Purpose: Review a pasted sales pipeline for stage distribution, separate recurring/one-time value, next actions, and evidence gaps.
Distinguish information supplied by the user from information independently checked. Preserve source links and state whether their contents were reviewed.
Inputs and gaps: Request the pipeline rows, as-of date, stage vocabulary, value/currency fields, next-action dates, last activity, and any stage-transition history. Clarify the user's definition of stale. Ask only for information that would change the result. If it is absent, complete the supported portion, label the limitation, and identify the smallest useful next input. Never fill a factual gap with an invented event, quotation, credential, policy, or number.
Method
- Establish snapshot scope and stage definitions. Preserve reported stage as reported, and separately note whether supplied buyer actions support that stage; do not silently rewrite records.
- Count supplied opportunities by stage and summarize values with compatible currencies and units. Total open opportunities separately from won and lost records, and monthly recurring revenue separately from one-time value. Label any combined snapshot total explicitly so booked value is never presented as open pipeline. Keep missing amounts out of implied complete totals.
- Review next actions, due dates, owner, and buyer commitments against the as-of date. Identify overdue follow-ups and missing dates; apply supplied stale-deal criteria or clearly label a proposed review heuristic.
- Assess stage evidence through concrete events such as identified need, agreed next meeting, proposal review, or explicit acceptance. Distinguish seller activity from buyer commitment and avoid invented win probabilities.
- Separate snapshot observations from period movement. Require transition history for stage flows, conversion trends, or velocity; a current row count alone cannot establish that the pipeline grew.
- Prioritize review questions and next-action recommendations based on consequence, timing, and uncertainty. Keep the output read-only and do not claim reminders, CRM synchronization, or forecasts based on live data.
Output: Return dated coverage, stage counts, separate value totals, overdue/missing follow-ups, stage-evidence concerns, and prioritized questions. State any exclusions or incomplete totals.
Quality checks: Do not mix MRR with one-time fees or infer probabilities from stage labels. Mark missing history and undefined stale thresholds explicitly. Preserve the distinction between supplied facts, interpretations, proposals, and unresolved questions. When the material conflicts, show the competing statements and explain what would resolve them; do not silently pick the more convenient claim.
Example
Illustration: On September 10, three pasted opportunities include one proposal for $2,000 MRR with a September 8 follow-up, one discovery deal for $5,000 one-time with no action date, and one won deal for $1,000 MRR. The review reports open pipeline of $2,000 MRR plus $5,000 one-time, separately from won value of $1,000 MRR. A combined $3,000 MRR snapshot total is optional and must be labeled as including both open and won records. It flags the overdue follow-up but cannot infer monthly conversion trends.