Leading Indicators — Measuring Progress When the Outcome Is 15 Years Away
If the final result is years away, what honest evidence of progress can you collect now?
Measure the earlier changes your explanation predicts, while continuing to track the eventual outcome. Early progress supports a hypothesis; it does not guarantee the final result.
What is a leading indicator?
A leading indicator is a measure intended to provide earlier information about a later result. In improvement work, it should have an explicit, testable connection to the change you expect—not simply be a convenient number available sooner.
Some indicators are predictors rather than causes. Others measure whether the intended intervention is reaching people or changing the work. Be clear which role each measure plays and how strong the evidence is for its connection to the outcome.
When the outcome may take 10–15 years, waiting for it alone is unhelpful. But replacing it with an easy activity count risks declaring success before the important result is known.
A six-step method
- Define the eventual result. Specify who should benefit, the outcome and the intended timescale. Avoid a goal so broad that almost anything counts as progress.
- Work backwards. Map the changes expected between today’s intervention and that result. Treat every arrow as an assumption to examine, not an established causal fact.
- Choose a few earlier measures. For each important link, ask what observable change would support or challenge the explanation. Include quality and possible harm, not only volume.
- Check feasibility. Confirm definitions, data access, collection burden and a suitable baseline. Do not promise a review date before checking when meaningful evidence could appear.
- Try to fool the measure. Could it improve while the intended outcome stays unchanged or becomes worse? Keep useful implementation measures, but do not mistake them for proof of benefit.
- Agree the review. Name an owner, predict the sequence and timing of changes, and state what would trigger investigation, adaptation or stopping.
This is a practical theory-of-change exercise: make the explanation visible so that evidence can test it.
Illustration: a Scottish football pathway
Suppose the long-term aim is a stronger pipeline of Scottish-developed players contributing to senior international performance. One proposed pathway might be:
- Improve the quality and accessibility of player development.
- Provide suitable competitive experience and progression opportunities.
- Develop players who establish themselves in senior professional football.
- Build a stronger pool for the national team.
This is an illustrative hypothesis, not a validated model of Scottish football. Selection, injuries, club opportunities and many other influences can affect progression. Qualification and tournament results also depend on opponents and chance.
What might you measure earlier?
- Development quality and access: assess the proposed coaching changes and who can use them. Accreditation alone may not show what players experience.
- Appropriate competitive experience: track participation and progression for defined player cohorts. More minutes alone do not establish better development.
- Sustained senior progression: track meaningful professional involvement, not just one-off debuts. Allow for the age and size of each cohort.
- Later contribution: follow sustained professional and international participation alongside the eventual performance goal.
These are candidate measures. Their definitions, data availability and relationship to later outcomes would need checking before use. Track exclusion, injury or other unintended effects where relevant.
As the programme matures, the questions change
Early reviews may ask whether the new arrangements were implemented and reached the intended group. Later reviews can examine progression and then the eventual outcome. Keep these stages connected rather than replacing yesterday’s measure whenever it disappoints.
An activity count is not automatically a “vanity metric”: it can show reach or delivery. The error is presenting more activity as proof of a better outcome without testing the intervening links.
Use charts to investigate, not certify success
Use time-ordered evidence where the data and question support it. A run chart, X-mR chart or retrospective step-change analysis serves different purposes; choose appropriately.
A fitted change does not prove that the programme worked. No fitted change does not, by itself, prove failure: check implementation, data, sensitivity and the expected delay. Avoid repeatedly changing the intervention in response to routine fluctuations or impatience.
Apply the same discipline elsewhere
In healthcare, a process measure might show that a new pathway is being used before patient outcomes could reasonably change. Pair it with access, quality and harm checks. The corridor-care evaluation example illustrates predictions and review—not an observed success.
Mixed results are a reason to investigate. If delivery improved but outcomes did not, possible explanations include a weak causal link, insufficient reach, a longer delay or another constraint. The pattern alone cannot tell you which is correct.
Keep the three questions connected
Right Problem: what prevents the desired result? Right Intervention: what change addresses it? Right Measures: what early and later evidence would test that explanation?
Use Types of Measures to distinguish implementation, outcomes and balancing effects. Keep the final outcome in view even when you must make decisions using earlier evidence.
How to take this further
Start with Right Measures in the improvement guide. Then use the other two questions to connect the explanation, the change and its evaluation.
Return to Why Improvement Programmes Fail for the complete framework.
Read your StepChange chart and decide what to check next → Use a chart alongside the intervention history and other evidence, not as a substitute for an explanation.
Method notes
The six-step method is this site’s practical synthesis, not a validated forecasting tool. Background: Carol H. Weiss, “Nothing as Practical as Good Theory” in New Approaches to Evaluating Community Initiatives (Aspen Institute, 1995); W. Edwards Deming, Out of the Crisis; Brian L. Joiner, Fourth Generation Management (1994). The football pathway and measures above are illustrative proposals, not claims about a current programme.