How a Purchase Stops Paying Off Without Anything Breaking

← Back · ACT I · Chapter 3: Value Drift

Every enterprise has machinery for things that break. Escalation paths, incident reviews, root cause analysis, an owner named on a page. Something stops working, an alarm sounds, and people assemble.

Almost no enterprise has machinery for things that diverge. A commitment that slowly stops resembling itself sets off nothing, because at no point does it break. It arrives at the end of the year as a smaller, softer, differently worded version of what was approved, and the record shows a series of reasonable adjustments rather than a failure.

That is Value Drift, and it is the reason most postmortems find nothing wrong.

Drift Is Produced, Not Suffered

Drift is often described as entropy, as though outcomes decay on their own. They do not. Drift is produced, reliably, by three forces, and each one is visible if you know its shape.

Interpretation. Intent is translated differently across teams. A decision to improve customer experience becomes a speed initiative in operations, a satisfaction initiative in service, and a retention initiative in sales. Every translation is defensible. None carries the original intent intact. The commitment has already moved before execution begins.

Substitution. Goals are replaced by easier proxies. Revenue generated per customer is hard to attribute, so the organization tracks customer touches instead, because touches can be counted this week. Goodhart named this in 1975: a statistical regularity tends to collapse once it is put under pressure for control purposes. A target does that to a measure, and people end up gaming the metric rather than the outcome underneath it. The goal changes without anyone declaring a change.

Relaxation. Standards weaken across review cycles. A twenty-five percent improvement target becomes approximately twenty percent at the first quarterly review, then meaningful improvement by midyear. Each adjustment is pragmatic on its own. Together they erode the commitment until the original is unrecognizable, and no meeting minutes record the moment it was abandoned.

An audit tabled in Australia in July 2026 shows the same shape in a real program. The Department of Foreign Affairs and Trade approved its plan for measuring the benefits of an A$394.3 million embassy security program eleven months after the government agreed the policy. In February 2021 one installation target was cut from 90 alarms to 50, and by August 2022 the program’s metrics had grown from 50 to 74, partly because new projects were added. The auditors found that “There was not a clear baseline established to measure program benefits against: metrics and targets changed throughout the life of the program.” In December 2023 the department closed the program “on the basis that it was an overall success, having achieved at least partial realisation of all intended benefits.” The audit office’s own conclusion was that the benefits had been “partly realised.” The report does not suggest anyone moved a target in bad faith, and it gives no reason for the alarm cut. What it shows is a program closed with no fixed line to measure it against.

Why Escalation Never Fires

Enterprise governance is calibrated for events, and failure is one of those. Drift is a rate, which is why it never shows up as an event at all.

An incident crosses a threshold and triggers a response. Drift never crosses one, because each individual adjustment sits below the level at which anyone is required to act. Interpretation happens in a kickoff. Nobody notices the substitution either, because it happens while someone is designing a dashboard, and the relaxation arrives in a quarterly review dressed as realism. There is no point at which a person with authority is handed a decision that looks like abandoning the commitment.

By the time the gap is large enough to be obvious, the original terms have been superseded by a chain of small, individually reasonable decisions, and nobody is in a position to say when it happened.

Drift Compounds

The compounding case for governed value is well understood: sustained execution, continuous measurement, and ongoing adjustment accumulate, and small improvements become large ones.

The same arithmetic runs in the other direction. Inconsistent execution, infrequent measurement, and absent adjustment accumulate too. Small deviations become permanent outcomes on exactly the same curve, and they compound quietly, which is why the endpoint is usually a surprise to the people who authorized the beginning.

The consequence reaches past any single initiative. Repeated drift teaches an organization that commitments do not hold. Targets get set lower on purpose. Business cases get written with room already built into them. Once a workforce stops expecting that promised outcomes will arrive, the discount is priced into everything proposed afterward.

It Survives Because It Belongs to Everyone

Drift persists for a structural reason rather than a cultural one. No single decision causes it, and no single role is accountable for correcting it.

The team that interpreted the intent was doing its job, and so was the team that chose the proxy metric. The executive who softened the target was exercising judgment against real constraints. Each actor behaved defensibly inside their own remit. Drift belongs to all of them and therefore to none of them, and distributed ownership is the same thing as no ownership when a verdict has to be issued.

What a Standard Does About It

Naming the three forces is the part an organization can act on without restructuring anything.

Drift becomes governable at the moment the original terms survive in a form that later decisions have to be measured against. A standard fixes the definition of the outcome, the evidence that will settle it, and the threshold below which an adjustment is no longer an adjustment. Interpretation is caught because the definition is written once and referenced rather than re-translated. A proxy cannot quietly take the seat, because the evidence that counts was named in advance. And relaxation turns into an explicit decision with an author, instead of a softer adjective in a slide.

None of that requires a new office or a new headcount. It requires that the terms be set before capital moves, which is the same discipline described in The Numbers Behind Big Purchases Are the Only Ones Nobody Audits, applied forward through execution rather than at the point of purchase.

Drift does not survive being named. It survives being distributed, and that is a decision an organization is making whether or not it knows it.

Governance Check: for your largest active initiative, what was the original target as written at approval, and who is authorized to change it?

Sources: Value Drift, the three forces of interpretation, substitution, and relaxation, and the drift-versus-failure distinction are developed in Chapter 3 of The Value Standard (Value Lattice, 2026). The substitution mechanism draws on Charles Goodhart’s 1975 observation that an observed statistical regularity tends to collapse once pressure is placed on it for control purposes. The compressed version usually quoted, that a measure which becomes a target ceases to be a good measure, is Marilyn Strathern’s 1997 wording rather than Goodhart’s own. The Australian example is from the Australian National Audit Office, Auditor-General Report No. 2 2026-27, DFAT’s Protection of Overseas Missions: Realisation of Benefits from the Security Enhancements Program (July 2026), paragraphs 8 and 1.19 to 1.21.

Keep reading: ACT I · Chapter 3: Value Drift