Turning Strategy into Five KPIs a Team Can Act On

Most strategies fail between the boardroom and the team. A short, well-defined set of KPIs — each with an owner, a definition and a response — is how a strategy becomes weekly work.

Kudakwashe Verah 30 October 2026 5 min read

Most organisations do not lack a strategy. They lack a way of turning it into something a team can act on this week. Somewhere between the strategy document and the people doing the work, the intent gets lost — and what survives is usually a long list of numbers that nobody connects to a priority.

The evidence that this gap is real is uncomfortable. In a survey of about 11,000 managers across more than 400 companies, Donald Sull and Rebecca Homkes asked people to list their organisation’s top priorities. Only about a third of senior managers could name the top three — and even people involved in developing the strategy did only slightly better.¹ If the people closest to the strategy cannot name it, a team three levels down has little chance of working towards it.

KPIs are meant to close that gap. Too often they widen it.

Why most KPI sets do not work

The typical KPI set grows by accumulation. Each department adds what it already measures, each new initiative adds a few more, and nothing is removed. The result is thirty or forty indicators of equal visual weight, many of them describing activity rather than progress, and none of them clearly owned.

Three problems follow:

No line of sight. A team cannot see which of its numbers relate to which strategic priority, so all of them feel equally important — or equally ignorable.
No definitions. Two managers quote different values for the “same” KPI because they calculate it differently, and the meeting turns into a debate about the number instead of the business.
No response. When an indicator turns red, nothing specific happens. A KPI that triggers no action is a statistic, not a management tool.

Why five

Five is not a magic number. It is a discipline. A team can keep five measures in mind, discuss all of them in a weekly meeting and know which ones it can actually move. Beyond that, attention spreads thin and the list starts to be read selectively.

The constraint is also useful because it forces choices. Agreeing which five measures matter most is, in practice, agreeing what the strategy means for this team. That conversation is often more valuable than the KPIs themselves.

From many measures to five. Each surviving KPI traces back to a strategic priority; the rest can still be reported, but they are not the team’s five.

From strategy to five: a working method

1. Start with two or three priorities, not the whole strategy. For a given team, ask which strategic priorities its work can genuinely influence. A field-service team might influence “reliable service” and “cost to serve”, but not “new market entry”.

2. For each priority, describe what success looks like in a sentence. “Customers’ faults are fixed on the first visit” is something a team can picture. “Operational excellence” is not.

3. Find the measure that would prove the sentence true. For first-visit fixes, that is the share of jobs closed without a repeat visit within a set number of days. Write it down before deciding whether the data exists — otherwise the available data quietly chooses the KPI for you.

4. Pair outcomes with drivers. Outcome measures (lagging indicators) tell you whether the priority is being achieved; driver measures (leading indicators) tell you early whether it is likely to be. A team needs both. “First-visit fix rate” is the outcome; “share of jobs dispatched with the right parts” is a driver the team can act on today.

5. Cut to five. Keep the outcome measures that matter most and the driver measures the team controls most directly. Everything else can still be reported elsewhere — it just is not one of the team’s five.

Define each KPI so it cannot be argued with

A KPI that two people can calculate differently will eventually be calculated differently. Before a KPI goes on a report, it needs a short definition card that everyone signs up to.

A KPI definition card. Every KPI in the five gets one; the owner, response and counter-measure fields are the ones most often missing. Example values are illustrative.

The fields that matter most are the ones usually left out. Owner is a named person, not a department. Response says what happens when the value crosses its threshold: who is told, what is checked first, and by when. And source names the system and the exact field, so the number can be traced and reconciled rather than trusted on faith.

Watch for the number becoming the goal

Economist Charles Goodhart’s observation — usually paraphrased as “when a measure becomes a target, it ceases to be a good measure”² — applies to every KPI set. If “average call duration” becomes the target, calls get shorter and problems go unsolved. If “tickets closed” becomes the target, tickets get closed and reopened.

Three habits keep this in check:

Pair measures that pull against each other. Speed with quality, volume with rework, cost with service level. Improving one by damaging the other then shows up immediately.
Review the definition, not only the value. Once a quarter, ask whether the KPI still reflects the priority it was chosen for, and whether anyone is finding ways to move the number without moving the outcome.
Keep the sentence visible. Every KPI on a report should carry the success sentence it was chosen to prove. It reminds people what the number is for.

Make the five part of the weekly rhythm

Five well-defined KPIs still do nothing on their own. They need a regular place where they are looked at, by the people who can move them, with time set aside for the ones that are off track.

A good weekly review is short and has a fixed shape: the five numbers against their targets; for anything outside its threshold, the owner explains what is happening and what they will do; actions are recorded and checked the following week. Over time the conversation shifts from explaining the numbers to changing them — which is the point.

A practical checklist

Each KPI traces to one named strategic priority, in writing.
There are no more than five per team, with at least one leading indicator.
Each has a definition card: formula, source system and field, frequency, target, threshold, owner and response.
Measures that could be gamed are paired with a counter-measure.
The five are reviewed weekly by the people who can move them, and the definitions are reviewed quarterly.

Closing thought

A strategy becomes real when a team can say which five numbers show whether it is working, who owns each one, and what they will do when one slips. Getting to that point takes fewer measures than most organisations expect — and more agreement than most are used to. That agreement is the work.

D. Sull, R. Homkes and C. Sull, “Why Strategy Execution Unravels — and What to Do About It”, Harvard Business Review, March 2015; survey summary in London Business School news, 7 December 2015. ↩
C. A. E. Goodhart, “Problems of Monetary Management: The U.K. Experience” (1975); the common phrasing is Marilyn Strathern’s (1997). ↩