Strategic planning trends 2027: facts and context

Nobody writing a plan for next year knows what next year holds, and every plan is written in a register that suggests otherwise. The interesting question is not how to forecast better. It is what a plan can honestly contain when the forecast is unavailable.

This page does not tell you what is coming. It sets out how to write a plan whose value does not depend on being right about that, and how to decide which decisions genuinely have to be made now.

What to take away

• Separate forecasts, commitments and branches. They are written in the same voice and have completely different failure modes, and mixing them is why plans cannot be revised.
• The only decisions that must be made far ahead are the ones with long lead times. Make a list of yours; it is shorter than the plan.
• Replace dates with triggers wherever you can. "By the third quarter" survives contact with nothing; "when the second warehouse hits capacity" survives everything.

Three kinds of sentence about the future

A plan mixes these freely and the mixture is what makes it unrevisable, because you cannot tell which parts were supposed to change when the world did.

A forecast is a claim about conditions you do not control. Demand, prices, what competitors do, what regulators decide. It can only be right or wrong, and you get no credit for either. Write forecasts as claims with owners and check dates, and expect roughly half of them to be wrong.

A commitment is something you will do whether or not the forecast holds. Hiring, building a capability, exiting a market. Commitments cost the same whether the world cooperates or not, which is why the list should be short and each item should be defensible in more than one future.

A branch is a decision you are deliberately not making yet, with the condition that will make it. "If the pilot converts at a rate we can live with, we hire two more; if not, we stop." Branches are the cheapest things in a plan and the rarest, because writing one requires admitting you do not know.

Go through any draft plan and mark each line with one of the three. The unmarkable lines, and there will be several, are the ones that were never decisions.

Cheap scenario work

Formal scenario planning is a large exercise most organizations cannot afford and do not need. The small version takes an afternoon and produces most of the value.

Pick the two uncertainties that would most change what you do, not the two that are most discussed. Cross them. You now have four boxes. For each box, answer one question only: what is our first move in this world?

Then look for the move that appears in more than one box. That is the thing to do now, and you have just found it without needing to know which box you are in. Moves that appear in only one box are branches, and they get a trigger rather than a date.

Two disciplines keep this from becoming an away-day exercise. Give each box an ugly name that describes the condition rather than a mood, so that later you can tell which one you are in. And write down what you would see first in each, because the value of scenario work is almost entirely in recognizing the world early, not in the plans themselves.

Lead time decides how far ahead you must commit

Most of the anxiety about long-range planning dissolves once you separate decisions by their lead time.

Make the list explicitly. What in your operation takes more than six months to change? Usually a small set: hiring a scarce skill, a lease, a certification or approval, a supplier relationship, a system migration, anything requiring training a lot of people. Those decisions have to be made early, in ignorance, and they deserve the planning effort.

Everything else can be decided later with better information, and deciding it now buys you nothing except the feeling of having a complete plan. Worse, it converts a future choice into a present commitment that somebody will defend.

The practical output is two documents. A short list of long-lead commitments with real dates, which is the part that must be right. And a longer list of intentions with triggers, which is allowed to change without anybody being wrong. Confusing the two is why revising a plan feels like an admission of failure.

Lead times are also worth measuring rather than assuming. Most organizations discover that the thing they treat as slow has become fast and the thing they treat as fast now takes a quarter, usually because a handoff was added; operating processes is where that gets diagnosed. The technique for laying out where the calendar time actually goes, as opposed to where the work happens, comes from value-stream mapping.

Triggers instead of dates

A date is a guess about how long something takes, dressed as a commitment. A trigger is a condition that will be observably true or not.

Compare "we will expand capacity in the third quarter" with "we expand capacity when the queue holds more than a fortnight of work for three consecutive weeks." The second one does not need to be revised when the year turns out differently. It also removes the argument, because the condition was agreed before anyone had a position.

Good triggers share three properties. They use something you already collect, or the trigger will quietly become a manual judgment. They are stated as a level and a duration, because a single reading is noise, and telling a run of readings apart from a single odd one is exactly what a run chart is for. And they are attached to a named action and a named person, otherwise the condition is met and everybody waits to see whether anyone else moves. Choosing numbers that can actually carry that weight is the subject of strategic planning metrics.

There is a limit. Some things cannot be triggered because the response takes longer than the warning: if the reaction has a nine-month lead time and the signal gives you six weeks, you need the commitment, not the branch. Knowing which of your responses are in that category is worth an hour on its own.

Preserving the ability to change your mind

Where two options cost about the same, prefer the one that is easier to reverse, and spend the difference in flexibility rather than in efficiency. That is not timidity; it is buying information.

Some concrete forms this takes. Contract lengths and break clauses. Doing something manually for one more quarter rather than automating a process that is still changing shape. Building the smaller version of the thing. Hiring one person and seeing what the work actually is before hiring four.

The counterweight is that optionality has a running cost, and organizations that keep every door open pay it forever without ever walking through one. At some point the information stops arriving and the choice is simply being avoided. A useful discipline is to name, when you preserve an option, what you are waiting to find out and when you expect to know. If you cannot answer, you are not keeping the option open, you are postponing.

What a plan for a distant year can honestly say

Strip out the forecasting and something usable remains:

• The long-lead commitments, with dates, and what each one costs if the world turns out otherwise.
• The capabilities you intend to have regardless of which scenario arrives, because they are useful in most of them.
• The work that stops, which is what makes room for any of it, and which the strategic planning pillar treats as the real test of whether anything has been decided.
• The branch list: conditions, actions, owners.
• The assumptions you are prepared to have proved wrong, with the dates on which somebody reports.

That document is shorter than a conventional plan and does not read as impressively. It also does not need rewriting when something surprising happens, which is the only test that matters.

Any of it that changes how people work will still need the same handling as any other change, and the fact that a trigger fired rather than a manager decided does not remove that; change management covers the part that people experience. And when you find yourself reaching for a claim about where the industry is heading, apply the tests in management foundations trends to it first, including to your own.

Common questions

Our board wants a three-year number. What do we give them?

Give the number, with the two or three assumptions it rests on stated beside it, and say which one you would check first. A single figure with no exposed assumptions is worse for the board than a figure with them, because when it turns out wrong nobody can tell which part was wrong.

Does this not just mean we never commit to anything?

The opposite. Naming the long-lead commitments makes them harder to avoid, because they are no longer buried in a list of thirty items of which most are intentions. Fewer commitments, held more firmly, is the intended result.

How often should the branch list be looked at?

Whenever the condition on any branch could plausibly have been met, which for most organizations is monthly and takes a few minutes. The failure is not looking too rarely; it is that the list sits in a document nobody opens, and the condition is met without anyone noticing for a quarter.