Diagnosis is knowing-through.
Distinguishing one thing apart from another through careful examination.
A list of everything wrong is an inventory. A diagnosis says which thing is the source and which things are downstream of it.
A real diagnosis is falsifiable.
If I am right, this intervention changes that measurement. If it does not, the diagnosis was wrong and gets revised. That is the difference between a diagnosis and an opinion.
Before you can accurately perceive, guide, or correct, you must first apply that same knowing-through to your own blind spots.
Systems that refuse to do this work miss the solution, because growth holds up a mirror.
Asking one artificial intelligence to check another does not make the result more trustworthy.
The second model brings no new evidence to the question. It brings another fluent answer. Agreement between two fluent answers is not confirmation of either.
Both systems learned from overlapping data and share the same failure modes. Both can make the same hidden mistake in the same place, at the same moment, in the same confident voice.
Two matching answers feel like a second opinion. Without understanding the shared blind spots, the agreement becomes evidence of nothing, believed as proof.
A diagnosis needs what neither model can supply: evidence from outside both of them. The record of what your system actually did.
These questions get asked when a diagnostic appointment is made.
If I spent a day watching someone perform this job, what would I see that is not written in your procedures?
Every question below is that question, asked from a different direction. Bring your best answers, including "I do not know."
Business context
- BC-01What prompted you to contact us today?
- BC-02What outcome are you trying to improve?
- BC-03Why is this important now?
- BC-04What happens if nothing changes?
The problem
- PR-01What do you believe is happening?
- PR-02How long has this been occurring?
- PR-03Who is most affected?
- PR-04Where in the organization does it occur?
- PR-05How do you currently know it is a problem?
Current process
- CP-01How is this work currently performed?
- CP-02Who performs it?
- CP-03What systems are involved?
- CP-04Is the process documented?
- CP-05Does the documented process match what actually happens?
Business impact
- BI-01How does this affect revenue?
- BI-02How does this affect cost?
- BI-03How does this affect risk?
- BI-04Have you already attempted to solve it?
- BI-05What happened?
Success
- SC-01If this engagement were successful, what would be different six months from now?
- SC-02How would you measure success?
- SC-03Who would decide the project was successful?
Engagement readiness
- ER-01Who owns this process?
- ER-02Who can authorize changes?
- ER-03Who should participate in the diagnostic?
- ER-04Are there known constraints we should understand before arriving?
"I do not know" is a finding, not a failure. Every unanswerable question above marks a place where the record is missing, and the missing record is where the diagnosis begins.
The diagnostic exists to understand how the work actually happens, to identify where measurable value may exist, and to determine whether an engagement is justified. Recommendations come afterward, if the evidence earns them.
Careful examination starts with a call.
Bring the questions above and whoever holds the system. If a diagnostic is not the right fit, that gets said on the first call.
brucetisler@quantuminquiry.org