← Back to SOPs & Operational Excellence

What is the difference between leading and lagging indicators?
Lagging indicators tell you what has already happened: missed deliveries, customer complaints, cost of rework, incidents. They matter, but by the time they move the damage is done. Leading indicators tell you whether the conditions for good outcomes are in place: are critical SOPs current, are people trained, are near misses being reported, are handovers complete? A healthy measurement set has both.
How do you connect measures to your mission?
Start with your Business DNA, not a list of popular KPIs. If your promise is next-day response, response time belongs on the dashboard; if it is technical accuracy, first-time-right rate does. I explain where those foundations come from in what Business DNA is. Your business sets its own objectives and targets; my work is helping the procedures that support them become visible and measurable.
Why isn't SOP adoption enough on its own?
It is tempting to measure how many procedures exist or how many people have read them. Those are activity measures, not performance measures. A team can sign every procedure and still deliver inconsistent work. Treat adoption as a leading indicator that should eventually show up in real outcomes; if it does not, the procedure may be the problem, which I discuss in why SOPs fail.
What should a balanced set of KPIs look like?
Here is an illustrative example, not a benchmark. The specific measures and targets should be yours:
- Customer promise (lagging): percentage of jobs delivered on the promised date.
- Quality (lagging): rework hours or cost per month, tracked as a trend.
- Handoffs (lagging): handover errors or rejected handovers between departments.
- Learning (leading): near misses reported per month — where a rise can be a good sign.
- Procedure health (leading): critical SOPs with a named owner and an in-date review.
- Readiness (leading): staff who completed the latest six-monthly refresher.
What does a practical dashboard look like?
Keep it to one screen and no more than eight measures. For each, show the current value, the trend over the last six months and a one-line comment from the owner explaining any movement. Group measures under the promises they protect, not under departments, so the conversation stays focused on the customer rather than on who is to blame.
How do you stop people gaming the numbers?
Any measure that becomes a target can be distorted. If incidents are punished, incidents stop being reported. Pair measures that pull against each other — speed with quality, low incidents with high near-miss reporting — and review trends alongside stories, not numbers alone. Aviation's reporting culture depends on people trusting that honest reports lead to learning; that is an analogy rather than a requirement for your business, but the logic holds. I cover reporting in more depth in how to manage operational incidents and near misses.
How often should you review the measures?
Review trends monthly with the owners, and revisit the measures themselves at each six-monthly update, retiring any that no longer drive useful conversations. Department leaders can apply the same approach within their own area, as described in aligning departments with company goals.
Frequently asked questions
How many KPIs should we track?
- Fewer than you think. Six to eight well-chosen measures, balanced between leading and lagging, usually give a clearer picture than dozens.
Is a rise in reported near misses bad?
- Not necessarily. It often means people trust the reporting process. Look at what is being reported and what is being learned.
Can you give us benchmark figures?
- I do not publish benchmarks. Your targets should come from your own baseline, mission and objectives.
Does the Company Operations Hub track KPIs?
- The Hub keeps procedures, owners, versions and review dates visible. Business performance measures remain in your own systems unless scoped otherwise.
Explore the Company Operations Hub, or book a discovery call to talk about what to measure first.


