The deliverable was never the analysis. It was the dependency.

Ask any capability leader what they actually bought the last time they commissioned an external diagnostic, and the honest answer is not the analysis.

They could describe the problem before the engagement started. Most of them had already named it in a leadership meeting, in plainer language than the final report used. What they could not produce, on the timeline the business runs on, was the artefact. The thing with a named methodology on it, a stated sample size, and enough structural credibility to survive a CFO asking where the number came from.

That artefact is the product. And the way it has been sold for thirty years has one structural feature that nobody puts in the proposal.

You cannot edit it.

The output arrives as a locked PDF. The methodology is described as proprietary, which is a polite way of saying you are not shown the working. The numbers cannot be changed, because the model that produced them never leaves the building it was built in.

So the first time a board member says "what happens to that figure if attrition runs three points higher," the answer is not a recalculation. The answer is a new conversation, a new scope, and a new fee.

Every follow-up question restarts the meter. That is not a bug in the model. That is the model.

I want to be precise about what I am and am not saying. The analysis in those engagements is frequently good. The people doing it are frequently excellent. The problem is not competence. The problem is that the commercial structure requires the client to remain unable to reproduce the work, and every design decision downstream of that requirement follows from it. Locked output is not laziness. It is architecture.

The second thing nobody prints.

Here is the part that bothers me more, and it is the reason I built what I built.

In any diagnostic of real size, some findings are solid and some are thin. The sample behind finding four is robust. The sample behind finding nine is eleven responses and a heavy assumption. Both arrive on the same slide, in the same font, with the same air of authority. The reader has no way to tell them apart.

Nobody grades their own confidence, because grading it looks like weakness in a room you are trying to win. So the client walks out treating a strong finding and a fragile one as equivalent, and allocates budget accordingly. I have watched organisations fund the wrong intervention for eighteen months on the strength of a number that should have carried a warning label.

A finding that does not tell you how much to trust it is not evidence. It is decoration.

What I did about it.

Judgment-Centered Leadership Design has always been a method before it was anything else. Diagnose before you prescribe. Classify root cause as skills, structure, measurement or context before you design a single intervention. Grade the evidence. Never present a figure you cannot walk someone through line by line.

That method is now software. It is called Treeng, and it went live this week.

Sixteen diagnostic engines. You bring your data, the engine runs a named methodology against it, and you get back an Excel workbook with live formulas and a deck you own outright. Kirkpatrick-Phillips for training ROI. Ishikawa and Pareto for root cause. Lean for process. ADDIE or SAM for instructional design. Not a proprietary black box with a brand name on it. Methods with literature behind them, applied properly, at speed.

And every single finding carries a grade:

• A. Sufficient sample, clean inputs, statistically meaningful. Act on it.
• B. The direction is reliable, the magnitude carries a wider band. Strong signal, not a settled figure.
• C. Needs more data, and the engine tells you precisely which data would move it up a grade.

That last part is the whole point. A consultant will not tell you which of their conclusions is thin, because the engagement is priced on confidence. A system has no such incentive. It can afford to say "this one is weak, and here is why."

The economics are a consequence, not the pitch.

Comparable external engagements in this category are commonly quoted somewhere in the fifty to two hundred thousand dollar range. That is a market observation about typical scope, not a figure from any engagement of mine, and what you would actually pay depends entirely on who you ask and for what.

Treeng starts free. A full analysis runs from nine dollars. Plans begin at forty-nine a month and do not auto-renew, because a subscription that renews itself is just dependency wearing a different hat.

But the price is the least interesting thing about it. The interesting thing is that when the analysis takes four minutes instead of six weeks, you stop rationing it. You stop reserving diagnostics for the moments important enough to justify a procurement cycle. You run one because you are curious, and curiosity is where most good organisational decisions actually start.

What this does not replace.

I should be clear, because the temptation with a launch is to claim it does everything.

A diagnostic tells you what is broken. It does not rebuild the architecture underneath it. Where the finding is structural, where judgment itself is the constraint rather than the measurement of it, you do not need a better report. You need somebody to redesign how decisions get made, and then to leave without the system falling over. That is still an engagement, and it always will be.

What Treeng removes is the six weeks and the six figures you currently spend finding out which of the two you need.

Run the free one first. Look at the grades before you look at the conclusions. If most of your findings come back C, that is not a failure of the tool. That is the most useful thing anyone has told you about your data in a while.

A question for the operators reading this: the last external diagnostic you commissioned, could you have rerun it yourself six months later when the assumptions changed? If not, what did you actually buy?