A portfolio needs scenarios, not a single annual plan

What the paper develops

A hiring freeze lands after the plan is approved. The plan assumed the people it needs, and it says nothing about what to do without them. Leaders call a meeting to reorder the portfolio's priorities, run in a hurry by whoever can attend, while each sponsor defends an initiative. I have supported portfolios that went through this. It is an observation from practice, not a measured finding.

This paper argues that a portfolio needs more than its plan: a small set of scenarios kept beside it. A scenario here is an alternative version of the portfolio, tested in advance so leaders know they could run it. Each one comes with a trigger, a signal someone can observe that tells leaders it is time to choose, and with the first moves leaders agree to make if it fires. The plan stays as the version everyone commits to, so "not a single plan" means not a plan alone. No source reviewed here compares portfolios that prepared alternatives with portfolios that re-planned after the surprise. The claim rests on reasoning and on existing methods, not on a measured result.

A re-forecast tells you what changed, not what to do by Friday. Only an alternative composed beforehand does. Scenario planning is an established method for preparing such answers. Paul Schoemaker describes it as a way to imagine possible futures that compensates for under- and overprediction of change. Hugh Courtney and his McKinsey coauthors pair a limited set of scenarios with trigger events to watch. Both write about strategy. This paper adapts the idea to the portfolio: alternatives composed against scarce people and dependencies, a named owner for each trigger, and first moves agreed in advance. That adaptation is untested as a whole.

Most parts hang on the alternatives. Start with two or three uncertainties that would change a choice, found by asking each initiative lead which assumption, if it broke, would change their plan. Build two alternatives you could actually run, which means real subtraction: a downside case that keeps every initiative is the same plan with a worse forecast. Give each alternative one page showing what it protects and what it gives up.

Then set triggers someone can see, each with one owner and a threshold chosen by lead time. "Market deterioration" is a mood. "Two straight quarters of pipeline below the planning level" is a fact. Agree the moves in advance, starting with a step you can reverse and the money that pays for it. Review the assumptions briefly, and retire scenarios whose uncertainty has resolved.

A good re-plan may do as well where surprises are slow and the portfolio has slack. The cycle earns its place where people are scarce, dependencies are shared, and decisions are slow. It costs the time to compose the alternatives, a regular check by each trigger owner, a short review, and money held back. The surprise that arrives may not be one you prepared, though composing the alternatives still shows which initiatives carry weight in every future. Sponsors may also bargain over a hypothetical cut, so the agreement is reopened only when a trigger fires or an assumption expires, and the one person with authority to move money and people decides.

Start with one portfolio where people and dependencies are scarce. Compose two honest alternatives first, because the triggers follow from them. Count the emergency reprioritizations, the hurried meetings that reorder priorities after a shock, for a year before you start. Then ask the first time a trigger fires: did leaders start from the prepared alternatives, or did the old emergency meeting come back? If no trigger fires, run the test on paper using the last real surprise.

The white paper covers what one plan costs when it breaks, the test that separates a scenario from a slide, trigger and move design, a vendor-risk illustration run from start to finish, and the objections and costs in full.