Pack the right inputs

For create a seven-minute decision deck for a six-week customer-support pilot, assemble only what changes the answer: decision requested, audience concerns, verified evidence, time limit, slide count, brand constraints, and likely objections. Label each item by authority and date. A source-of-truth document outranks a memory-based note; a current error log outranks a description of last month's behavior. State conflicts instead of letting the model blend them.

Use a four-part packet: task, evidence, constraints, and output contract. Put untrusted content inside clear delimiters and say that it is data, not instruction. Include representative examples, especially one normal case and one boundary case. Omit irrelevant history; excess context can hide the one line that controls the result.

Create a six-slide outline for leaders deciding whether to assign two staff to a six-week support pilot. Use only my evidence. Each title must be a complete takeaway sentence; add 2–3 support bullets, a suggested visual, and speaker notes under 40 words. Mark missing evidence; never invent metrics or quotes.

Before sending, annotate the packet. Mark which values are verified, which are illustrative, and which are unknown. If a screenshot is involved, transcribe critical small text. If structured data is involved, include headers, units, software version, and null behavior. If creative material is involved, record ownership and permitted use. This is how context becomes operational rather than decorative.

A useful response would look like this: A title-only story that moves from current support delay to bounded pilot, measurement plan, risk controls, and a final staffing decision. That description is intentionally observable. “Looks good” is not acceptance. The operator must read titles alone for logical flow, trace every number to a source, inspect chart axes, rehearse to seven minutes, and test contrast, font size, and alt text. Keep the source material beside the draft so review means comparison, not memory.

The privacy boundary is specific: remove confidential strategy, customer names, internal forecasts, and unreleased product information before using an external tool. Create the smallest synthetic example that preserves the problem. Replace names and identifiers consistently so relationships remain testable. Redaction is not merely drawing a box: crop surrounding notifications, remove metadata where relevant, and check that hidden sheets, comments, or revision history are not included.

Poor packets lead to predictable failures: decorating before structuring; invented metrics; slide titles that name topics rather than conclusions; overcrowded notes; unlabeled generated imagery. Another common failure is silently changing the source packet mid-run. Save a version or hash of the inputs beside the output, especially when another person will reproduce the work.

Rate a packet from zero to two on six dimensions: relevance, authority, recency, completeness, privacy, and reproducibility. A score below two on authority or privacy blocks the run. A low completeness score does not invite invention; it creates a question for the owner.