ODAM
A dry earthen ditch running between rows of orchard trees.
One outcome at a time

Maturity profile

Assess seven practices for one outcome against the evidence you can show, then give the next improvement to an owner. Never added up into one number.

Start with one outcome →Details: grades, rules and full examples
Seven questions

Ask each one of a single outcome.

01
Aim
What must stay true, in what range, and who owns it?
02
Measure
How will we know it's holding, and what decision does the number drive?
03
Evidence needed
What do we have to see to know?
04
Coverage
Do we actually see it, and is what we see fresh?
05
Detection
When it breaks, do we notice, and has that been tested?
06
Bounded work
Who or what may act on it, within what limits, and who checks?
07
Loop
Do we learn, and are repairs verified rather than assumed?
Five grades, plus Unknown

Graded by what you can show, not by what you own.

  1. AbsentAbsentNothing exists.
  2. StatedStatedWritten down, with an owner.
  3. In placeIn placeIt exists and runs.
  4. CheckedCheckedA recorded review by someone other than the person who did it, against a stated criterion. It doesn't mean every test passed, or that the review was independent.
  5. HeldHeldStill true within its stated horizon, and reconfirmed after the last change.
  6. UnknownUnknownNobody can find out yet. A finding, never a zero.

Grades go down as well as up: when a horizon passes or something changes, the answer has to be shown again.

Everyone answers the same questions, but that doesn't make profiles comparable across organizations, or give them the same authority.

Details: when grades go down
One profile (illustrative)

"Residents can pay their water bill online."

A three-person county IT shop, one outcome. Example data, not a real account.

How the county shop answers each of the seven questions for this outcome, the grade it earns, and the evidence behind the grade.
PracticeGradeWhat the team can show
AimWhat must stay true, in what range, and who owns it?HeldThe aim is written down and owned by the utility director. It covers one budget year and was reconfirmed after the payment vendor changed.
MeasureHow will we know it's holding, and what decision does the number drive?StatedThe measure is named, but no one has collected the numbers yet.
Evidence neededWhat do we have to see to know?StatedThe team has listed what it needs: the vendor's settlement report and the import log.
CoverageDo we actually see it, and is what we see fresh?UnknownNo one knows whether failed imports are logged. That gap is the main finding.
DetectionWhen it breaks, do we notice, and has that been tested?AbsentFailures are only noticed when residents call.
Bounded workWho or what may act on it, within what limits, and who checks?In placeOnly the billing clerk can rerun an import, but no one reviews the reruns.
LoopDo we learn, and are repairs verified rather than assumed?AbsentProblems are discussed when they happen, but nothing is written down.

Grades, from least to most: Absent, Stated, In place, Checked, Held. Unknown means no one can find out yet; it is a finding, not a zero. What each grade means.

The next improvement

Find out whether failed imports are logged. The billing clerk does the work; the utility director owns the result.

A second example, a state agency with forty outcomes, is on the reference page.

Details: both examples, with full evidence
Read beside the profile, never folded into it
Days at Risk
How long known risk stays. Where to look: Bounded work and Loop.
Time to Good Decision
How fast a decision that holds arrives. Where to look: Detection and Coverage.
Ambient Breach
Whether repair keeps up with arrival. Where to look: Coverage and Loop.