Signposts and Early-Warning Indicators: Know When to Change Course

Every plan rests on beliefs about the future: that a key customer will stay, that a supplier will keep delivering, that demand will hold. When one of those beliefs starts to fail, the warning signs are often visible well before the damage. The trouble is that nobody was assigned to look for them, or nobody agreed on what they would mean.

Signposts fix that. A signpost is a specific, observable event or threshold that tells you an important assumption is weakening. This guide draws on two well-documented sources: RAND's Assumption-Based Planning method, developed for the U.S. Army and published in 1993, and the indicators technique in the U.S. government's 2009 Tradecraft Primer on structured analytic techniques.

Start with the assumptions your plan depends on

RAND's report, A Planning Tool for Very Uncertain Times by James Dewar, Carl Builder, William Hix and Morlie Levin, begins with what can be known: the assumptions on which current plans are based. It defines an assumption as an assertion about some characteristic of the future that underlies an organization's current operations or plans. An assumption is important if its negation would lead to significant changes in those operations or plans.

The report sets out five steps:

1. Identify important assumptions.
2. Identify which of them are vulnerable within your planning time horizon.
3. Define signposts.
4. Define shaping actions.
5. Define hedging actions.

The first two steps matter more than they look. RAND observes that it is rare for an organization to set out all of its important assumptions explicitly. Most have to be dug out of documents, interviews and observation, and some unstated assumptions only surface on reflection. A useful exercise is to read your strategy or annual plan line by line and ask, for each commitment, "What would have to be true for this to work?"

What makes a good signpost

RAND defines a signpost as "an event or threshold that clearly indicates the changing vulnerability of an assumption." The report adds that the fundamental challenge is making sure the indications are unambiguous and not the product of a deliberate attempt to mislead.

In practice, that suggests a short checklist for each signpost:

• Observable: someone can actually see it, in your own data or in public information.
• Specific: it names a threshold or an event, not a mood. "Order volume from our top three customers falls more than 15 percent for two consecutive months" works; "the market softens" does not.
• Sourced: you know where the information will come from and how often it updates.
• Owned: one named person checks it on a set schedule.
• Hard to fake: it does not rely on a single source that has a reason to tell you what you want to hear.

Pair every signpost with an action

A signpost is only useful if everyone already knows what happens when it fires. RAND's method pairs signposts with two kinds of action.

A shaping action is designed either to avert or to cause the failure of a vulnerable assumption. It is about using whatever influence you have over the situation. A hedging action is meant to better prepare the organization for the failure of an important assumption. RAND describes it as an act of replanning: you rethink your plans as though the assumption had already failed.

The report notes that monitoring the signposts and taking these near-term actions are the plan elements that best prepare an organization for an uncertain future. Writing the response down in advance also takes some of the emotion out of the moment. When the threshold is crossed, the conversation is about executing an agreed step, not about whether things are really that bad.

Building an indicators list

The 2009 Tradecraft Primer, prepared by the U.S. government for intelligence analysts, describes a closely related technique called indicators or signposts of change. Its method is short:

1. Identify a set of competing hypotheses or scenarios.
2. Create separate lists of potential activities, statements or events you would expect to see for each one.
3. Regularly review and update the lists to see which indicators are changing.
4. Identify the most likely hypotheses or scenarios based on the number of changed indicators observed.

Two details are worth borrowing. First, the primer suggests developing two lists for each hypothesis or scenario: indicators that a development is emerging and indicators that it is not. That guards against only noticing evidence that confirms what you already believe. Second, it says an indicators list can "depersonalize" an argument when views are sharply divided, by shifting attention to an agreed set of objective criteria. If your leadership team disagrees about where things are heading, agreeing on the indicators first can turn a debate about opinions into a debate about evidence.

The primer's worked example rates each indicator quarter by quarter on a scale from negligible to serious concern, and also lists separate "trigger mechanisms" that could bring about a sudden shift. A simple spreadsheet with the same structure works for most organizations.

Linking signposts to your scenarios

If you have built scenarios, signposts are how you keep them alive. The UK Government Office for Science's Futures Toolkit notes that working through scenarios helps you recognize early warning signs as they happen. Shell's Scenarios: An Explorer's Guide describes an exercise in which participants collect evidence from the media that one or other scenario is currently developing, and it encourages people to scan for signals that the dynamics behind their scenarios are actually happening.

A practical approach is to write three to five signposts for each scenario: the early events you would expect to see if that world were starting to arrive.

A worked example

For example, imagine a hypothetical 12-person commercial cleaning company. Its plan assumes that its two largest clients, together a large share of revenue, will renew their contracts next year. That assumption is clearly important: if it failed, staffing and cash plans would change.

• Signpost: either client issues a request for proposals, asks for a price reduction of more than 10 percent, or changes its facilities manager.
• Shaping action: schedule a service review with each client before renewal season, and fix any recurring complaints.
• Hedging action: start building a pipeline of smaller clients now, so that losing one large contract would not force layoffs.
• Owner and cadence: the owner checks the signposts at a monthly account review.

Common pitfalls

• Vague signposts that no one can say for sure have been crossed.
• Too many indicators. A handful that are watched beats dozens that are not.
• Watching only for confirming evidence. Keep the "is not happening" list too.
• No owner, so nobody looks.
• Seeing the signpost and then debating it for months. Decide the response in advance and review it when you review the plan.

Key takeaways

• Signposts start with assumptions: list what must be true for your plan to work, then find the ones that could fail within your horizon.
• RAND defines a signpost as an event or threshold that clearly indicates an assumption's changing vulnerability.
• Make signposts observable, specific, sourced, owned and hard to fake.
• Pair each signpost with shaping and hedging actions agreed in advance.
• Keep separate lists of signs that a development is and is not emerging, and review them on a schedule.

Sources

• RAND Corporation, Dewar, Builder, Hix and Levin, Assumption-Based Planning: A Planning Tool for Very Uncertain Times, MR-114 (1993)
• U.S. Government, A Tradecraft Primer: Structured Analytic Techniques for Improving Intelligence Analysis (March 2009)
• UK Government Office for Science, Futures Toolkit (2024 edition)
• Shell International, Scenarios: An Explorer's Guide (2008)