Why Having an AI Strategy Matters – Especially in Customer Success

AI is influencing almost every part of how we work. In Customer Success, where time, trust and targeted action matter AI has the potential to become a powerful asset, but only if we use it with intention.

Too often, AI adoption is reactive. A shiny tool appears, someone says “let’s try it,” and before long, there’s confusion about what it’s for, who’s using it, and whether it’s delivering any real value. That’s the result of acting without a strategy — and it can be costly.

What is an AI strategy, and why does it matter?

An AI strategy isn’t about being an AI expert. It’s about having a clear plan:

• What problems are you trying to solve?
• Which tools are worth your time?
• How will you measure success?
• And how will you avoid introducing risk - to data, clients, or trust?

Without clarity, AI initiatives become symptoms of shiny object syndrome, often falling into the category of “good intentions, poor outcomes.”

Building your own personal AI strategy

In CS, we’re already juggling a mix of high-touch client relationships, internal data reviews, and increasingly automated processes. At its most basic level AI has a place in helping us:

• Summarise meetings and notes
• Draft personalised, accurate emails
• Spot usage or engagement trends
• Prepare renewal decks with less manual effort

But AI won’t deliver this automatically. We have to guide it. Here's how:

A practical framework to get started:

1. Start with tasks you repeat Choose 2–3 things you do weekly. For example, do you often turn calls into recaps or create content for client updates?
2. Select tools with purpose Don't use five apps for the same task. Find one or two that integrate with your existing workflow and test them properly.
3. Define your goal Be specific: “Save one hour per client preparing QBRs” is better than “use AI more.”
4. Track what works and share it Save successful prompts or templates. This is how a strategy becomes sustainable and scalable within your team.

When companies get AI strategy wrong

The risk of skipping strategy isn’t theoretical - it’s already happened, and at scale.

Take IBM’s Watson for Oncology. Once hailed as the future of cancer diagnosis, it was meant to support clinicians with treatment decisions. But the system was trained on hypothetical cases, not real patient data. The result? Inaccurate and even dangerous recommendations, including suggesting bleeding medications to patients already suffering from bleeding disorders.

After $62 million in investment and years of development, the project was scrapped. The core problem? A mismatch between ambition and execution - no clear strategy for data governance, no alignment with frontline clinical reality, and no guardrails in place for trust or safety.

The lesson is clear: even world-leading companies can stumble when they prioritise novelty over planning.

Now, it's unlikely that in your day to day CS role that you'll make a $62 million mistake, but the advice we take ourselves leads into the advice we give our clients, and when focussed the wrong way can impact us all. This is particularly important if you work in an industry where AI is part of your product suite.

What this means for our clients

Many of our clients are starting to explore AI. Some are piloting tools in learning or performance management. Others are still in the “we should probably look at this soon” phase.

Very few have a complete AI strategy. That creates risk - but also opportunity. As CS partners, we can help bridge the gap between ambition and execution.

Ask:

• “What are you trying to solve with AI?”
• “How are you measuring success?”
• “Who’s involved in shaping your approach?”

These questions often spark productive conversations and position you as a strategic partner, not just a platform expert.

Five practical tips to build your AI strategy

If you’re ready to take a more deliberate approach to AI, here are five things you can do this week:

• Audit your workflow Pick two tasks where AI could save time or improve consistency.
• Trial one tool properly Use it for a full client cycle. Measure what it actually helps with, not just how “impressive” it looks.
• Be specific about outcomes For example: “reduce prep time for QBRs by 30% using AI-generated usage summaries.”
• Encourage team conversations Build a shared library of successful prompts, use cases and lessons learned.
• Start AI conversations with clients Ask what they're exploring, where they need guidance, and how they’re approaching risk or governance.

Final Thoughts

You don’t need a 30-slide roadmap or a dedicated AI budget to get started. You just need to be intentional. AI will increasingly shape our roles - but whether it creates more value or more noise depends on the strategy behind it.

In Customer Success, that strategy starts with understanding how AI can help you serve clients better, faster, and more insightfully - while keeping trust, context, and human judgement front and centre.

Have you started shaping your own AI approach in CS? I’d love to hear what’s working (or what’s not).