesting is that choosing the field determines the data you'll collect over the next six weeks. Skip this, and you gather a little bit of everything, ending up with plenty of material and little basis for decisions.

The assignment isn't a checklist to work through. It tells you which analyses you actually need in the next step. If your focus is the portfolio, you need numbers for each unit. If it's competitive advantage, you need customers telling you why they buy.

How do scope and focus fit together?

Not every field fits every level. A portfolio question at business unit level is usually a company-wide question in disguise, and a competitive question at group level almost always leads back to the individual market.

So check the combination explicitly. If you're developing a business unit strategy and the room starts talking about acquisitions, you're working on the wrong topic, or someone is missing from the table.

A common case in mid-sized companies: Management wants to grow and means revenue. Sales hears new customers in the existing market, development hears new products for existing customers. Both are growth, and both call for different investments. The word alone doesn't carry the decision.

How long should a strategy hold?

Three to five years at most is the range that has settled out in practice. There's no reliable forecast for how volatile the next few years will be, so this is experience, not calculation.

More important than the number is recognizing that the parts age at different speeds. A product strategy can be outdated after a year under competitive pressure, while the production strategy holds for five. The fast-aging parts belong on shorter review cycles.

More important than any date is writing down the assumptions the strategy rests on. Interest rates, raw material prices, availability of skilled workers, a law that's coming or isn't. Then the review is no longer a ritual in the calendar but a question with an answer: does the assumption still hold?

Company size sets part of the pace. Experience shows that large organizations need two to three years before a new strategy is developed, understood across the organization, and noticeable to customers. In mid-sized companies it moves faster, and that's an advantage that rarely gets used.