Two kinds of measure
Some measures tell you what already happened. Others hint at what may happen next. Both are useful, and confusing one for the other leads to either false comfort or needless alarm.
Lagging indicators report outcomes after the fact. They are usually what the organization ultimately cares about, but by the time they move, the causes are already in the past. Examples include revenue for a quarter, the number of security incidents last year, or the share of orders delivered late last month.
Leading indicators are earlier signals that tend to move before an outcome does. They are not guarantees; they are clues. Examples include the age of the oldest open request, the number of overdue maintenance tasks, or the share of staff who have completed training before a system launch.
A driving analogy
When driving, the speed you travel is something you can influence now, while the distance you cover by the end of the trip is the outcome. The fuel gauge is a signal that you may have a problem later. The analogy is imperfect, but the idea holds: you steer using signals you can act on, and you judge success by outcomes.
Illustrative examples
The pairs below are examples only. Whether a given early signal really predicts a given outcome depends on the organization, and should be checked rather than assumed.
| Area | Possible leading indicator | Related lagging indicator |
|---|---|---|
| IT support | Age of the oldest open request; backlog growth | User satisfaction scores; number of escalations |
| Change delivery | Share of staff trained before launch | Adoption rates and errors some weeks after launch |
| Procurement | Share of purchases following the agreed process | Spend leakage found at quarter end |
| Supply chain | Supplier confirmation rates on open orders | Late deliveries to customers |
| Finance | Number of reconciliations open as period end approaches | Time taken to complete the financial close process |
| Security | Share of accounts with overdue access reviews | Incidents traced to excess access |
In each row the early signal is something a team can act on before the outcome is fixed.
Strengths and weaknesses
Lagging indicators
- Strength: they measure what actually matters, and are usually harder to dispute.
- Weakness: they arrive late, often too late to change the result. They can also be affected by many things outside the team’s control.
Leading indicators
- Strength: they give time to act and are usually closer to the team’s own work.
- Weakness: the link to the outcome may be weaker than it looks. A team can improve a leading indicator without improving the result, particularly if people start to focus on the indicator instead of the goal.
That last risk is worth taking seriously. If, for example, staff are measured on training completion, they may complete training quickly without learning much. The completion rate rises, while the thing it was meant to predict does not improve. The article on target side effects covers this in more depth.
Checking whether an early signal really leads
Treat a leading indicator as a hypothesis. A reasonable approach:
- Write down why you think the signal should predict the outcome.
- Look at past data, if available, to see whether the two have moved together, and with what delay.
- Remember that moving together is not proof of cause. A third factor may drive both.
- Revisit the link periodically. A relationship that held under one process may change after a reorganization or a new system.
If you have a performance baseline, it is easier to see whether a signal’s movements actually preceded changes in the outcome.
Pairing the two
A balanced set of performance metrics usually includes a few of each, connected in a simple chain: an action, an early signal, then an outcome. For example, a team might track how quickly new starters receive their equipment (early), and later the time until those starters can work independently (outcome). If the outcome disappoints, the earlier measures help narrow down where the process broke.
Pairing also offers a sanity check. If the leading indicators look healthy and the lagging indicators keep worsening, the early signals are probably the wrong ones, or something unmeasured is going on.
Practical cautions
- Do not over-read short-term movement. Early signals can be noisy. A single week’s jump may not mean much.
- Allow for delay. An improvement in an early measure may take weeks or months to show up in outcomes, and that delay should be understood before judging a change too soon.
- Keep the set small. A long list of early signals makes it harder to see which ones matter.
- Define both kinds carefully. A leading indicator with a vague definition is easy to adjust until it looks good.
- Talk to the people doing the work. They often notice early signs long before any report does.
Some measures can be either kind depending on viewpoint. Customer satisfaction is an outcome of service quality, but it can also be an early indicator of customer retention. What matters is being clear about which role the measure plays in a given discussion.
Using them in reviews
In a regular review, early signals are often the better starting point for discussion because they point to action: what is trending, what needs attention this month. Outcome measures then confirm, over a longer period, whether those actions had the intended effect. Both belong in the agenda, in different proportions depending on how fast the work moves.