Practical guide
AI fluency for sales operations: reconcile pipeline aging
Reconcile a fictional pipeline export by removing version duplicates and calculating age from the correct stage-entry event.
Sales operations analysts can use AI to prepare pipeline reporting while checking record identity and dates. This fictional exercise reconciles duplicated exports and calculates stage age for three opportunities. It focuses on denominator and event discipline, not revenue prediction, quota evaluation or a judgment about real sellers.
Choose the record and date definitions
Pipeline age needs a unit of analysis and a defined starting event.
The fictional report counts one current row per opportunity and defines stage age as report date minus the latest supported entry into the current stage. The report date is September 1. Creation date and last general edit are not substitutes for stage entry under this definition.
Ask AI to restate the rule and list required fields. Keep missing entry dates visible. Do not use today from the computer clock when the exercise supplies a report date, and do not infer that any record edit restarted stage age.
Remove version duplicates without losing history
Two export rows can be versions of one opportunity rather than two opportunities.
The packet has A version 1 and A version 2 with the same stable identifier. Version 2 has the later export timestamp and a corrected amount. Count A once in the current snapshot, while retaining both versions in the reconciliation note. B and C each have one row.
Do not deduplicate by organization name alone because different opportunities may belong to the same organization. Ask AI to identify duplicates using the supplied stable key, then inspect conflicts between versions. A smaller total is not automatically correct if distinct opportunities were merged.
Calculate stage age from dated events
The arithmetic should be reproducible and use the same day convention.
A entered its current stage August 28, B entered August 30 and C entered September 1. Using elapsed whole calendar days to the September 1 report date, their ages are four, two and zero days. These dates and conventions are exercise inputs.
Show each subtraction and label the convention. Do not average the ages until missing or disputed dates are resolved. If the reporting system uses another boundary or timezone, that is a different definition requiring its own checked calculation.
Treat a stage correction as a data change
Correcting the current stage can change which entry event is applicable.
Now a source note shows B was moved to Evaluation in error and remains in Discovery, which it entered August 25. Update both its stage and relevant entry date; its corrected age becomes seven days. Do not keep the August 30 date merely to minimize disruption to the report.
Create a before-and-after row with source and reason. AI can recompute the distribution, but verify that it does not alter A or C. Preserve who approved the correction in real use; this synthetic packet provides only a fictional data owner.
Deliver the reconciled snapshot and exceptions
A useful report exposes the data decisions behind its totals.
Include three current opportunities, the duplicate-version disposition, ages four, seven and zero, and the correction history. List any records excluded for missing stable keys or unsupported dates. Avoid describing stage age as account health or likelihood to close.
Review record identity, date selection and response to the corrected event. Revenue operations defines valid lifecycle states; this analyst exercise builds a consistent snapshot from those states without turning descriptive data into a forecast.
Sources and scope
NIST: AI risk management. Skills England: workplace AI foundations.
These references provide background, not validation or endorsement of this exercise. The case details, calculations and suggested review questions are original instructional material. Use them to discuss observable work, not to infer customer outcomes, professional credentials or performance in every setting. Before adapting the exercise, confirm the relevant facts, approved tools, data permissions and decision owners. If you change the case, revisit the expected answers and checks as well. These examples describe practice tasks, not a promise that a particular product includes the fictional features.
Sources: [1] [2]
Sources
- 1.AI RMF Core · NIST
- 2.AI foundation skills for work benchmark · Skills England