The strategic plan is adopted. In the planning retreat, staff picked twenty or so performance indicators, every one of them sensible at the time. Six months later, half the fields are empty, and nobody is quite sure who was supposed to fill them in.
This is almost never a discipline problem. It is a design problem. An indicator that does not get updated was usually the wrong indicator to begin with.
Why indicators stop getting updated
Three causes come up again and again in cities, counties, school districts and public agencies.
First, the data does not exist yet. Staff chose to measure something nobody compiles, so every update turns into a small project. Second, the indicator belongs to a department rather than a person. A department does not enter data, a person does. Third, and most quietly damaging, nobody ever looks at the result. An indicator nobody reads dies without a sound, and it usually takes others with it.
The fix fits in one sentence: choose fewer indicators, and choose them knowing where the data comes from, who enters it and who will read it.
Three families of indicators, three jobs
1. Base indicators
These describe the state of a plan element with no configuration at all: progress (from not started to complete), assessment (from very problematic to ahead of schedule) and priority. They do not measure a result, they describe a situation.
That is exactly what a leadership team needs for a portfolio review. One view shows what is behind and what is moving. They require no data collection, so they do not decay.
2. Custom indicators
These measure a real result, with a target, thresholds and a chosen display (gauge, traffic light, number against a target, symbols). Each one attaches to a specific element of the plan: a goal, a strategy, an action.
This is where you demonstrate results to a city council, a county board or a school board. It is also where the collection effort concentrates, which is why you have to be selective.
3. Global, or cross-cutting, indicators
A cross-cutting indicator is a single indicator linked to several elements of the plan that keeps the same value everywhere. One entry, one definition, one version of the truth.
It answers the most common failure in tracking spreadsheets: the same number entered in three places, eventually showing three different values. When that surfaces in a public meeting, it stops being a data question and becomes a credibility question.
Five tests for an indicator that lasts
- The data source already exists. If the data has to be created from scratch, the indicator will not survive its second update. Start from what the organization already compiles for other reasons, including state and federal reporting.
- A named person is responsible. Not a department, not a team. A person, who knows this is part of their job and how often it is due.
- A realistic frequency, agreed with that person. Monthly, quarterly, annual: it matters less than whether it is sustainable and written down.
- A target and thresholds set in advance. An indicator without a target says nothing. Deciding after the fact what counts as a good result drains the measure of its value.
- An audience. Who will read this result, and in which document? If the answer is vague, the indicator does not yet belong in the plan.
How many indicators do you need?
Fewer than the retreat produced. The rule that works best: one result indicator per measurable goal, not one per action.
Actions track perfectly well through base indicators and task completion. Keeping custom indicators at the goal level concentrates the collection effort where it informs a decision.
An organization that keeps ten indicators current all year reports better than one that defined sixty and documents twelve in the week before the annual report is due.
Three common mistakes
- Measuring activity instead of results. The number of meetings held is not a result. It becomes useful only once the intended result is measured somewhere else.
- Changing a definition mid-cycle without recording it. A definition that shifts without a trail makes comparison impossible, and the question always gets asked eventually.
- Saving indicators for the annual report. An indicator reconstructed once a year justifies decisions rather than informing them. Its value comes from its ability to trigger a decision during the year.
A note on dashboards
Public dashboards have become a standard expectation, and that is a good thing. But a dashboard is a display layer. It shows whatever the underlying indicators contain, including their gaps.
So the order matters more than it seems: fix the indicators first, publish second. A rich dashboard fed by empty fields does more damage than a plain page fed by ten indicators that are genuinely current.
Updating an indicator is a management moment, not paperwork
In organizations where tracking actually works, updating an indicator is not an isolated task emailed out two weeks before the report is due. It is attached to a meeting that already exists: a team meeting, a leadership huddle, a departmental review.
The responsible person shows up with the value, explains the variance if there is one, and records the context in a note. That note is often worth more than the number. When the annual report comes around, it saves staff from reconstructing from memory why a target slipped six months earlier.
This is also what separates a tracking tool from a data entry file. When the note, the data and the plan element live in one place, the context stays with the organization. When they live in three separate files, the context stays in one person's head, and it leaves when they do.
Match the measure to the horizon of the plan
The planning horizon shapes the kind of indicator that makes sense. Cities, counties and agencies responsible for public infrastructure often plan over a decade, because their decisions commit the community for a long time. Many school districts and nonprofits plan over three to five years instead.
Whatever the horizon, the same logic holds: the vision stays stable, the annual targets get revised. A result indicator attached to a ten-year goal needs interim milestones, or it says nothing for nine years. An indicator attached to an annual action plan needs to be closeable and replaceable without breaking the history.
Where to start
Take the indicator list from your current plan and run all five tests on each one: the data exists, a person is named, the frequency is agreed, the target is set, someone will read it. Any indicator that fails two or more needs rework or removal.
The exercise takes two hours with your leadership team, and it is usually the single best use of that time before a new planning cycle begins.
Building indicators that last
Volume 3 of our guide series, From Strategy to Measurable Action, covers how to build indicators and targets that hold up to rigorous tracking : https://planivore.app/en-us/guides/
