How to evaluate a diagnostic before you commission one
Operators commission diagnostics for a specific reason. Performance has slipped, growth has stalled, an acquisition is being considered, a new investor is coming in, the founder is preparing for a transition. The decision is consequential, and the diagnostic is supposed to inform it.
Most do not. The reason is that the operator has no framework for evaluating the quality of a diagnostic before commissioning it, and the firms that produce diagnostics are not incentivized to provide one. The operator hires based on relationship, brand, or fee, and discovers the quality of the work only after the engagement is delivered. By then, the decision has been made or missed.
There are five questions an operator can ask before commissioning a diagnostic that will predict, with reasonable accuracy, whether the work will produce useful findings.
1. Where will the work be done.
The diagnostician's location during the engagement is the single strongest predictor of finding quality. A diagnostic conducted from a conference room, with data delivered by the operator, will reflect the operator's existing view of the business. A diagnostic conducted inside the operation, with the diagnostician embedded in the work, will surface the dependencies and decision rights that the operator's existing data does not capture.
If the firm cannot tell you specifically where their team will be, who they will spend time with, and what they will observe, the diagnostic will be desk work dressed up as fieldwork. The findings will be thin.
2. Who is doing the work.
The seniority of the team that actually shows up determines what they can see. A junior associate cannot diagnose a governance problem because they have not run businesses with governance problems. A senior operator can, because they have. The diagnostic that matters is conducted by people whose pattern recognition is calibrated against operations of similar size and complexity.
If the firm pitches with senior people but staffs with associates, the work will produce a competent assessment and miss the diagnosis. Ask who will be in the operation, by name, and what they have run.
3. How long will the work take.
The time-box is a feature, not a constraint. A diagnostic that takes 90 days will produce more pages and fewer decisions than a diagnostic that takes 14. Length is not a proxy for rigor. Often it is the opposite.
A diagnostic that does not have a defined endpoint will drift. The scope will expand, the deliverables will multiply, and the operator will pay for analysis that arrives after the decision needs to be made. Ask what the deliverable looks like and when it lands. If the answer is vague, the engagement will be vague.
4. What is the deliverable.
A diagnostic should produce specific artifacts, named in advance, that the operator can use to make decisions. An operational baseline. A constraint map. A sequenced plan. A decision memo. The operator should know what they are getting before they commit, and the deliverables should be structured as inputs to decisions rather than as comprehensive descriptions of the business.
If the firm describes the deliverable as a "report" or a "presentation" without further specificity, the work is not yet defined. The artifacts should be named, scoped, and committed to in writing before the engagement begins.
5. What happens after.
A diagnostic that produces a list of recommendations and ends there is the most common failure mode. The recommendations get filed. The operator moves on. Nothing changes.
A diagnostic that produces a sequenced plan with a defined first 90 days, ownership for each initiative, and measurable outcomes is structured to drive action. The operator finishes the engagement with a decision, not a document. Ask what the firm expects the operator to do in the 30 days after the diagnostic is delivered. If the answer is "review the findings," the diagnostic is not designed to produce action.
These five questions take 15 minutes to ask and answer. They will tell the operator more about the quality of the engagement than any pitch deck or case study. Most firms will not answer them well, because most firms are selling assessments and calling them diagnostics.
The ones that can answer them, with specificity, are the ones worth commissioning.
- DIAGNOSTIC
The Composition Diagnostic
Most middle market businesses have been assessed before. The assessments produced documents. The documents identified problems. The problems mostly persisted. This is not because the assessors were wrong. It is because an assessment is not a diagnostic, and the two have been conflated for so long that operators no longer expect the distinction to matter.
5 MIN READ - DIAGNOSTIC
Twelve things we have learned about diagnosing middle market businesses
Most performance problems are governance problems wearing operational costumes. The constraint is rarely where the dashboard points. Twelve field notes from current engagements.
1 MIN READ