Diagnose the System, Not the Project
Consultants get hired to fix delivery and then discover nobody agrees on how delivery currently works. The PMO describes stage gates; the teams describe sprints; the ticket history describes neither. This diagnostic captures how work actually flows, in the client’s own words, before anyone prescribes how it should — a baseline you can measure against, not a workshop flipchart that evaporates by Friday.
Eight Probes That Expose the System
- Stated vs lived methodology. Which framework does the organization claim — then what actually happens when a deadline slips? The gap between those two answers is the engagement.
- On-time record. A number question on the last three projects’ delivery against plan. Memory is honest in aggregate.
- Decision latency. An opinion scale rating how long approvals typically stall. Long-latency organizations don’t have a methodology problem first; they have an authority problem.
- Tool inventory. A multiple choice of everything in use. More than four tools without integration is a finding in itself.
- The hero test. One yes/no: does delivery currently depend on one person’s overtime? Hero dependency is the most common single point of failure in mid-size delivery organizations.
- Change control reality. How do scope changes enter the plan — a form, a hallway conversation, or not at all?
- Reporting cadence. Who sees status, how often, in what format — and does anyone trust it?
- Risk ownership. Name the top delivery risk and its owner. A blank here means risk lives in nobody’s job description.
Reading Between the Answers
The diagnostic earns its keep through patterns: methodology theater (rich vocabulary, no artifacts), approval pneumonia (everything waits on one committee), spreadsheet glue (six tools connected by one exhausted coordinator). Each pattern points at a different intervention — and the diagnostic names it before the first workshop, which is precisely what separates a diagnostic from a questionnaire.
One Instrument, Comparable Baselines
Use duplicate to spin up a copy per engagement and keep the probes identical — comparability is the product. After your intervention, recall information pulls the client’s original baseline into the re-assessment, so improvement gets measured rather than asserted. Across engagements, your anonymized diagnostics quietly become benchmark data no competitor can shortcut.
Sharing the Findings
Viewer seats let client sponsors read results without edit rights, and webhooks push responses straight into whatever analysis stack your practice runs.
Frequently Asked Questions
Self-reported data is unreliable — why collect it?
Because the distortions are the data. What people believe about their process drives their behavior, and the gaps between departments’ answers locate the friction precisely.
How long should the diagnostic take?
About twelve minutes. Every added minute past twenty costs you mid-level respondents — the people who actually know where work gets stuck.
Who should complete it?
A spread: one sponsor, two delivery leads, two practitioners. The organization permissions keep response-level detail within the consulting team.
When do we re-run it?
Ninety days after the intervention, same instrument, no edits. The sibling evaluation forms collection covers adjacent assessment formats.