Home / Blog / In Paris, AI Is Scaling Customer Success Capacity, Not Strategy

header for In Paris, AI Is Scaling Customer Success Capacity, Not Strategy

In Paris, AI Is Scaling Customer Success Capacity, Not Strategy

Customer Success teams are not struggling to adopt AI. They are struggling to trust it. AI in Customer Success strategy is already reshaping post-sales workflows, with notes automated, emails drafted, and account summaries appearing instantly.

The operational lift is real, measurable, and already part of the daily rhythm of most teams. But something more fundamental is not happening. Customer Success is not becoming more strategic as a result.

At the latest Customer Success Snack event in Paris, hosted at SAP and led by city leads Aurélie Lemaire and Rachel Herlihy, around 70 post-sales professionals came together to exchange how AI is actually being used today and where it is falling short. What emerged across sessions was not a tooling gap. It was a trust ceiling.


The Trust Ceiling Is the Real AI Bottleneck

Most teams in the room described themselves as being in a phase of active experimentation, particularly when defining how AI in Customer Success strategy should operate in practice. AI is present across workflows, yet far from autonomous in practice. The use cases are already there, from summaries and email drafting to QBR preparation and RFP automation.

Some teams are even pushing AI into more proactive territory, using it to monitor early warning signals such as usage drops, campaign underperformance, or shifts in engagement. But as Caroline Bresson, Senior Customer Success Manager at Smartway, highlighted in her session on AI agents, the friction starts the moment AI produces an output.

Teams do not execute on it. They review it, validate it, and often rewrite it before it reaches the customer. This creates a consistent pattern. Time saved on execution is immediately lost to verification, creating what many described as a “verification tax.” It limits how far teams are willing to trust AI, especially in situations that require judgment or customer sensitivity.

Several teams also shared what not to do. Sending AI-generated emails without review, relying on AI to assess relationship quality, or assuming summaries are always accurate consistently led to errors and loss of trust.

Until that changes, AI remains an assistant, not a decision-maker.


Data Quality Is Quietly Dictating AI Outcomes

If trust is the visible barrier, data quality is the invisible one shaping everything underneath. AI performance is directly tied to the quality of the CRM. And most CRMs are not ready.

In her CRM-focused session, Julie Cotovio, Manager – Customer Success Consulting | Enterprise Segment at LinkedIn, reframed the issue clearly: the CRM must be treated as the single source of truth, or AI will scale inconsistency instead of insight. Teams highlighted recurring issues such as fragmented account context, inconsistent notes, and missing historical data. This lack of structure makes it difficult for AI to produce reliable outputs.

The consequence is predictable. AI does not just reflect poor data quality. It amplifies it. One of the most practical shifts discussed was the need to build structured, AI-ready knowledge bases before deploying agents. Without that foundation, trust does not improve. It erodes.


AI Is Replacing Work Faster Than It Is Redefining It

There is no question that AI is already eliminating large portions of manual work. Teams have stopped writing notes manually, building decks from scratch, and searching across disconnected tools. As Alexia Narlian, Head of Customer Success at Brevo, demonstrated in her session on AI-powered workflows, meeting preparation, summaries, and follow-ups are now largely automated, significantly reducing administrative overhead. This shift is changing execution, but not yet redefining AI in Customer Success strategy.

This extends beyond meetings. Teams are increasingly using AI to generate structured account context documents that consolidate history, stakeholder mapping, and strategic priorities, dramatically accelerating QBR preparation and internal alignment.

In parallel, Giorgia Pedenzini, Senior Customer Success Manager at Gainsight, emphasized how teams are leveraging AI to accelerate internal processes such as account planning and knowledge retrieval, reducing dependency on other functions. Some are even using AI to extract competitive intelligence from sales calls and RFP data, feeding battlecards that inform renewal and expansion conversations and strengthening how teams approach aligning Customer Success with Sales and Product.

The efficiency gains are real. But the strategic shift is not following. The time saved is not being reinvested into deeper customer engagement. Instead, it is being absorbed into larger portfolios and increased activity expectations.

AI is increasing capacity. It is not increasing strategic depth.


The “Magic Wand” Gap Reveals the Real Ambition

When participants were asked what they actually wanted AI to do, the answers shifted. The focus moved away from automation and toward decision-making. This is where Charlotte Cobbaert, Enterprise Account Executive at Gainsight, introduced what the group described as the “magic wand” gap. The difference between what AI is doing today and what teams actually need.

They are not asking for faster execution. They are asking for better judgment. Teams want AI that can anticipate customer needs, prioritize work based on risk signals, and guide decision-making across accounts.

Complementing this, Valeriya Titlinova, Regional Customer Success Leader EMEA at SAP, highlighted the importance of feeding AI with structured historical insights, such as churn signals and win/loss data, to move from reactive outputs to proactive intelligence. This is a fundamentally different expectation. It requires context, accuracy, and reliability at a level that current implementations do not consistently deliver.


Attention, Not Automation, Is Still the Hardest Problem

Not all discussions focused on AI. A major theme that surfaced was the increasing difficulty of capturing customer attention in an already saturated environment.

In sessions led by Elie Setruk, Head of Customer Success at Riot, the challenge was framed clearly: customers are not lacking information; they are lacking attention. This shifts the role of Customer Success from delivering updates to delivering relevance. Overly generic communication, excessive data, and meetings without clear outcomes all contribute to disengagement. More communication does not mean better engagement.

Sally Hamdan, Team Lead, Customer Success Southern and Emerging Markets at Asana, expanded on this by emphasizing the importance of role-specific personalization and high-impact messaging. Generic outreach is no longer effective, and even small changes like subject line clarity or channel selection can determine whether a message is seen at all.

At the same time, Nathalie Berger, Sr Director, Customer Success, highlighted that attention is constrained by competing priorities inside customer organizations. When a product is not mission-critical, Customer Success must work harder to demonstrate value in a concise and targeted way.

Some teams are starting to use AI to support this effort, for example by surfacing sentiment patterns or highlighting potential bias in how customer interactions are interpreted. But even here, AI acts as a support layer, not a replacement for judgment.

AI can help produce communication faster, but it cannot make it meaningful.


CRM Is Becoming a Strategic Asset Again

The CRM is quietly returning to the center of Customer Success strategy. Not as a reporting tool, but as the operational backbone of decision-making.

The principle is simple: if it is not in the CRM, it does not exist. But this only works if the CRM is structured, clean, and shared across teams. In her session, Julie Cotovio emphasized that the goal is not to capture more data, but to capture the right data in a way that enables action and collaboration.

This is where more advanced AI use cases begin to emerge. Some teams are connecting internal systems to create knowledge agents that can instantly answer questions like “What are the risks on this account?” or “What happened in the last QBR?” without manual digging.

AI plays a role here, but only as an enabler. It can structure and summarize information, but it cannot fix broken inputs. Once CRM data feeds AI systems, any inconsistency becomes scalable, making data quality not just an operational issue, but a strategic one.


What This Means for Customer Success Leaders

The discussions in Paris point to a clear shift in how AI in Customer Success strategy should be approached. The question is no longer how to use AI, but how to build an organization that can actually leverage it.

Three implications stand out:

  1. AI strategy must start with data strategy. Without structured, reliable data, AI adoption will plateau quickly.
  2. Efficiency gains must be reinvested intentionally. If not, they will default to increased volume rather than increased impact.
  3. Trust is built through governance, not adoption.

As Caroline Bresson emphasized, keeping humans in the loop is not a temporary safeguard. It is a permanent design principle for using AI in customer-facing environments.


The Shift Is Already Underway

What stood out in Paris was not the presence of AI, but the clarity around its limitations. There were no inflated promises, just operators comparing what actually works and what doesn’t.

This is where Customer Success strategy is evolving. Through shared experience, not theory.


Join the Conversation

If you want to stay close to how Customer Success is actually evolving, these are the conversations that matter. The ones grounded in real constraints, real trade-offs, and real execution. The next Customer Success Snack Paris takes place on October 9, at the Partoo office, bringing together post-sales leaders for another morning of peer-led discussions.

Find your hyper-local community

Find your local community in Dublin, Amsterdam, Stockholm, Malmo, Copenhagen, Berlin, Munich, Hamburg, Ghent, Paris, Madrid, Zurich, Bucharest and more.

Get invited to Events, Job alerts & More. Straight to your inbox.

close-link