Opportunity Solution Tree / Discovery with a visible thread

Connect the goal
to what you test next.

Opportunity Solution Tree: visual rendering of the saved fictional demonstration output
Fictional demonstration. A visual rendering of the actual saved Opportunity Solution Tree output from the source-guided Codex preparation run. Targets and proposals are not achieved business results. Select the image to open the full-resolution image.
Actual generated discovery artifact / synthetic research
One desired outcome60% → 75%First volunteer session within seven days
Fictional baseline / proposed target
O1 / fixture score 0.72A clear first stepNotes R1, R3
Solution S1

A one-screen first-visit checklist.

E1, E2 / proposed; not run
Solution S2

A short visual welcome card showing the next action.

Solution S3

A plain-text confirmation that includes one next action.

O2 / fixture score 0.595Booking certaintyNotes R2, R4
Solution S4

A concise confirmation summary with status and reply-by time.

E3 / proposed; not run
Solution S5

A consistent pending/confirmed status label in the current booking message.

Solution S6

A manually maintained confirmation queue with an owner and review time.

O3 / Meeting pointScore 0.400 · Deferred branchO4 / ReschedulingScore 0.240 · Deferred branch

One outcome → four opportunities → six solution ideas on two selected branches → three proposed tests. No experiment was executed.

Build a four-level map from one desired outcome to customer needs, multiple solution options and proposed experiments.

Outcome → Opportunity → Solution → Experiment

Fictional controlled demonstration. Quotes, ratings and targets are supplied fixtures. No customer research, implementation or experiment result is claimed.

For teams deciding what to try.

Useful for product, service and operations teams that already have customer feedback and want to avoid jumping straight to one feature.

Provide a single outcome and authorized research. Receive 3–7 customer-framed opportunities, prioritized branches, at least three options per selected branch and test designs.

A reasoning workflow.

The original PM Skills instruction file structures discovery; it does not bundle analytics, survey collection, a diagram editor or service integration.

A model/assistant is required and can have costs. Supplied-material work needs no extra service API key. Reading uploads needs host file tools.

What was verified

Codex produced one outcome, four traceable opportunities, two prioritized branches, six solution options and three proposed experiments. Python recomputed the fixture scores and passed ten structure/evidence checks.

Not a prediction or a substitute for research. Missing data can use qualitative assessment; an opportunity score is only as trustworthy as its inputs.

Original Pawel Huryn / PM Skills source, MIT. Framework attribution to Teresa Torres is retained in the original instructions. Original copyright and license are preserved; no endorsement implied.