AI prompts for sales managers

Steps

1. 1

   Export your pipeline.

   From your CRM, pull open deals with rep, stage, amount, close date, last activity date and the last note. A spreadsheet or CSV is fine.

2. 2

   Ask AI to find the risks.

   Use the pipeline review prompt below. It should flag deals with no recent activity, close dates that keep moving, and stages that do not match the notes.

3. 3

   Pull call notes or transcripts for each rep.

   A few recent calls per rep is enough to spot patterns.

4. 4

   Get one coaching point per rep.

   Not ten. One thing they can change this week, with an example from their own call.

5. 5

   Build the meeting.

   Use AI to turn the above into a short agenda, so the weekly meeting is about decisions, not reading numbers out loud.

Prompts you can copy

Pipeline review

You are a sales manager reviewing the pipeline before the weekly meeting.

Here is the open pipeline export with rep, deal, stage, amount, close date, last activity date and last note:
[paste]

Today is [date]. Flag deals that:
1. Have no activity in [number] days.
2. Have a close date in the past or pushed more than once, if the notes show it.
3. Sit in a stage the notes do not support, like "proposal sent" with no proposal in the notes.

For each flagged deal give: rep, deal, the problem in one line, and one question I should ask the rep. Use only what is in the data. If something is unclear, say so. Sort by amount, largest first.

Coach a rep from their calls

Here are notes or transcripts from [rep name]'s last [number] calls:
[paste]

Act as a sales coach. Tell me:
1. One thing they do well, with a short quote from the calls.
2. The one thing that would help most to fix, with a short quote showing it.
3. A drill they can practice for 10 minutes a day this week.
Keep it under 200 words. Be direct and kind. Do not make up quotes, only use words from what I pasted.

Weekly meeting agenda

Build a 30 minute weekly sales meeting agenda from this:
Pipeline risks: [paste output from the pipeline review]
Wins this week: [list]
Team numbers this week: [calls, conversations, meetings, deals, whatever you track]

Give me: 5 minutes of wins, 15 minutes on the top 3 deals at risk with one decision each, 10 minutes on one skill to practice as a team. List what each rep should bring. Plain words, under 150 words.

Forecast sanity check

Here is my rep's forecast for this month: [paste deals they say will close]
Here are the notes and last activity for those deals: [paste]

For each deal, rate it likely, maybe, or unlikely based only on the notes and activity, and give one line why. Then list what proof would move each maybe to likely. Do not guess at amounts or dates I did not give you.

Example

This is a made-up example.

Pasted pipeline row: "Rep: Carlos. Deal: Tri-County Electric, $42,000, stage Negotiation, close date moved from Aug 30 to Sep 30 to Oct 31. Last activity Sep 12. Note: waiting on owner sign off."

A good flag:

"Carlos, Tri-County Electric, $42,000. Close date moved twice and no activity in about 6 weeks while in Negotiation. Ask: When did you last talk to the owner directly, and what does he need to see to sign?"

Tips

• Clean data in, useful answers out. If last activity dates are wrong in your CRM, fix that first. See data health.
• Coach one thing at a time. Reps cannot fix ten things in a week.
• Share the AI output with the rep before the one on one, so the talk is a conversation, not a surprise.
• Run the pipeline prompt every week with the same format so you can compare.
• More prompts for this role are in sales manager prompts.

What not to do

• Do not use AI scores to rank or punish reps. Use them to start a talk.
• Do not paste data your company does not allow in outside tools. Ask which AI tools are approved, or use the one built into your work apps, like Copilot in Microsoft 365 if your company has it.
• Do not trust a flag without checking the deal. The CRM is often behind what the rep knows.

FAQ

Can AI build my sales forecast?

It can sort deals and flag weak ones from your notes, but it cannot know what a buyer is thinking. Use it as a second opinion and ask your reps the questions it raises.

How often should I run a pipeline review with AI?

Once a week before your sales meeting works for most teams. It takes a few minutes once your export and prompt are saved.

What if my reps don't write good notes?

Then the AI has little to work with. Ask reps to log three things after every call: what the buyer said, next step, and date. AI can help clean up rough notes, see call notes.

Which AI tool should a sales manager use?

Use the one your company approves. ChatGPT, Claude, Gemini and Copilot can all read spreadsheets and notes. If your CRM has AI built in, try that first since the data is already there.

Rather have it done for you? We build the system that handles your outreach, one time, and you own it. See how it works.

First posted October 28, 2026. Last updated October 28, 2026. Free to copy and use, CC BY 4.0.