2026 GOVERNMENT BENCHMARK SERIE

S

THE STATUS RULE PLAYBOO

K

The governments that

catch problems first

track 9× fewer measures.

What the plans of 199 governmen t & public-sector organizations reveal about why performance problems surface at qua rter-end — and the 90- day plan to fix it. Measured, not surveyed.

1

9

government & publ ic- sector organizati ons analyzed

194,601

performance measures in the sample

87 vs 754

median measures: b est vs. worst quartile on statu s- rule coverage

43.3%

median share of measure s with a status rule

Extracted from the ClearP oint production platform · August 13, 2 026

clearpointstrategy .com/government-performance-benchmark

THE HEADLINE FINDIN

G

87 vs. 754: the portfolio-size trap.

We expected the best-instrumented governments to be the biggest, best-staffed ones. The data says the opposite: the strongest predictor of status-rule coverage in our sample isn’t tooling or headcount — it’s how few measures the organization tries to track.

▲ TOP QUARTILE — COVERAGE ≥ 65%

Median measures trac ked

87

Median status-rule c overage

75.0%

Orgs using reminder email s

24%

Orgs using notification s

26%

Phantom-owner rate

67.2%

▼ BOTTOM QUARTILE — COVERAGE ≤ 13%

Median measures track ed

754

Median status-rule cov erage

5.3%

Orgs using reminder emails

50%

Orgs using notifications

52%

Phantom-owner rate

73.1%

Quartiles c omputed on status -rule coverage (ev aluated measures ÷ to tal measures) acros s 199 organization s · n = 50 per qu artile · August 13 , 2026. The mechanics are mundane. Setting a defensible threshold takes a real conve rs ation: what is our baseline, how much variance is noise, at what p oint does someone act. A performance office can have that conversation 87 times. Nobody can have it 754 times. So the oversized portfolio ships without rules, and the rul es never catch up. “You can’t threshold 754 measures. You can threshold 87. The prune is the control system.” Note what this finding is not : it is not “measure less, know less.” The top quartile still rep orts everything it needs to — but it separates the managed portfolio (fewer measures, every one with a rule and an owner) fro m reference data (context numbers that don’t need a threshold). The bottom quartile mixes the two , a nd the signal drowns.

The tooling trap

Bottom-quartile organizatio ns lean on reminders and notifications at twic e the top quartile’s rate. It makes sense — when no rule watches the num ber, you have to nag a human to look at it. But a reminder checks that som eone updated; a status rule checks what the number says. The first is hy giene. Only the second is control.

THE DAT

A

Where does your coverage actually sit?

DISTRIBUTION · STATUS-RULE COVERAGE PER ORGANIZATION

How much of their measure portfolio do gove rnments actually evaluate? 0–10% of measures evaluated

43 orgs

10–25%

33 orgs

25–50%

41 orgs

50–75%

57 orgs

75%+ — the citable ceiling

25 orgs

Source: ClearPo int Government Performa nce Benchmark · 199 government & public-se ctor organizations with at least one me asure · August 13, 2026.

Find your percentile

IF YOU EVALUATE

…

YOU ARE AT…

READ IT A

S

≤ 13% of your measures Bottom quartile Your dashboard is a scrapboo k. Numbers exist; signals don’t. ~ 43%

Median Half your portfolio can raise its hand. The other half fails silently.

≥ 65%

Top quartile Drift gets caught in-cycle. This is the workin g standard.

≥ 77%

Top decile The measure portfolio is fully instrumented. Rare — 1 org in 10.

THE DATA, BY SEGMEN

T

No segment is exempt from the pattern.

SEGMENT

ORGS

MEDIAN STATUS-RULE

COVERAGE

MEDIAN

MEASURES

PHANTOM-OWNER

RATE

Counties

65

46.4%

435

73.8%

Utilities, district s & authorities

1 1

39.9%

319

58.4%

Cities & towns

95

32.2%

330

76.3%

Other public sector

25

59.1%

328

83.4%

74.5%

of gov plan owners never log an update (n = 4,605)

78.7%

of measures, pooled, carry no status rule at all

43.3%

median organization’s coverage — the number that matter s

The playbook in one sentence:

Prune the portfolio until ev ery measure can state its own red line, s et the rule once, name a real owner — and let the platform do the watc hing. The next two pages give you the 90-day plan and the threshold worksheet.

THE PLAYBOO

K

The 90-day status rule program.

Sized for a five-department rollout with one performance l ead. The order matters: pruning comes first, because every later step costs effort per measure — and the data shows the portfolio’s size is wha t kills coverage. 1 WEEKS 1–2 · PRUNE

Split “managed” from “reference.”

Run every measure through o ne test: “Can we state, in numbers, the point at which some one must act?” If yes, it’s a managed measure. If no — it’s reference data: ke ep reporting it, but move it out of the managed portfolio. Benchmark target: a managed portfolio near the top-quartile median ( ~90–150 meas ur es for a mid-size government), not the bottom-q uartile 754. 2 WEEKS 3–4 · THRESHOLD

Set the rule in a room, not in a spreadsheet.

One workshop per department. For each managed measure, fix three numbers with the workshee t on the next page: the baseline, th e caution line, the action line. Write down who chose the threshold an d why — that sentence is what survives the co uncil question “why is this red?” 3 WEEKS 5–8 · OWN

Ownership is agreed, not assigned.

74.5% of owners never update because ownership was typed into a field, no t accepted by a person. Re- confirm each managed measure’s own er face-to-face, with the cadence stated (“monthly, by the 5th”). One owner per measure; a committee is a p hantom with extra steps. 4 WEEKS 9–12 · RUN ONE CYCLE

Let the first reds happen.

Run a full reporting cycle with rules live. Expect 10–20% of managed measures to flag — tha t’s the system working, not failing. Triage each flag in th e leadership meeting: real drift, wrong threshold, or stale d ata. Adjust thresholds once, then fre eze them for two quarters.

What “done” looks like

Every managed measure has a rule, an owner, and a dated update from the last cycle. On the ClearPoint benchmark that puts you at ≥65% coverage — top quartile — and, more us efully: the next time a number drifts, you find out that week, not at quarter-end.

THE WORKSHEE

T

The 6-step threshold worksheet.

STEP

QUESTION TO ANSWER

WORKED EXAMPLE — ILLUSTRATIVE, “POTHOLE RESPONSE TIME”

1 · Direction Is lower better, higher better, or is there a target band?

Lower is better (days to repai r).

2 · Baseline Median of the last 8–12 periods — no t the best month, the median.

Median of last 12 months : 5.1 days.

3 · Noise band How much did the measure move in periods where nothing was actually wrong?

Normal seasonal swing: ±1 day .

4 · Caution line Baseline + noise. Crossing it means “watch this,” not “act.”

Amber at > 6 days.

5 · Action line The value at which a nam ed person must do something specific. Tie it to a commitment (charter, budget book, council goal ) when one exists.

Red at > 8 days — the respons e standard published in the serv ice charter.

6 · Rationale sentence

One sentence: who set it, on what basis, when it’s next reviewed.

“Set by Public Works leade rship from FY24–25 baseline; review July 2027.” Why we don’t publish “benchmark thresholds” per KPI. Our platform data tells us whethe r a measure has a rule — deliberately, it does not tell us what your pothole target should be. A threshold copied from another city is a threshold you can’t defend. The defensible number comes from your baseline, your noise band, and your published commitments — which is exactly what this worksheet extracts. Any v endor selling you universal government KPI targets is selling folklore.

“You can’t threshold 754 mea sures. You can threshold 87. The prune is the control system.”

METHODOLOG

Y

Where these numbers come from.

Population. 199 organizations classified as governme nt or public-sector (by organization-name pattern: city, county, town, state, district, authority, agency, transit, police, co uncil, ministry, public) with at least one performance measure, out of ClearPoint Strategy’s prod uction platform. The sample covers 194,601 performance measures and 4,605 named plan owners . Extraction. Aggregated, de-identified counts extracted August 13, 2026. No customer is i dentifiable; segment cuts are suppressed below n = 5 (stat e agencies, n = 3, are excluded from segment tables). Definitions. A status rule is an evaluation threshold configu re d on a measure. Coverage is evaluated measures ÷ total measures, computed per org anization; distribution statistics are medians and percen tiles across organizations, not pooled averages — except where marked “pooled.” A phanto m o wner is an active user who owns at least one plan ele ment and has zero update records. Quartiles are computed on coverage (n = 50 per quartile). Honesty notes. The pooled “78.7% of measures have no rule” and the median-o rg anization “43.3% coverage” are both true — the pool is dominated by a few very large portfolios; we report both so you can’t be mi sled by either. Segment assignment by name-matc hing is a heuristic and can misclassify hybrid entities. Correlatio ns are Pearson r on organization-level valu es; with n = 199, |r| = 0.01 is indistinguishable from zero. We re-extract quarterly; figures carry their extraction date. Worked examples in the worksh eet (pothole response time) are illustrative, not customer data.

© 2026 ClearPoi nt Strategy · The St atus Rule Playbook, 20 26 Government Edition · Benchmark figures ma y be cited with attributio n: “ClearPoint Government Performance Benchmark, August 2026.” · clearpointstrategy .com

NEXT STEP

Bring one measure that

went red last quarter.

In a 20-minute walkthrough we’ll set the status rule, name the owner, and show you the council view and the resident view reading from the same record — on your data, not a demo. If it doesn’t beat your spreadsheet, tell us so. We’d rather know.

Book the 20-minute walkthrough

We’ll set one status rule live, on one of your own measures.

Prefer the raw numbers first? The full benchmark — dataset, definitions, quarterly re-extract s — is free, no form: clearpointstrategy.com/government-performance-benchmark