.svg)
Solving the CSM Conundrum with AI: From Catch-All Chaos to Strategic Clarity
Customer Success Managers (CSMs) are under siege. They’re asked to do everything (onboard, drive adoption, manage renewals, deliver QBRs, field escalations, collect feedback, surface insights, build relationships) and still show up with a smile.
The issue runs deeper than work volume; it’s a structural mismatch between what leaders expect and how CSMs are deployed. What CEOs say they want (strategic guidance, domain expertise, customer impact) is at odds with the way CSMs are actually deployed.
The results are always the same. Teams burn out. Value realization lags. And the CSM function, ironically, gets blamed for not delivering what it was never set up to do in the first place.
That’s the heart of the “CSM Conundrum.”
In our latest AI Advantage webinar, Momentum’s Head of GTM Growth Jonathan Kvarfordt (aka Coach K) and Director of Customer Success Alphonso Calanoc were joined by Dione Hedgpeth, former Chief Customer Officer at Sumo Logic and a seasoned CX leader with over 25 years of experience building CS teams from the ground up.
What followed was an unfiltered conversation about how AI can help fix what leadership failed to define, why the “unicorn CSM” is a fantasy, and how the future of CS hinges on structured data, role clarity, and orchestration, not more hustle.
Let’s break it down.
The Role Has Morphed. The Expectations Haven’t.
Dione didn’t sugarcoat the reality: “The CSM role today is broken and not because of the CSMs. It’s a leadership failure.”
She’s seen the cycle too many times. CEOs slash CS headcount in cost-cutting moves. Then six months later, when renewals drop and adoption plummets, they scramble to rebuild what they dismantled without ever addressing the root cause.
The problem is that most CEOs can’t define what a CSM actually does.
“Everyone agrees we need CSMs,” Dione said, “but nobody agrees on what they’re accountable for. You ask five companies, you get five different motions. There’s no shared framework.”
What’s missing isn’t effort. It’s structure.
A Framework for Clarity: The 5 Core Skill Sets
To replace chaos with clarity, Dione introduced a framework for mapping the distinct skill sets required across the customer lifecycle. Here’s how she broke it down:
- SS1: Sales Core
High-skill prospecting, executive engagement, discovery; traditional seller skills. Typically owned by AEs. - SS2: Sales Transactional
Renewal execution and pricing negotiations. Doesn’t require prospecting but still sales-oriented. - SS3: Adoption Process
Project management, onboarding, and value realization. This is the traditional “day job” of most CSMs. - SS4: Domain Expertise
The most prized yet most underdelivered layer. CSMs who can speak the customer’s language and serve as trusted advisors. - SS5: Technical Expertise
Feature configuration and troubleshooting. Often owned by solutions engineers or technical account managers.
This framework immediately exposed the crux of the conundrum: most CSMs are expected to juggle SS2, SS3, and SS4 simultaneously. And if we're being honest… That’s not a job, it’s three jobs pretending to be one.
The Myth of the Unicorn CSM
“We tried to hire a unicorn,” Dione confessed. “Brilliant domain expert. Incredible communicator. But six months in, he told me he was leaving. Why? He was drowning in admin work and couldn’t use any of his expertise.”
This story resonated deeply with Alfonso and Coach K. The assumption that one person can simultaneously execute complex renewals (SS2), manage dozens of projects (SS3), and serve as a strategic advisor (SS4) is simply unrealistic.
Even worse, it leads to missed value across the board. CSMs get stuck in low-leverage work. Customers get no strategic guidance. And leadership wonders why expansion isn’t happening.
“If you define the role as everything,” Dione said, “you end up delivering very little. Even AI won’t fix that.”
But AI can rebalance the equation.
How AI Helps Solve the Conundrum, But Only With the Right Foundation
Let’s get one thing straight: AI doesn’t fix broken org design. It accelerates whatever structure you already have, good or bad.
That’s why Dione emphasized: the starting point for AI isn’t tooling. It’s doing a work breakdown of what CSMs are actually doing. “Go ask your team. Survey them. List every activity they’re responsible for. Then calculate how many hours are being wasted on low-leverage tasks.”
Once you do that, the roadmap becomes clear:
- Start with SS3. That’s the administrative drag that’s easiest to automate: QBR prep, internal notes, follow-ups, feedback routing, status updates.
- Build AI agents for SS2 and SS5. Transactional renewals and basic technical support are perfect candidates for autonomous workflows.
- Support SS4, don’t replace it. AI can’t be your domain expert. But it can turn your domain expert into a force multiplier.
As Coach K put it: “Momentum’s north star is simple. Let humans be the source of relationships and excellence, not data entry.”
It goes beyond belief. It’s built into the architecture of how teams execute.
From AI Assistant to Co-Pilot: Enabling Strategic CSMs
Alphonso offered a practical example of how this looks inside Momentum. When a deal hits the “finalizing paperwork” stage, CS gets an instant handoff note. Not just from the rep, but from the AI.
Pulled from two quarters of call transcripts and email threads, this note summarizes:
- Stakeholders involved
- Key deliverables
- Execution risks
- Timelines and expectations
- Sentiment and signals of risk
No chasing notes. No Slack scavenger hunts. Just instant clarity.
“That handoff sets us up to validate value, ask the right questions, and immediately start planning a high-impact kickoff,” said Alphonso. “And we do it without wasting a week gathering context.”
More importantly, this is AI that moves past transcription and it turns noise into signal. With Momentum, all of that signal-rich customer context is turned into clean, structured data fields that flow into Salesforce or Snowflake. The result is scalable insights across every segment, not just anecdotal hunches.
As Dione noted, “Structured data is what unlocks AI’s next level. That’s how you move from automation to orchestration.”
What Real AI Orchestration Looks Like in CS
Speed is just the baseline. The real power of AI is in unlocking what wasn’t possible before:
- Predicting churn months in advance
Momentum surfaces patterns from closed-lost and churn data to flag risks before they’re visible to humans. - Running precision playbooks
Whether it’s building a champion, responding to a procurement blocker, or driving expansion, Momentum triggers the right follow-up at the right moment. - Segmenting smarter
Dione floated a provocative idea: what if SMB accounts become more attractive because their AI usage gives them leverage they never had before? “Don’t assume your segmentation stays the same,” she said. “AI might change your ideal customer profile.” - Hiring differently
When repetitive tasks are handled by AI, you can hire for what actually drives outcomes: domain fluency, emotional intelligence, and trust-building.
Momentum’s belief is simple: your CS team should spend 70% of their time in the room with customers, not preparing to be in the room.
And with AI, that’s finally possible.
The Path Forward: From Generalists to Domain Specialists
What happens when AI removes the administrative weight of SS3 and automates the transactional grind of SS2? You free up the one skill set that actually unlocks revenue: SS4, domain expertise.
Dione framed it clearly: “In the future, you won’t search LinkedIn for ‘Customer Success Manager.’ You’ll search for people who understand your industry better than anyone else.”
That doesn’t mean turning every CSM into a seller. It means enabling them to become what customers truly want: trusted partners who understand their day-to-day and can connect product capabilities to real-world outcomes.
But here’s the catch: even the most brilliant domain expert can’t do their job if they’re buried in repetitive work. Solving this takes more than headcount. It takes clarity, focus, and a system that supports real execution.
That’s what Momentum was built to do.
Final Takeaways: Orchestrating the Future of Customer Success
Dione, Alfonso, and Coach K closed the conversation with a shared outlook: we’re entering a new era of customer success.
Resource constraints aren’t new. What matters is how clearly your team can focus on what moves the needle.
Here’s what the best teams are already putting into practice:
- Stop asking CSMs to do everything. Define roles around actual skill sets, not catch-all titles.
- Attack the SS3 burden first. This is where AI delivers immediate relief and sets the stage for strategic work.
- Automate the edges. SS2 and SS5 are ripe for AI agents, especially in long-tail or SMB segments.
- Structure your data. If your AI can’t move insights into your CRM or warehouse, it’s not orchestration, it’s shelfware.
- Elevate your CSMs. With AI handling logistics, give your team the time and tools to build relationships, not just decks.
At Momentum, we’re building toward a world where domain experts can thrive, data works for you (not the other way around), and your team finally has the operating system it needs to scale human impact, not just headcount.
Want to See the Future of AI-Powered CS in Action?
If this conversation hit home, we invite you to see how leading CS teams are using Momentum to:
- Automatically generate structured customer insights from every interaction
- Trigger risk signals and renewal workflows in real time
- Scale domain expertise through orchestration, not more meetings
Book a Momentum demo and turn your CS function into the strategic growth engine it was meant to be.
Or follow us on LinkedIn for more expert insights and webinar recaps like this.
Let’s fix the CSM conundrum together.