Beyond the Hype to Actual Use Cases.
Everyone is talking about AI in customer success. Blog posts promise revolutionary transformation. Vendors tout AI-powered solutions. Conference sessions are packed with discussions about the future.
Yet when you ask CS professionals what’s actually changed in their day-to-day work, the answers are surprisingly modest. There’s a massive gap between the hype and the reality, and understanding that gap matters.
The Trust Problem Nobody Wants to Discuss
Here’s a thought experiment: would you trust AI to write your university essays? Probably yes. Would you trust AI to brush your teeth? Sure, why not. Would you trust AI to perform brain surgery? Absolutely not.
This isn’t about AI’s capabilities, it’s about human psychology. Anything high-risk, anything where the stakes matter deeply, we don’t yet trust algorithms. And much of what happens in customer success falls into this category.
Building customer relationships. Making strategic recommendations. Navigating complex organisational politics. Deciding when to escalate and when to absorb risk. These are high-stakes activities where the cost of error is significant.
This is why so many AI applications in CS remain back-office. They’re handling tasks where mistakes are recoverable, where human oversight is easy, where the risk is manageable.
Where AI Actually Works Today
The unglamorous truth about AI in customer success is that most current applications are incremental improvements rather than revolutionary changes.
Automated email responses for common questions. Predictive churn models that surface at-risk accounts. Data analysis that identifies usage patterns. Content recommendations based on customer behaviour. Basic sentiment analysis on support tickets.
These are valuable. They save time. They surface insights that might otherwise be missed. But they’re not transforming the fundamental nature of CS work.
The reason is simple: the truly valuable applications of AI in CS require something that doesn’t exist yet – widespread trust in AI’s judgement for high-stakes decisions.
The Goodwill Hunting Principle
There’s a scene in the film Goodwill Hunting where a therapist tells a brilliant young man: You’re just a kid. You don’t have the faintest idea what you’re talking about…If I asked you about art, you’d probably give me the skinny on every art book ever written…But I bet you can’t tell me what it smells like in the Sistine Chapel.
The point is, you can memorise information without understanding experience. You can recite facts without wisdom. You can process data without judgement.
This is precisely where AI sits in customer success right now. It can tell you what customers are doing. It can’t tell you what it feels like to be in their position. It can analyse patterns but can’t truly understand context, nuance, and the human elements that drive customer behaviour.
Until AI can bridge that gap, and it’s not clear it ever will, human judgement remains essential for the most valuable CS work.
The Persona and Relationship Gap
The CS community has become so distracted by AI discussions that we’ve stopped talking about something more fundamental: personas and people.
At the core of customer success is human behaviour. How people make decisions. What builds trust. How relationships develop. These elements are governed by things like status, certainty, autonomy, relatedness, and fairness, the psychological framework known as SCARF.
The relatedness component – trust between people – is what AI struggles with most. Customers might accept AI-generated insights, but they still want strategic guidance from humans they trust.
This isn’t going to change quickly. It’s deeply embedded in how we’re wired as social creatures. We trust people who demonstrate experience, who share similar challenges, who understand our context in ways that feel authentic rather than algorithmic.
The Application Question
When evaluating AI in customer success, the critical question isn’t “what can AI do?”, It’s “what use cases are compelling enough that the technology becomes almost irrelevant?”
For example: imagine AI could predict with 90% accuracy which customers will churn in the next 90 days and why. That’s a compelling use case regardless of how it’s achieved. The technology becomes secondary to the outcome.
But we haven’t seen many use cases at that level yet. Most AI applications in CS are incremental, helpful but not transformative. They make existing processes faster or more efficient without fundamentally changing what CS teams can achieve.
This is why the AI conversation feels frustrating to many practitioners. There’s endless hype about potential with limited delivery of concrete, meaningful improvements.
The Knowledge Commerce Alternative
Whilst everyone focuses on AI, there’s a more immediate opportunity: knowledge commerce. This is the ability to bring expertise that customers would pay for independently of your software.
If you left your job tomorrow and offered to consult for your customers, would they hire you based on what you know? That’s knowledge commerce.
Building this requires deep industry expertise, battle-tested experience applying that expertise, and the ability to articulate value with impact. These are inherently human capabilities that AI complements but doesn’t replace.
Ironically, whilst the industry obsesses over AI, the most successful CS professionals are doubling down on distinctly human skills: building relationships, navigating complex organisations, delivering insights that come from experience rather than data.
The Sales-CS Technology Divide
Here’s an unexpected consequence of AI: it might widen the gap between sales and CS rather than bridging it.
When one team becomes expert in using AI tools whilst the other doesn’t, it creates asymmetry. The team with AI expertise makes different decisions, works at a different pace, and operates with different assumptions than the team without it.
This only reinforces existing tensions. Sales might use AI to identify expansion opportunities CS doesn’t see. CS might use AI to flag risk in deals that sales wants to close. Without shared understanding of the tools and data, these become points of conflict rather than collaboration.
The solution isn’t to avoid AI, it’s to ensure that tools serve both functions equally, with shared data sets and aligned incentives.
What Would Actually Help
Rather than chasing AI for its own sake, CS teams should focus on use cases that solve real problems:
Customer health scoring that actually predicts behaviour rather than just aggregating metrics. Automated playbook suggestions based on similar customer situations. Real-time conversation guidance during customer calls. Intelligent routing of customer requests to the right resources. Proactive identification of expansion opportunities based on usage patterns.
These use cases exist. Some vendors deliver them reasonably well. But they’re not revolutionary, they’re evolutionary improvements that make CS work more efficient without fundamentally transforming it.
And that’s fine. Not everything needs to be revolutionary. Incremental improvements compound over time.
The Generational Divide
There’s an emerging divide between CS professionals who grew up with AI and those who didn’t. Younger entrants to the profession treat AI as a natural tool, like previous generations treated email or CRM systems.
Older professionals often approach AI with more scepticism, having seen multiple waves of technological hype that promised more than they delivered.
Neither approach is wrong, but the divide creates challenges. Teams need to find common ground on how AI fits into their workflow without becoming either dismissive or overly credulous.
The Remote Work Factor
AI adoption in CS is happening alongside another major shift: remote work becoming standard. This combination creates interesting dynamics.
Remote teams often rely more heavily on asynchronous communication and documented processes, exactly where AI can add value. But they also struggle more with relationship-building and trust development, areas where AI can’t help much.
The most effective remote CS teams use AI to handle administrative overhead whilst being deliberately over-invested in human connection. They don’t assume AI can replace the water cooler conversations and casual interactions that build team cohesion.
What Actually Matters
Here’s what matters more than AI in determining CS success over the next few years:
- Commercial acumen: understanding business models and speaking the language of outcomes.
- Industry expertise: bringing knowledge that customers can’t get from the software alone.
- Relationship skills: building trust that survives economic uncertainty and competitive pressure.
- Cross-functional collaboration: connecting different parts of the organisation to drive customer value.
- Strategic thinking: moving beyond tactics to influence actual business strategy.
AI might enhance these capabilities. It won’t replace them.
Final Thoughts
AI in customer success will continue evolving. Tools will improve. Use cases will become more compelling. Trust will gradually develop as people see consistent results.
But the timeline is longer than most people think. And the transformation will be more gradual than the hype suggests.
In the meantime, CS professionals should treat AI like Tony Stark’s Iron Man suit – it makes you more capable, but you’re still the one making decisions. The suit enhances your abilities; it doesn’t replace your judgement.
Focus on developing the fundamentally human skills that create value: expertise, relationships, strategic thinking, commercial acumen. Use AI where it helps, ignore it where it doesn’t, and maintain healthy scepticism about revolutionary promises.
The future of customer success isn’t about AI replacing humans. It’s about humans who understand AI working alongside humans who understand customers. The combination is more powerful than either alone.
This post was inspired by Episode 1 of the Doing It for Retention Podcast. Watch and listen online now.
About ABR Talent
ABR Talent specialises in connecting exceptional professionals with leading SaaS companies. Whether you’re making your first move into tech or seeking your next leadership role in customer success or sales, we understand the unique challenges and opportunities in the market.
Get in touch to discuss your career journey or hiring needs.
