← All Playbooks
Crawl 1–3 weeks
Introduce AI Copilots for CSM Productivity
Get CSMs using AI tools for call summarization, email drafting, and meeting prep to cut administrative overhead and reclaim time for customer work.
Why This Matters
Crawl-stage teams are lean: 1–3 CSMs doing everything. A significant portion of CSM time goes to notes, follow-up emails, and internal prep, not actual customer conversations. AI copilots handle the most repetitive writing tasks with minimal setup. The goal is not transformation; it's reclaiming two to four hours per week per CSM so they can cover more accounts or go deeper on at-risk ones.
Action Plan
- 01 Audit where CSMs spend time on admin: post-call notes, follow-up emails, QBR prep, internal status updates. Pick the top two to three time sinks
- 02 Select one AI tool to pilot (your existing stack may already have one: Gong, Chorus, Salesforce Einstein, Copilot for M365, or a standalone like Otter.ai)
- 03 Run a two-week pilot with one or two CSMs: use AI for post-call summaries and one type of customer email
- 04 Define a human-review step: every AI-drafted output should be reviewed before sending to a customer
- 05 Capture time saved per week and qualitative feedback from the piloting CSMs
- 06 If the pilot shows value, roll out to the full team with a one-hour onboarding session on prompt hygiene and review practices
- 07 Document which use cases work well and which produce unreliable outputs. This becomes your first AI usage guide
Metrics to Watch
Related Principles
Common Pitfalls
- Sending AI-generated emails without human review. Hallucinations and wrong customer details erode trust fast
- Piloting too many tools at once. Pick one, prove value, then expand
- Measuring adoption (did CSMs use it?) instead of impact (did it save time and maintain quality?)