clean up a messy CRM export and build a forecast the sales leader will actually trust for the board meeting

• Export the active pipeline report from the CRM into a spreadsheet.
• Filter out deals with past close dates that reps forgot to push.
• Cross-reference notes and Slack messages to check if deals set to close this month are actually moving.
• Manually adjust close dates and probabilities based on gut feel and recent manager check-ins.
• Pivot the cleaned data by rep and stage to calculate the expected weighted forecast.
• Copy and paste the final numbers into a slide deck for the sales leader's review.

Arguing with sales managers about whose deals are real and whose are wishful thinking, while manually fixing date formats and blank fields row by row.

1. You

   1. Gather the pipeline evidence

   Export the active pipeline with deal owner, amount, stage, probability, close date, required fields, and notes. Add the relevant manager check-ins and Slack messages for deals expected to close this quarter; AI cannot retrieve or verify private context that you do not provide.

2. AI

   2. Standardize and flag the CRM data

   Give AI the export and supporting context, then have it normalize date and currency formats, identify blank required fields, detect inconsistent stage values, and flag close dates set to the quarter's final day. It should preserve the original values and label uncertainty rather than inventing missing information.

3. AI

   3. Create a deal-level evidence table

   Have AI match each in-scope deal with the supplied notes and messages, summarize evidence of movement or blockage, and assign a data-quality and confidence flag. Treat summaries as decision support: AI cannot reliably determine whether a rep's forecast is true without current human confirmation.

4. You

   4. Approve the forecast treatment

   Sales managers and RevOps decide which deals remain in the quarter, which move to a later period, and which are excluded from the forecast, using the evidence table and direct deal knowledge. They also approve the stage and probability policy that will determine the board forecast; these choices change the reported revenue and remain human-owned.

5. AI

   5. Calculate the approved forecast

   Have AI apply the approved dates, stages, probabilities, and exclusions to calculate weighted revenue by rep, stage, and close period. Ask it to produce commit, best-case, and excluded or at-risk views where the approved rules support them, while retaining an audit column showing the source and change for every adjusted deal.

6. You

   6. Sign off and prepare the board version

   The sales leader confirms that the totals, major deal movements, assumptions, and exceptions match the business situation, then signs off on the numbers or sends specific deals back for correction. Use the approved summary and tables in the board slide deck rather than copying unverified raw CRM totals.

• Find and merge duplicate contacts and companies, and flag stale deals before the monthly pipeline review
• Document the sales process so a new sales rep can follow it without constantly asking questions