AI for Customer Success Managers: How CSMs Can Use AI in Their Day-to-Day Work
AI for Customer Success Managers: How CSMs Can Use AI in Their Day-to-Day Work
AI for Customer Success Managers: How CSMs Can Use AI in Their Day-to-Day Work
Key Takeaways
For high-call-volume CSMs, the biggest AI win is often meeting prep. A CSM running 30 calls per week can spend up to 15 hours on pre-call research and post-call follow-up, though actual time varies by call type and prep depth. AI compresses that to under 3 hours, returning the equivalent of almost two full working days per week.
You don't need a new platform to start. The six prompts in this article work with ChatGPT, Claude, or any AI tool. If your CS platform has native AI, the context is already there and you skip the copy-paste step.
Start with post-call follow-up emails. It's the fastest win, you do it every day, and the time saving is immediately obvious. Add pre-call briefs in week 2, then the Friday review in week 3.
QBR preparation is where AI saves the most time per task, from 3-4 hours down to 45 minutes. AI compiles the data; you decide how to frame it for your specific executive.
AI doesn't make your relationships feel less human, it makes you more prepared for the moments that are human. You arrive at calls knowing the context, not scrambling to remember it.
Key Takeaways
For high-call-volume CSMs, the biggest AI win is often meeting prep. A CSM running 30 calls per week can spend up to 15 hours on pre-call research and post-call follow-up, though actual time varies by call type and prep depth. AI compresses that to under 3 hours, returning the equivalent of almost two full working days per week.
You don't need a new platform to start. The six prompts in this article work with ChatGPT, Claude, or any AI tool. If your CS platform has native AI, the context is already there and you skip the copy-paste step.
Start with post-call follow-up emails. It's the fastest win, you do it every day, and the time saving is immediately obvious. Add pre-call briefs in week 2, then the Friday review in week 3.
QBR preparation is where AI saves the most time per task, from 3-4 hours down to 45 minutes. AI compiles the data; you decide how to frame it for your specific executive.
AI doesn't make your relationships feel less human, it makes you more prepared for the moments that are human. You arrive at calls knowing the context, not scrambling to remember it.
What AI Can Do for You as a CSM Right Now
The biggest immediate wins for CSMs are: pre-call research (30 minutes → 3 minutes review), post-call follow-up emails (20 minutes → 3 minutes review-and-send), and QBR preparation (3-4 hours → 45 minutes). The biggest time saver is meeting prep, at 30 calls per week, compressing prep from 30 minutes to 3 minutes each returns up to 13 hours per week. Start with post-call follow-ups using the prompts in this article.
The CSM Time Tax: Where Your Hours Actually Go (And Where AI Gets Them Back)
Here's the math that convinced a CS leadership team at a 40-CSM organization to invest in AI meeting prep: their CSMs were averaging 6 calls per day, 30 calls per week. Each call required roughly 30 minutes of prep and follow-up. That's up to 900 minutes per CSM per week, roughly 15 hours for a 6-calls/day schedule, spent on tasks that don't require a human relationship. Actual time varies by call complexity and volume. They were spending nearly half their week on research and writing, not customers.
You probably recognize this. The manual prep cycle, open CRM, check last call notes, scan the email thread, check usage data, try to remember what was discussed three weeks ago, takes longer than the call itself. And the post-call cycle, write notes, draft a summary, type the follow-up email, create tasks, eats another 20-30 minutes per call.
AI doesn't eliminate these tasks. It compresses them from 'time you spend doing' to 'time you spend reviewing.' The research happens in seconds. The draft is ready. You review, tweak, send. The work is still yours. It just doesn't take 30 minutes.
The Six Tasks Where AI Saves the Most Time
Task | Time Before AI | Time With AI | How AI Handles It |
|---|---|---|---|
Pre-call account research | 20-30 min per call | 3 min review | Compiles health trend, last call summary, open issues, renewal date into a structured brief |
Post-call summary & notes | 10-15 min per call | Auto-generated | Transcribes call, extracts key points, decisions, and action items automatically |
Follow-up email to customer | 15 min per call | 3 min review-and-send | Drafts email from call summary in your tone; you review and personalize |
QBR/EBR preparation | 3-4 hours per review | 45 min | Compiles quarter's health data, call themes, objective completion, usage trend, you customize the narrative |
Weekly account health review | 1-2 hours scanning | 20 min with 3 queries | Natural language queries surface accounts that need attention instead of dashboard scanning |
Handover brief for coverage | 45 min per account | Auto-generated | Compiles account context from calls, emails, and CRM into a structured coverage brief |
Kelly Poydence, a CSM at Basis Technologies, describes the direct benefit:
"I'm able to access all of those data points in one system and really feel competent and prepared for each and every interaction with that customer." — Kelly Poydence, CSM |
“I'm able to access all of those data points in one system and really feel competent and prepared for each and every interaction with that customer.”
Kelly Poydence
Director, Customer Success
Basis Technologies
What AI Can Do for You as a CSM Right Now
The biggest immediate wins for CSMs are: pre-call research (30 minutes → 3 minutes review), post-call follow-up emails (20 minutes → 3 minutes review-and-send), and QBR preparation (3-4 hours → 45 minutes). The biggest time saver is meeting prep, at 30 calls per week, compressing prep from 30 minutes to 3 minutes each returns up to 13 hours per week. Start with post-call follow-ups using the prompts in this article.
The CSM Time Tax: Where Your Hours Actually Go (And Where AI Gets Them Back)
Here's the math that convinced a CS leadership team at a 40-CSM organization to invest in AI meeting prep: their CSMs were averaging 6 calls per day, 30 calls per week. Each call required roughly 30 minutes of prep and follow-up. That's up to 900 minutes per CSM per week, roughly 15 hours for a 6-calls/day schedule, spent on tasks that don't require a human relationship. Actual time varies by call complexity and volume. They were spending nearly half their week on research and writing, not customers.
You probably recognize this. The manual prep cycle, open CRM, check last call notes, scan the email thread, check usage data, try to remember what was discussed three weeks ago, takes longer than the call itself. And the post-call cycle, write notes, draft a summary, type the follow-up email, create tasks, eats another 20-30 minutes per call.
AI doesn't eliminate these tasks. It compresses them from 'time you spend doing' to 'time you spend reviewing.' The research happens in seconds. The draft is ready. You review, tweak, send. The work is still yours. It just doesn't take 30 minutes.
The Six Tasks Where AI Saves the Most Time
Task | Time Before AI | Time With AI | How AI Handles It |
|---|---|---|---|
Pre-call account research | 20-30 min per call | 3 min review | Compiles health trend, last call summary, open issues, renewal date into a structured brief |
Post-call summary & notes | 10-15 min per call | Auto-generated | Transcribes call, extracts key points, decisions, and action items automatically |
Follow-up email to customer | 15 min per call | 3 min review-and-send | Drafts email from call summary in your tone; you review and personalize |
QBR/EBR preparation | 3-4 hours per review | 45 min | Compiles quarter's health data, call themes, objective completion, usage trend, you customize the narrative |
Weekly account health review | 1-2 hours scanning | 20 min with 3 queries | Natural language queries surface accounts that need attention instead of dashboard scanning |
Handover brief for coverage | 45 min per account | Auto-generated | Compiles account context from calls, emails, and CRM into a structured coverage brief |
Kelly Poydence, a CSM at Basis Technologies, describes the direct benefit:
"I'm able to access all of those data points in one system and really feel competent and prepared for each and every interaction with that customer." — Kelly Poydence, CSM |
“I'm able to access all of those data points in one system and really feel competent and prepared for each and every interaction with that customer.”
Kelly Poydence
Director, Customer Success
Basis Technologies
What AI Can Do for You as a CSM Right Now
The biggest immediate wins for CSMs are: pre-call research (30 minutes → 3 minutes review), post-call follow-up emails (20 minutes → 3 minutes review-and-send), and QBR preparation (3-4 hours → 45 minutes). The biggest time saver is meeting prep, at 30 calls per week, compressing prep from 30 minutes to 3 minutes each returns up to 13 hours per week. Start with post-call follow-ups using the prompts in this article.
The CSM Time Tax: Where Your Hours Actually Go (And Where AI Gets Them Back)
Here's the math that convinced a CS leadership team at a 40-CSM organization to invest in AI meeting prep: their CSMs were averaging 6 calls per day, 30 calls per week. Each call required roughly 30 minutes of prep and follow-up. That's up to 900 minutes per CSM per week, roughly 15 hours for a 6-calls/day schedule, spent on tasks that don't require a human relationship. Actual time varies by call complexity and volume. They were spending nearly half their week on research and writing, not customers.
You probably recognize this. The manual prep cycle, open CRM, check last call notes, scan the email thread, check usage data, try to remember what was discussed three weeks ago, takes longer than the call itself. And the post-call cycle, write notes, draft a summary, type the follow-up email, create tasks, eats another 20-30 minutes per call.
AI doesn't eliminate these tasks. It compresses them from 'time you spend doing' to 'time you spend reviewing.' The research happens in seconds. The draft is ready. You review, tweak, send. The work is still yours. It just doesn't take 30 minutes.
The Six Tasks Where AI Saves the Most Time
Task | Time Before AI | Time With AI | How AI Handles It |
|---|---|---|---|
Pre-call account research | 20-30 min per call | 3 min review | Compiles health trend, last call summary, open issues, renewal date into a structured brief |
Post-call summary & notes | 10-15 min per call | Auto-generated | Transcribes call, extracts key points, decisions, and action items automatically |
Follow-up email to customer | 15 min per call | 3 min review-and-send | Drafts email from call summary in your tone; you review and personalize |
QBR/EBR preparation | 3-4 hours per review | 45 min | Compiles quarter's health data, call themes, objective completion, usage trend, you customize the narrative |
Weekly account health review | 1-2 hours scanning | 20 min with 3 queries | Natural language queries surface accounts that need attention instead of dashboard scanning |
Handover brief for coverage | 45 min per account | Auto-generated | Compiles account context from calls, emails, and CRM into a structured coverage brief |
Kelly Poydence, a CSM at Basis Technologies, describes the direct benefit:
"I'm able to access all of those data points in one system and really feel competent and prepared for each and every interaction with that customer." — Kelly Poydence, CSM |
“I'm able to access all of those data points in one system and really feel competent and prepared for each and every interaction with that customer.”
Kelly Poydence
Director, Customer Success
Basis Technologies
The AI-Powered CSM Day: A Practical Walkthrough
Here's what your day looks like when AI handles the research and drafting layer. Second person, present tense, because this is actually doable starting Monday.
Morning (8:30-9:00am): Account Prioritization Without the Dashboard Scan
You open your CS platform. Instead of scanning 80 accounts manually, or looking at a wall of health scores and deciding what matters, AI has already surfaced 5-7 accounts that need attention today based on health changes, overdue tasks, and upcoming calls.
If your platform has natural language AI, one query saves the rest of the morning review: 'Which of my accounts had health changes in the last 7 days?' You get the list instantly. You scan the upcoming calls and identify which need real prep versus routine check-ins. You set three intentions for the day, one per priority account. Total time: 20 minutes.
Without AI: you spend 60-90 minutes scanning dashboards, reading through account lists, and trying to remember which accounts you were worried about last week. The output is roughly the same. The time cost isn't.
Before Each Call (30 min → 3 min): The Pre-Call Brief
Fifteen minutes before a call, you use Prompt 1 (below), either in your CS platform if it has native AI, or in ChatGPT/Claude with the account context pasted in. In under 2 minutes, you have: key context about this account, what was discussed in the last call and what the agreed next steps were, open issues you should follow up on, and what to focus on in this specific call given the account's current situation.
You arrive at the call knowing the account. Not scrambling to remember what you discussed in November. Not hoping the customer doesn't ask about the task you haven't completed yet. Prepared.
Time check: Pre-call prep: 30 minutes → 3 minutes. At 6 calls per day, that's 2.5 hours returned to you daily.
After Each Call (30 min → 5 min): AI-Powered Follow-Up
The traditional post-call cycle: write up notes while the call is still fresh, draft a summary, write a follow-up email, create tasks for the action items mentioned. Minimum 20 minutes, usually 30.
With AI: the call is transcribed. AI generates a summary (key discussion points, decisions, action items with owners). AI drafts the follow-up email. You review, add any personal touches, and send, often within 15 minutes of the call ending. The customer gets their follow-up the same day instead of the next morning. The action items are in your task system before you've forgotten them.
For high-call-volume CSMs, the compound math: 25 minutes saved per call × 30 calls per week = up to 12.5 hours per week returned to actual customer work. This, combined with prep savings, is where the 15 hours/week figure comes from.
End of Week, Friday (2 hrs → 30 min): The AI-Assisted Review
Three queries replace the two-hour Friday dashboard scan:
1. 'Which of my accounts dropped health score this week?', gives you the at-risk accounts to plan for next week.
2. 'Which accounts have renewals in the next 90 days with health below 75?', gives you where to focus renewal attention.
3. 'Which of my accounts have had no engagement activity in the last 3 weeks?', surfaces the silent accounts that need a check-in before they become problems.
With these three answers, you have a prioritized plan for next week in 20-30 minutes. You spend the remaining time on actual planning, deciding what you'll say in each outreach, not finding who to reach out to.
Jolt's CS team describes the shift this enables:
"I would say we were 80/20 reactive, so we were reactive a majority of our day — and now I can confidently state we're 70% proactive."
— Adam Cooney, Jolt
QBR and EBR Preparation: From 4 Hours to 45 Minutes
QBR prep is the most painful regular task in a CSM's calendar, and the one with the most to gain from AI. The traditional approach: gather usage data from the analytics tool, pull support ticket history, review last quarter's call notes, write up value delivered, prepare the business review slides, draft the executive summary. At 3-4 hours per QBR, a CSM doing 10 QBRs per quarter spends 30-40 hours/quarter just on preparation.
AI doesn't write the QBR for you, executives will notice if it's generic and it will undermine the relationship you've built. What AI does is handle all the research and first-draft work, so you focus on customization and strategy. Pam Dickson-Fishman at Basis Technologies describes the time shift when AI handles review preparation:
"Planhat saves us more than 30 hours per week automating and streamlining partner review preparations, lifecycle handoffs and time tracking administration, hours that can now be spent on driving value for our customers.", Pam Dickson-Fishman, Basis Technologies
Five Data Points AI Compiles in Under 5 Minutes
1. Health score trend over the quarter with inflection points, when did health change and what drove it.
2. Key themes from the quarter's calls, AI reads all transcript summaries and surfaces the recurring topics.
3. Objective completion status, which success plan milestones were achieved, which weren't.
4. Usage trend, is the customer using more or less? Which features?
5. Support history summary, major issues raised, how quickly they were resolved.
Each of these feeds the QBR narrative. Your job shifts from 'spend an hour finding this data' to 'decide what to emphasize and how to frame it for this specific executive at this specific moment in the relationship.'
The AI-Powered CSM Day: A Practical Walkthrough
Here's what your day looks like when AI handles the research and drafting layer. Second person, present tense, because this is actually doable starting Monday.
Morning (8:30-9:00am): Account Prioritization Without the Dashboard Scan
You open your CS platform. Instead of scanning 80 accounts manually, or looking at a wall of health scores and deciding what matters, AI has already surfaced 5-7 accounts that need attention today based on health changes, overdue tasks, and upcoming calls.
If your platform has natural language AI, one query saves the rest of the morning review: 'Which of my accounts had health changes in the last 7 days?' You get the list instantly. You scan the upcoming calls and identify which need real prep versus routine check-ins. You set three intentions for the day, one per priority account. Total time: 20 minutes.
Without AI: you spend 60-90 minutes scanning dashboards, reading through account lists, and trying to remember which accounts you were worried about last week. The output is roughly the same. The time cost isn't.
Before Each Call (30 min → 3 min): The Pre-Call Brief
Fifteen minutes before a call, you use Prompt 1 (below), either in your CS platform if it has native AI, or in ChatGPT/Claude with the account context pasted in. In under 2 minutes, you have: key context about this account, what was discussed in the last call and what the agreed next steps were, open issues you should follow up on, and what to focus on in this specific call given the account's current situation.
You arrive at the call knowing the account. Not scrambling to remember what you discussed in November. Not hoping the customer doesn't ask about the task you haven't completed yet. Prepared.
Time check: Pre-call prep: 30 minutes → 3 minutes. At 6 calls per day, that's 2.5 hours returned to you daily.
After Each Call (30 min → 5 min): AI-Powered Follow-Up
The traditional post-call cycle: write up notes while the call is still fresh, draft a summary, write a follow-up email, create tasks for the action items mentioned. Minimum 20 minutes, usually 30.
With AI: the call is transcribed. AI generates a summary (key discussion points, decisions, action items with owners). AI drafts the follow-up email. You review, add any personal touches, and send, often within 15 minutes of the call ending. The customer gets their follow-up the same day instead of the next morning. The action items are in your task system before you've forgotten them.
For high-call-volume CSMs, the compound math: 25 minutes saved per call × 30 calls per week = up to 12.5 hours per week returned to actual customer work. This, combined with prep savings, is where the 15 hours/week figure comes from.
End of Week, Friday (2 hrs → 30 min): The AI-Assisted Review
Three queries replace the two-hour Friday dashboard scan:
1. 'Which of my accounts dropped health score this week?', gives you the at-risk accounts to plan for next week.
2. 'Which accounts have renewals in the next 90 days with health below 75?', gives you where to focus renewal attention.
3. 'Which of my accounts have had no engagement activity in the last 3 weeks?', surfaces the silent accounts that need a check-in before they become problems.
With these three answers, you have a prioritized plan for next week in 20-30 minutes. You spend the remaining time on actual planning, deciding what you'll say in each outreach, not finding who to reach out to.
Jolt's CS team describes the shift this enables:
"I would say we were 80/20 reactive, so we were reactive a majority of our day — and now I can confidently state we're 70% proactive."
— Adam Cooney, Jolt
QBR and EBR Preparation: From 4 Hours to 45 Minutes
QBR prep is the most painful regular task in a CSM's calendar, and the one with the most to gain from AI. The traditional approach: gather usage data from the analytics tool, pull support ticket history, review last quarter's call notes, write up value delivered, prepare the business review slides, draft the executive summary. At 3-4 hours per QBR, a CSM doing 10 QBRs per quarter spends 30-40 hours/quarter just on preparation.
AI doesn't write the QBR for you, executives will notice if it's generic and it will undermine the relationship you've built. What AI does is handle all the research and first-draft work, so you focus on customization and strategy. Pam Dickson-Fishman at Basis Technologies describes the time shift when AI handles review preparation:
"Planhat saves us more than 30 hours per week automating and streamlining partner review preparations, lifecycle handoffs and time tracking administration, hours that can now be spent on driving value for our customers.", Pam Dickson-Fishman, Basis Technologies
Five Data Points AI Compiles in Under 5 Minutes
1. Health score trend over the quarter with inflection points, when did health change and what drove it.
2. Key themes from the quarter's calls, AI reads all transcript summaries and surfaces the recurring topics.
3. Objective completion status, which success plan milestones were achieved, which weren't.
4. Usage trend, is the customer using more or less? Which features?
5. Support history summary, major issues raised, how quickly they were resolved.
Each of these feeds the QBR narrative. Your job shifts from 'spend an hour finding this data' to 'decide what to emphasize and how to frame it for this specific executive at this specific moment in the relationship.'
Copy-Paste AI Prompts for Customer Success Managers
Six prompts you can use today, with any AI tool or your CS platform's native AI.
⚠️ Privacy note: Before pasting account data into any public AI tool, check your company's data policy. Avoid including customer PII, contract terms, or confidential pricing in prompts sent to external AI services. If your CS platform has native AI, use it, the data stays within your system. |
If your CS platform has native AI, you often don't need to paste in the context because it's already there. Either way, these work.
📋 PROMPT 1, Pre-Call Account Brief I have a call with [Company Name] in 30 minutes. Here is the account context:- Health score: [X], trend: [up/flat/down over last 30 days]- Last call date: [date]. Key points discussed: [paste 3-5 bullet points]- Open tasks/issues: [list]- Renewal date: [date]- Key contact: [name, role]Generate a pre-call brief covering: (1) Key context to remember about this account, (2) What was agreed last call and whether it was completed, (3) Open issues to follow up on, (4) What to focus on in this call given the account's current status. Keep it under 200 words. |
📋 PROMPT 2, Post-Call Summary & Follow-Up Email Here is a summary/transcript from my customer call with [Company Name]:[Paste call notes or transcript]Generate: (1) A 5-bullet summary of what was discussed, (2) Key decisions made, (3) Action items with owners and deadlines, (4) A follow-up email to send to the customer covering key points and confirming next steps.Email tone: professional but warm. Max 150 words. Start with something specific from the call, not a generic 'thanks for your time.' |
📋 PROMPT 3, QBR Executive Summary I need to prepare a QBR for [Company Name]. Account context:- Health score trend this quarter: [X → Y]- Key wins: [list 2-3 specific achievements with numbers if available]- Challenges addressed: [list]- Objectives completed: [X of Y], incomplete ones: [list]- Usage: [up/flat/down, which features most used]- Renewal date: [date]- Attendees: [executive names and roles]Generate a QBR executive summary that: (1) Opens with business impact, not product metrics, (2) Covers 2-3 wins with specific outcomes, (3) Addresses 1-2 challenges and the response plan, (4) Proposes clear focus for next quarter with measurable goals. Tone: confident, executive-ready, under 400 words. |
📋 PROMPT 4, At-Risk Account Re-Engagement Email I need to reach out to [Contact Name] at [Company Name]. The account has been declining:- Health context: [health score, main signal, e.g., usage dropped 35%, no response to last email 3 weeks ago, last call was 6 weeks ago]- Last interaction: [date and outcome]- Their priorities (as I understand them): [list 1-2 things they care about]Write a re-engagement email that: (1) Doesn't explicitly mention that I noticed low engagement (which sounds surveillant), (2) Offers something specific and relevant to their priorities, (3) Proposes a 20-minute conversation with a clear purpose, (4) Feels personal to this account, not templated. Max 100 words. |
📋 PROMPT 5, Expansion Conversation Starter My customer [Company Name] has just [achieved a milestone or hit a usage threshold, e.g., reached 85% seat capacity / completed onboarding 30% faster than average / used the platform for their first cross-team report].Write an email that: (1) Opens by acknowledging this specific behavior or achievement, (2) Asks a natural question about what comes next for their team (not pushy), (3) Briefly mentions how other customers at this stage typically grow their usage and what they get from it, (4) Suggests a 15-minute conversation to explore. Tone: consultative, curious. Max 120 words. |
📋 PROMPT 6, Coverage Brief / Handover I need to brief a colleague who will cover my accounts. For [Company Name]:- Recent context: [last call summary, key relationship notes]- Current status: [health score, main risk or opportunity, where we are in the relationship]- Upcoming meetings: [dates, purpose, what needs to happen]- Open tasks: [list with priority and deadline]- Key stakeholders: [names, roles, relationship notes]- Renewal date: [date]Generate a coverage brief my colleague can read in 5 minutes and feel prepared. Include: 3-sentence account context, current status and main priority, meeting notes, task list, and stakeholder guide. |
Copy-Paste AI Prompts for Customer Success Managers
Six prompts you can use today, with any AI tool or your CS platform's native AI.
⚠️ Privacy note: Before pasting account data into any public AI tool, check your company's data policy. Avoid including customer PII, contract terms, or confidential pricing in prompts sent to external AI services. If your CS platform has native AI, use it, the data stays within your system. |
If your CS platform has native AI, you often don't need to paste in the context because it's already there. Either way, these work.
📋 PROMPT 1, Pre-Call Account Brief I have a call with [Company Name] in 30 minutes. Here is the account context:- Health score: [X], trend: [up/flat/down over last 30 days]- Last call date: [date]. Key points discussed: [paste 3-5 bullet points]- Open tasks/issues: [list]- Renewal date: [date]- Key contact: [name, role]Generate a pre-call brief covering: (1) Key context to remember about this account, (2) What was agreed last call and whether it was completed, (3) Open issues to follow up on, (4) What to focus on in this call given the account's current status. Keep it under 200 words. |
📋 PROMPT 2, Post-Call Summary & Follow-Up Email Here is a summary/transcript from my customer call with [Company Name]:[Paste call notes or transcript]Generate: (1) A 5-bullet summary of what was discussed, (2) Key decisions made, (3) Action items with owners and deadlines, (4) A follow-up email to send to the customer covering key points and confirming next steps.Email tone: professional but warm. Max 150 words. Start with something specific from the call, not a generic 'thanks for your time.' |
📋 PROMPT 3, QBR Executive Summary I need to prepare a QBR for [Company Name]. Account context:- Health score trend this quarter: [X → Y]- Key wins: [list 2-3 specific achievements with numbers if available]- Challenges addressed: [list]- Objectives completed: [X of Y], incomplete ones: [list]- Usage: [up/flat/down, which features most used]- Renewal date: [date]- Attendees: [executive names and roles]Generate a QBR executive summary that: (1) Opens with business impact, not product metrics, (2) Covers 2-3 wins with specific outcomes, (3) Addresses 1-2 challenges and the response plan, (4) Proposes clear focus for next quarter with measurable goals. Tone: confident, executive-ready, under 400 words. |
📋 PROMPT 4, At-Risk Account Re-Engagement Email I need to reach out to [Contact Name] at [Company Name]. The account has been declining:- Health context: [health score, main signal, e.g., usage dropped 35%, no response to last email 3 weeks ago, last call was 6 weeks ago]- Last interaction: [date and outcome]- Their priorities (as I understand them): [list 1-2 things they care about]Write a re-engagement email that: (1) Doesn't explicitly mention that I noticed low engagement (which sounds surveillant), (2) Offers something specific and relevant to their priorities, (3) Proposes a 20-minute conversation with a clear purpose, (4) Feels personal to this account, not templated. Max 100 words. |
📋 PROMPT 5, Expansion Conversation Starter My customer [Company Name] has just [achieved a milestone or hit a usage threshold, e.g., reached 85% seat capacity / completed onboarding 30% faster than average / used the platform for their first cross-team report].Write an email that: (1) Opens by acknowledging this specific behavior or achievement, (2) Asks a natural question about what comes next for their team (not pushy), (3) Briefly mentions how other customers at this stage typically grow their usage and what they get from it, (4) Suggests a 15-minute conversation to explore. Tone: consultative, curious. Max 120 words. |
📋 PROMPT 6, Coverage Brief / Handover I need to brief a colleague who will cover my accounts. For [Company Name]:- Recent context: [last call summary, key relationship notes]- Current status: [health score, main risk or opportunity, where we are in the relationship]- Upcoming meetings: [dates, purpose, what needs to happen]- Open tasks: [list with priority and deadline]- Key stakeholders: [names, roles, relationship notes]- Renewal date: [date]Generate a coverage brief my colleague can read in 5 minutes and feel prepared. Include: 3-sentence account context, current status and main priority, meeting notes, task list, and stakeholder guide. |
How to Start Without Overhauling Your Workflow
The most common reason CSMs try AI once and stop: they try to change everything at once, hit friction, and revert. The correct approach is one task, one week, one small win, then expand.
Week 1, Post-Call Follow-Up Emails
After every call this week, use Prompt 2 with your AI tool of choice. You're not changing how you run calls. You're changing how you write the follow-up.
Time saved: approximately 15 minutes per call × 6 calls per day = 90 minutes per day. By Friday, you've saved over 7 hours without changing anything else about your workflow. The output quality will be better, more consistent, faster, and the customer gets their follow-up the same day instead of next morning. Both are wins.
Don't try to perfect the prompt in week 1. Use it as written, review the output, tweak what doesn't sound like you. Speed over perfection.
Week 2, Pre-Call Briefs for Your Three Most Important Calls
Don't start with all calls, pick your three most important calls of the week and use Prompt 1 before each. Notice the difference: you arrive knowing the context instead of flipping through notes in the two minutes before the call starts.
At the end of the week, compare: how much time did you spend on manual prep this week versus last? Most CSMs find the brief actually makes them more present in the call, not less human. You stop mentally tracking 'what did we discuss last time' and start actually listening.
Week 3, The Friday Review Routine
Replace your Friday manual account review with the three queries from Section 2 (End of Week). The first time, run both approaches, do your usual manual review AND the AI-assisted version. Compare what each surfaced.
Almost every CSM who does this comparison finds that AI caught something they missed, an account that went quiet, a health change they hadn't noticed, a renewal coming up faster than expected. After that comparison, most never go back to the manual review. Friday afternoon shifts from two hours of dashboard scanning to 30 minutes of actual planning.
When Your CS Platform Has Native AI: What Changes
The prompts in Section 4 work with any AI tool, ChatGPT, Claude, or any other LLM. The one thing that changes when your CS platform has native AI is that you skip the copy-paste step entirely. Instead of pasting account health, last call notes, and renewal dates into a prompt, you just ask.
In Planhat, for example, Conversational AI already knows your account's health trend, the summaries from the last three calls, your open tasks, and your upcoming renewal dates. 'Give me a pre-call brief for [Account]' takes 10 seconds and requires zero data pasting. Writing Assistant drafts follow-up emails with full account context already loaded. Email & Call Intelligence generates the post-call summary automatically from the recording. The prompts become one-liners instead of context-heavy pastes. If you're on a platform with this capability, start there. If not, the prompts in this article work exactly as written.
How to Start Without Overhauling Your Workflow
The most common reason CSMs try AI once and stop: they try to change everything at once, hit friction, and revert. The correct approach is one task, one week, one small win, then expand.
Week 1, Post-Call Follow-Up Emails
After every call this week, use Prompt 2 with your AI tool of choice. You're not changing how you run calls. You're changing how you write the follow-up.
Time saved: approximately 15 minutes per call × 6 calls per day = 90 minutes per day. By Friday, you've saved over 7 hours without changing anything else about your workflow. The output quality will be better, more consistent, faster, and the customer gets their follow-up the same day instead of next morning. Both are wins.
Don't try to perfect the prompt in week 1. Use it as written, review the output, tweak what doesn't sound like you. Speed over perfection.
Week 2, Pre-Call Briefs for Your Three Most Important Calls
Don't start with all calls, pick your three most important calls of the week and use Prompt 1 before each. Notice the difference: you arrive knowing the context instead of flipping through notes in the two minutes before the call starts.
At the end of the week, compare: how much time did you spend on manual prep this week versus last? Most CSMs find the brief actually makes them more present in the call, not less human. You stop mentally tracking 'what did we discuss last time' and start actually listening.
Week 3, The Friday Review Routine
Replace your Friday manual account review with the three queries from Section 2 (End of Week). The first time, run both approaches, do your usual manual review AND the AI-assisted version. Compare what each surfaced.
Almost every CSM who does this comparison finds that AI caught something they missed, an account that went quiet, a health change they hadn't noticed, a renewal coming up faster than expected. After that comparison, most never go back to the manual review. Friday afternoon shifts from two hours of dashboard scanning to 30 minutes of actual planning.
When Your CS Platform Has Native AI: What Changes
The prompts in Section 4 work with any AI tool, ChatGPT, Claude, or any other LLM. The one thing that changes when your CS platform has native AI is that you skip the copy-paste step entirely. Instead of pasting account health, last call notes, and renewal dates into a prompt, you just ask.
In Planhat, for example, Conversational AI already knows your account's health trend, the summaries from the last three calls, your open tasks, and your upcoming renewal dates. 'Give me a pre-call brief for [Account]' takes 10 seconds and requires zero data pasting. Writing Assistant drafts follow-up emails with full account context already loaded. Email & Call Intelligence generates the post-call summary automatically from the recording. The prompts become one-liners instead of context-heavy pastes. If you're on a platform with this capability, start there. If not, the prompts in this article work exactly as written.
Frequently Asked Questions
Which AI tools should I use as a CSM?
Start with what you already have. If your CS platform has native AI, start there, no context pasting required. If not, any general AI assistant (ChatGPT, Claude, or similar) works for every prompt in this article. Pick one task, post-call follow-up is the fastest win, and use it for a week before adding anything else. One workflow change at a time.
How do I use AI to prepare for customer calls?
Use Prompt 1 (pre-call brief) with your account context pasted in. In under 2 minutes you have: key context, last call highlights, open issues, and what to focus on today. If your CS platform has native AI, you often just ask directly, the platform already has the data. The output is a structured 150-200 word brief you read once before the call. That's it.
How do I speed up QBR preparation with AI?
Four steps: query your platform for the five data points (health trend, call themes, objective completion, usage trend, support history); paste into Prompt 3 with the context; review and customize, the output is a first draft, not a final, and the editing is where your relationship knowledge adds the most value; use AI for the QBR follow-up email as well. Total: 45 minutes instead of 4 hours.
What are the best AI prompts for customer success managers?
The six prompts in Section 4 cover the highest-frequency CSM tasks: pre-call brief, post-call summary and follow-up, QBR executive summary, at-risk account re-engagement, expansion conversation starter, and coverage handover brief. Start with Prompt 2 (post-call follow-up), it's the one you'll use most often and the easiest to build a habit around. Save all six somewhere accessible.
Will using AI make my customer relationships feel less personal?
No, and for most CSMs, it's the opposite. When AI handles the research and drafting, you have more presence in the actual conversation. You arrive at calls knowing the context instead of scrambling to recall it. You follow up within the hour instead of the next morning. You spend the 30 minutes you saved on prep actually thinking about what the customer needs, which is what makes a relationship feel personal in the first place. The parts that make customer success feel human, empathy, judgment, genuine care, are exactly what AI doesn't do. Those remain entirely yours.
How much time can I realistically save with AI as a CSM?
Conservative estimates based on actual CS team usage: post-call follow-up saves 12-15 minutes per call × 30 calls per week = 6-7.5 hours per week. Pre-call prep saves 20 minutes per call × 20 key calls per week = 6-7 hours per week. QBR prep saves 2-3 hours × 10 QBRs per quarter = roughly 2.5 hours per week. Combined: up to 14-17 hours per week if you apply AI across all three and have a high call volume. Even at half (if you only use it for some tasks), 7-8 hours is a meaningful change to how your week feels.
Frequently Asked Questions
Which AI tools should I use as a CSM?
Start with what you already have. If your CS platform has native AI, start there, no context pasting required. If not, any general AI assistant (ChatGPT, Claude, or similar) works for every prompt in this article. Pick one task, post-call follow-up is the fastest win, and use it for a week before adding anything else. One workflow change at a time.
How do I use AI to prepare for customer calls?
Use Prompt 1 (pre-call brief) with your account context pasted in. In under 2 minutes you have: key context, last call highlights, open issues, and what to focus on today. If your CS platform has native AI, you often just ask directly, the platform already has the data. The output is a structured 150-200 word brief you read once before the call. That's it.
How do I speed up QBR preparation with AI?
Four steps: query your platform for the five data points (health trend, call themes, objective completion, usage trend, support history); paste into Prompt 3 with the context; review and customize, the output is a first draft, not a final, and the editing is where your relationship knowledge adds the most value; use AI for the QBR follow-up email as well. Total: 45 minutes instead of 4 hours.
What are the best AI prompts for customer success managers?
The six prompts in Section 4 cover the highest-frequency CSM tasks: pre-call brief, post-call summary and follow-up, QBR executive summary, at-risk account re-engagement, expansion conversation starter, and coverage handover brief. Start with Prompt 2 (post-call follow-up), it's the one you'll use most often and the easiest to build a habit around. Save all six somewhere accessible.
Will using AI make my customer relationships feel less personal?
No, and for most CSMs, it's the opposite. When AI handles the research and drafting, you have more presence in the actual conversation. You arrive at calls knowing the context instead of scrambling to recall it. You follow up within the hour instead of the next morning. You spend the 30 minutes you saved on prep actually thinking about what the customer needs, which is what makes a relationship feel personal in the first place. The parts that make customer success feel human, empathy, judgment, genuine care, are exactly what AI doesn't do. Those remain entirely yours.
How much time can I realistically save with AI as a CSM?
Conservative estimates based on actual CS team usage: post-call follow-up saves 12-15 minutes per call × 30 calls per week = 6-7.5 hours per week. Pre-call prep saves 20 minutes per call × 20 key calls per week = 6-7 hours per week. QBR prep saves 2-3 hours × 10 QBRs per quarter = roughly 2.5 hours per week. Combined: up to 14-17 hours per week if you apply AI across all three and have a high call volume. Even at half (if you only use it for some tasks), 7-8 hours is a meaningful change to how your week feels.
AI
Load More
