01 the business
Contract manufacturer
120 staff, sales team of seven
The sales manager spends Monday mornings exporting CRM data and comparing it with last week's spreadsheet. Directors get a slide deck that shows the total pipeline value, but not why deals moved. The weekly meeting turns into everyone reading out their deals.
02 before
The problem
The sales manager spends Sunday evening or Monday morning exporting CRM data into a spreadsheet to work out what changed. The numbers explain what the pipeline is worth, but not why deals moved or stalled. Owners and directors get a slide deck that is out of date by Wednesday.
- Half a day each week goes on exports, lookups and formatting.
- Deals that slip quietly are noticed weeks late.
- Directors and the sales team look at different versions of the numbers.
03 the build
What the AI does
- Step 1: Takes a snapshot of every open deal in your CRM at the same time each week and compares it with the week before.
- Step 2: Lists new deals, deals that moved stage, deals that went quiet and close dates that were pushed back.
- Step 3: Reads the latest activity notes on each changed deal and explains the reason for the change in one line.
- Step 4: Flags deals that look at risk, such as no contact in a long time or a close date already in the past.
- Step 5: Posts the summary to Slack or Teams and updates a Looker Studio or Power BI page with the same figures.
paused: waiting for a person to approve
How it was put together
- Took a snapshot of every open deal at the same time each week and stored it for comparison.
- Listed new, moved, stalled and slipped deals, with a one-line reason drawn from the latest notes.
- Flagged deals at risk using simple rules the manager chose, such as no contact in 21 days.
- Posted a summary to Slack and refreshed the same figures on a Looker Studio page.
04 human in the loop
Where a person stays in charge
The sales manager reads the draft first and can add comments or correct a deal note before it goes to the wider leadership channel.
- The manager reviews the draft before it reaches the leadership channel.
- Each reason links to the CRM note it came from.
- The report only reads the CRM. It never changes a deal.
05 rollout
From kickoff to live in 2–3 weeks
- Week 1Agree the snapshot time, risk rules and the summary layout.
- Week 2Run alongside the manual report and compare the two.
- Week 3Switch the weekly meeting to the new summary.
06 after
What changes
The weekly pipeline meeting starts from a shared, current picture. The manager stops building reports by hand, and the team talks about the few deals that need help rather than reading out the full list.
Assume the manager spends 4 hours each week on the report and the summary takes 20 minutes to review.
4 hours − 20 minutes = 3 hours 40 minutes saved a week, or about 190 hours a year.
What to measure
- Hours spent preparing the weekly report.
- Deals flagged at risk that were recovered.
- Forecast accuracy at quarter end.
- Length of the weekly pipeline meeting.
07 evidence
What the research says
A survey of 4,050 sales professionals found the average seller spends about 40% of their time actually selling, and 51% of sales leaders using AI say disconnected systems are slowing their AI work.
Salesforce, State of Sales report, 2026 edition (opens in a new tab) · 2026
Gartner research put the average cost of poor data quality at $12.9 million a year per organisation.
Gartner, Data quality: why it matters and how to achieve it (opens in a new tab) · 2020
In a study of 453 professionals doing writing tasks from their own jobs, access to ChatGPT cut the time taken by 40% and raised quality, as rated by evaluators, by 18%.
08 lessons
Mistakes to avoid
- Explaining changes without evidence. Every reason should point to a note.
- Too many risk rules. Start with two or three that the manager trusts.
- Letting reps game the notes. The summary reflects what is logged, so logging habits matter.
09 questions
Common questions
- What if our CRM data is messy?
- The summary shows the mess quickly, which helps. Many teams pair this with automatic CRM logging.
- Can it go to Microsoft Teams instead of Slack?
- Yes, and the dashboard can be Power BI instead of Looker Studio.