A book in progress
The analysis is
not the product.
Analysis that cannot change which alternative is chosen is decoration, and accurate decoration is still decoration. This is the working reference for the other job: name the choice, earn the evidence, install the rule that runs when you are not in the room.
Free and complete. Nothing on the method pages is gated.
The method
Three parts, and a study is what you get when one is missing
Ordinary words, used precisely. They are never abbreviated into an acronym, because a method of this kind should be readable in a meeting without a legend.
01
Choice
The alternatives, including the status quo. Who can commit. By when. Once, or repeating.
When it is missingA missing choice looks like a tour: interesting cuts, no owner, no freeze date, a dashboard that can be admired.
02
Evidence
Only what could change the ranking of those alternatives. Not what would be interesting.
When it is missingA missing piece of evidence looks like a sales document: a pick with no killable reasons, or reasons that could not change the ranking if they were false.
03
Rule
What the operating process does now, who runs it, and when it is revisited.
When it is missingA missing rule looks like a good meeting: agreement in principle, and Tuesday unchanged.
The premise
The failure is systematic, which is what makes it fixable
A vice president asks for a dashboard by Friday. A competent team extracts, explores, models, charts, and presents. Forty-seven slides later someone asks the only question that mattered at the beginning: what, exactly, are we supposed to do? The meeting agrees the work was thorough. The replenishment rule does not change.
“Inventory and service” licenses a dashboard, a forecast, a segmentation, and a tour of every distribution center. A choice sentence licenses only work that could change the ranking, in time. Everything else becomes visibly optional, which is usually the first honest thing that has happened to the project.
Evidence assembled against no candidate ranking has no stopping rule, which is why exploratory phases expand without bound. Write down what you would pick and what would prove you wrong, then go looking. A candidate that dies in week two is a successful week two.
No named owner who can change a parameter. No cadence. No specification of what happens when an input is missing. No priced exception path. The recommendation was accepted and the operating process never heard about it.
Free, and complete
The method
Eighteen entries, one per chapter. Each gives the definition, the mechanics, a worked example from operations, and the failure mode that most often breaks it. The book carries the cases and the politics; these carry the method.
- 01The Analysis Is Not the ProductThe deliverable is a commitment among alternatives, not an explanation of what happened.
- 02Name the ChoiceA single sentence naming who will choose among which alternatives, by when, and to what end.
- 03Problem, Decision, OutcomeThe separation of what is wrong, what we are choosing among, and what we want to be true afterward.
- 04Invent the AlternativesThe step of enlarging the option set before ranking it, on the grounds that decision quality is usually more sensitive to the menu than to the mathematics that ranks it.
- 05Refuse the First BriefThe practice of returning an underspecified request rather than absorbing it into analysis.
- 06Work Backward from the ChoiceStarting from a provisional answer and finding the cheapest evidence that could overturn it.
What it argues
Analysis that cannot change which alternative is chosen is decoration, and accurate decoration is still decoration.
- 1A decision has three parts, and a study is what you get when one of them is missing.You specify the choice: the alternatives including the status quo, who can commit, by when, and whether this fires once or repeats. You assemble the evidence that could change the ranking of those alternatives. You specify the rule: what the operating process does now, who runs it, and when it is revisited.
- 2The work runs backward and the presentation runs forward, and technical teams routinely reverse both.Backward means starting from the choice and finding the cheapest evidence that could change your mind. Forward means opening with the recommendation. The default is chronological in both directions: extract, clean, model, caveat, and conclude if there is time.
- 3The test is whether the organization can make the decision again without you in the room.The work is not finished when the analysis is correct. It is finished when a named owner runs a specified rule on a stated cadence, with a priced exception path and a date to revisit.
Writing
Recent articles
Forty-Seven Slides and the Rule Did Not Change
The most common outcome of competent technical work is a meeting that agrees the work was thorough and an operating process that continues exactly as before. The cause is not effort, and more effort will not fix it.
Read
Your Model Is Not the Answer
Model quality is an intermediate artifact, and presenting it as the deliverable is the central structural error in applied data science. The remedy is not clearer communication. It is a different definition of finished.
Read
How to Refuse a Brief Without Becoming the Obstacle
The first request is almost never the decision. Absorbing it anyway is the most expensive habit in technical work, and the standard objection to pushing back has a specific, practical answer.
Read
What Would Change the Ranking
One note every other Tuesday.
One piece of the method, one worked example from real operations, and one thing to try before you open a notebook. No news, no roundups, no launch announcements.
Subscribe by emailSign-up form pending. The link opens a pre-filled message.
Work with me
When the method needs to land in your organization, not just your inbox
Individual coaching for technical people whose analysis is correct and whose recommendations are not getting run, and team engagements where the problem is the operating process rather than any one analyst.