If you interview a leadership team separately and ask each member what the organization ought to be doing over the next three years, the answers usually converge more than anyone expects. The insight is present and reasonably well distributed. What is absent is agreement on order, and on what will be given up to make the order possible.
This is why strategy documents so often describe a direction everyone already accepted and produce very little change.
Capacity is the binding constraint, and it is rarely modelled
Strategic plans are typically constrained against capital. Capital is measurable, its availability is known, and finance owns the model. Execution capacity — senior attention, engineering throughput, the number of significant changes an operation can absorb in a year without degrading — is rarely constrained at all, because no function owns it and it is uncomfortable to quantify.
The result is a portfolio of initiatives that is affordable and undeliverable. Each initiative has a sponsor, a budget and a business case. Collectively they require more senior attention than exists. Everything proceeds at sixty percent, nothing reaches the threshold where it starts returning, and the following year’s plan attributes the shortfall to execution.
Sequencing decisions have compounding structure
The order matters more than the list because initiatives are not independent. Some create the conditions for others; some consume the same scarce resource; a few actively foreclose options that a later initiative would need.
Three relationships are worth mapping explicitly before a plan is approved:
Enabling. Which initiatives make later ones cheaper or possible? A pricing capability built first makes three subsequent commercial moves straightforward. Attempted concurrently, all four contend and none land.
Contending. Which draw on the same constrained resource — usually the same twelve people, not the same budget line? Two initiatives that both require the head of operations for forty percent of the year are, in practice, one initiative and a delay.
Foreclosing. Which decisions remove future options? Exclusive distribution agreements, platform commitments, and acquisitions all narrow the subsequent decision space, sometimes considerably. Taking them early because they are available is how organizations arrive at year three with a strategy they can no longer execute.
What to stop is the harder half
Every strategy is a resource allocation, which means every strategy is also a set of withdrawals. In practice the withdrawals are usually left implicit — the plan describes what will be added and assumes the organization will find room.
It does not find room. It spreads the same capacity more thinly and the new work performs worse than it should, which then reads as evidence against the strategy rather than against the allocation.
Naming the stops is unpleasant because they belong to people. A product line, a market, a channel, a customer segment — each has an internal advocate who will argue, often correctly, that it remains profitable. The relevant question is not whether it is profitable. It is whether the senior attention it consumes is worth more elsewhere. That is a genuinely difficult judgement and it is the one the plan exists to make.
A practical test before approval
Before a strategy is signed off, three questions expose most of the sequencing weakness:
- If we could only do the first two initiatives, which two, and what does the plan look like at the end of year one? Plans that cannot survive this question are lists rather than sequences.
- Which named individuals are required by more than one initiative, and for what proportion of their year? This is usually where the plan quietly breaks.
- What are we stopping, and who has been told? If the answer is nothing, capacity has not been allocated — it has been assumed.
The recommendation
Strategy work that produces a clearer articulation of a direction the organization already held has, in most cases, produced very little. The value sits almost entirely in the sequence, the capacity model behind it, and the explicit list of withdrawals that makes it feasible.
That work is less satisfying to present and considerably more useful to own.