.svg)
Welcome back to our series on "Ignite Your GTM With AI." We are moving deep into the post-sale world. If Chapter 1 was about the landscape, Chapter 11 is about the engine that keeps the lights on: Customer Success.
In this chapter, we partner with two heavyweights in the field: Dione Hedgpeth, Momentum Advisor and former Chief Customer Officer at Sumo Logic, and Alphonso Calanoc, Director of Customer Success at Momentum.
Together, they expose a fundamental contradiction sitting at the epicenter of the SaaS business model.
Organizations invest massive amounts of capital to acquire customers. They spend months nurturing leads, flying out for dinners, and engineering complex deals. Yet, the moment the contract is signed, the machinery breaks down. The intelligence that won the deal evaporates, and the team responsible for keeping the revenue—Customer Success—is left to operate in the dark.
The data supports this grim reality. Across the SaaS industry, Net Revenue Retention (NRR) has declined from 115% to 105%, and Gross Revenue Retention (GRR) has dropped to 89-90%.
These numbers are indicators that organizations are failing to extract growth potential from their existing customer base. They are treating Customer Success as a cost center focused on retention, rather than the primary growth engine for the business.
In Chapter 11, "Account & Customer Expansion: Handoff Intelligence, Retention & Upsell Signals," we dismantle the old playbook of "health scores" and "QBRs" and replace it with a new framework: Intelligence Architecture.
The CSM Conundrum: Quarterback vs. Domain Expert
We begin by addressing the elephant in the room: Role Confusion.
For the last decade, companies have tried to solve retention problems by throwing bodies at them. They hire Customer Success Managers (CSMs) and give them an impossible mandate. They ask CSMs to be "trusted advisors" while simultaneously burying them in administrative process management.
Dione Hedgpeth captures this reality with stark clarity in the book. She notes that CSMs often become "process quarterbacks, a jack-of-all-trades, versus domain experts."
This is the "CSM Conundrum."
Customers do not want to talk to professional project managers. They want to talk to peers—experts who understand their specific domain, their market challenges, and how to use the product to solve them. That is how deep relationships are built. That is how cross-sell opportunities are unlocked.
But the CSM cannot be a domain expert because they are too busy running the internal motions—scheduling meetings, updating CRM fields, chasing down support tickets, and manually aggregating data for reviews.
The quarterbacking is necessary for the business, but it is undervalued by the customer. The domain expertise is valued by the customer, but it is underdelivered by the business because there is no time left in the day.
Most leaders try to solve this by splitting the roles—adding Account Managers, Renewal Managers, and Onboarding Specialists. This breaks the bank and confuses the customer with too many faces.
The solution is not more people. It is a fundamental architectural shift that allows AI to handle the quarterbacking so humans can handle the strategy.
The "Amnesia" of the Sales-to-CS Handoff
The role confusion is compounded by a structural data failure that occurs the moment a prospect becomes a customer.
During the sales cycle, your team generates an immense amount of intelligence. They map stakeholder dynamics. They uncover implementation fears. They identify competitive threats. They learn exactly what "value" means to that specific buyer.
Then, the deal closes.
In 99% of organizations, this intelligence evaporates. The handoff process is typically a single transition meeting where Sales "downloads" their knowledge through a subjective summary. Critical context about decision-making patterns and strategic priorities gets filtered through individual interpretation rather than preserved as structured data.
Customer Success teams inherit the customer, but they do not inherit the intelligence that made that customer a successful prospect.
They start from zero. They ask the customer the same questions Sales asked three months ago. This is not just inefficient; it damages trust. It signals to the customer that your organization does not listen and does not share information.
This structural amnesia makes it impossible to build an effective early warning system. Without the historical context of why they bought, you cannot accurately predict when they might leave.
Why Your "Green" Health Scores Are Lying to You
This lack of context leads to the second major failure point discussed in Chapter 11: the broken signal system.
Every CS leader has experienced the "Green Account Surprise." This is when a customer who has a health score of 95/100—perfect usage, bills paid on time, zero support tickets—suddenly sends a non-renewal notice.
We have relied on health scores, usage data, and sentiment signals for years, yet they continue to fail us.
Dione explains that one reason for this is that CS leaders often haven't defined what the customer experience should look like. They lack a definition of "Moments of Truth"—the specific milestones a customer must hit to derive real value.
Without a map of these moments, you are rudderless. You are measuring activity, not impact.
Furthermore, traditional signals are lagging indicators. By the time a "usage drop" triggers an alert in your dashboard, the decision to churn was likely made weeks ago. A new executive with ties to a competitor may have entered the picture. A champion may have left. These are the signals that matter, but they are buried in unstructured data—emails, Slack channels, and Zoom calls—that your traditional health score calculator ignores completely.
The Solution: Intelligence Architecture
To stop the bleeding and restart the growth engine, we must move from reactive customer management to Intelligence Architecture.
This approach redesigns how information flows through the entire customer lifecycle. It creates compound benefits that multiply across functions. As Dione emphasizes, the goal isn't just to improve how CSMs execute static playbooks. The goal is to create intelligent systems that can detect behavior patterns, predict outcomes, and trigger interventions based on real-time signals.
Chapter 11 details the four dimensions of this architecture:
1. Context Engineering
AI is useless without a taxonomy. You cannot just point a Large Language Model at your Salesforce instance and hope for magic. You must establish standardized definitions for adoption, sentiment, and relationship data.
Dione outlines a rigorous approach to scoring "Relationship Level." It’s not enough to know if the customer is happy (Sentiment). You must know who is happy. A strong relationship with a user does not offset the absence of a connection with the Economic Buyer. Context engineering ensures that your data structure reflects these nuances, allowing AI to distinguish between a "noisy but safe" account and a "quiet but deadly" one.
2. Human-AI Orchestration
This is the answer to the CSM Conundrum. We must explicitly define which tasks are for the machine and which are for the human.
AI handles data synthesis, pattern recognition, and routine communications. It preps the Account Review. It scans the support tickets. It identifies the risk. Humans provide the contextual interpretation and the emotional intelligence.
As Dione notes, Enterprise customers demand trusted advisors. They stay because of loyalty and partnership. SMB customers, conversely, often prefer self-serve efficiency. An effective architecture uses AI to service the long-tail and arm the Enterprise CSMs with superpowers.
3. Signals Mapping
We need to move from "Churn Analysis" (looking at why they left) to "Signals Mapping" (detecting they are about to leave).
With AI, you can create Smart Tags for known churn risks—like "competitor mention," "budget freeze," or "executive departure." AI can then monitor every email and transcript for these tags.
But it goes deeper than keyword matching. Intelligence Architecture enables triangulation. It can flag a risk when a specific combination of events occurs: The champion hasn't attended the last two meetings AND usage of a core feature dropped by 10% AND the term "consolidation" was used in a support ticket.
This allows you to run plays mid-quarter, in real-time, rather than waiting for a post-mortem.
4. Transparent Reasoning
This is critical for adoption. If an AI system tells a CSM "Account X is at risk," the CSM will likely ignore it. The system must provide Transparent Reasoning.
It must say: "Account X is at risk because the Economic Buyer has not opened an email in 45 days, and the main admin mentioned 'budget cuts' in a call on Tuesday. Click here to see the clip."
This evidence trail allows humans to validate the AI's logic. It builds trust in the system and creates a feedback loop that improves the model over time.
The Six-Step Implementation Framework
Chapter 11 doesn't just theorize; it provides a roadmap. We outline a specific Six-Step Implementation Process to build this architecture in your organization.
While we detail every step in the book, the philosophy centers on "Compounding Intelligence."
- Map Value Milestones: Move beyond generic "onboarding" to specific "Moments of Truth" (e.g., "First global stakeholder review completed").
- Implement Intelligence Transfer: Automate the extraction of behavioral intelligence from Sales to CS.
- Establish Behavioral Classification: Tag every interaction at the source.
- Build Dynamic Detection: Replace static scores with pattern recognition rules.
- Create Automated Triggers: Set confidence thresholds for when AI acts vs. when it escalates to a human.
- Cross-Customer Learning: Use outcome data to refine the predictions.
This last step is where the magic happens. In this architecture, every churn event and every upsell becomes training data. The system gets smarter with every interaction, creating a competitive moat that static software can never replicate.
The Infrastructure for Customer Expansion
You might be reading this and thinking, “This sounds great, but my CRM can’t do this, and I don’t have the engineering resources to build a custom AI layer.”
You are right. Your CRM was built for a world of static data entry, not dynamic intelligence flow.
This is why we built Momentum.
Momentum is the Intelligence Architecture described in this chapter, fully realized. We solve the Handoff Gap and the Signals Failure by automating the flow of unstructured data into structured insights.
We Automate the Sales-to-CS Handoff: Momentum captures the "soft data" from the sales cycle—the stakeholder map, the pain points, the competitor analysis—and automatically structures it into your Customer Success platform. Your CSMs don't need a download meeting; they open the account and see exactly what the deal was built on.
We engineer the Context: Our workflow engine allows you to define the taxonomies Dione mentions. You can automate the "Relationship Scoring" by tracking interaction frequency across specific stakeholder roles (Champion vs. Economic Buyer) automatically via email and calendar integration.
We Operationalize the Signals: Momentum doesn't just give you a dashboard; it gives you action. When our system detects a risk pattern—like a stalled implementation or a ghosting executive—it doesn't just log it. It triggers a workflow. It can alert the VP of CS, draft an email for the CSM to review, or update the forecast category in Salesforce instantly.
We Enable Transparent Reasoning: Because Momentum is grounded in your actual communication data (Slack, Email, Zoom, CRM), every insight is backed by evidence. We don't give you a black box score; we show you the conversation that moved the needle.
The "CSM Conundrum" is solvable. But you cannot solve it by asking your people to work harder or by buying more disconnected point solutions. You solve it by changing the architecture of how your company learns, remembers, and acts.
Stop treating Customer Success as a cost center. Start treating it as your primary growth engine.
Ready to see what Intelligence Architecture looks like in practice? Book a demo with Momentum today.