Diagnosis, Then a Recommendation

A diagnosis is a mechanism you could intervene on. It is not a restatement of the symptom with more decimal places.

“Fill rate is down nine points” is an observation. “The current rule allocates safety stock in proportion to volume, and the misses are concentrated in low-volume items whose stockouts strand a repair job” is a diagnosis, because it names something you could change.

The test is whether the sentence implies an intervention. If it does not, you have measured rather than diagnosed, and any recommendation attached to it is a guess in professional clothing.

Why teams refuse to recommend

The refusal is rarely stated. It shows up as a slide titled “Options” with three balanced columns, or a conclusion that “the data suggests” something without saying what to do.

Three causes, all honest.

The first is epistemic scruple: the analysis has assumptions, and a flat recommendation feels like overclaiming. The answer is to make the recommendation and attach the flip condition, which is more honest than a balanced menu, not less. A menu transfers the judgment to someone with less information than you.

The second is political exposure. A recommendation can be wrong in a way that a summary of options cannot. This is true and it is the job. The person who never recommends is never wrong and is also never the reason anything happened, and that trade is visible to everyone senior.

The third is a belief that recommending exceeds the analyst’s authority. It does not. Recommending is not deciding. The owner decides. Withholding a recommendation does not preserve their authority; it deprives them of the input they asked for.

Make an actual recommendation

The shape:

We recommend [action], for [scope], starting [when]. Expected consequence. Major tradeoff. Whether you are willing to run this, and for how long.

The last clause is the one that separates a recommendation from a suggestion. “We are willing to run this for a quarter and would revisit if bypass exceeds eight percent” is a commitment. “This appears promising” is not.

Recommending the status quo

Keeping the current rule is a legitimate recommendation and it is systematically under-produced, because a project that concludes with no change feels like a project that failed.

It is not. A study that establishes that the current rule is adequate, and names the condition under which that would stop being true, has saved the organization an implementation and given it a trigger to watch. That is a good outcome, and it is worth stating in exactly those terms so that nobody mistakes it for an absence of findings.