AI is not going to revolutionise your CRM strategy. Just yet.

Since stepping away from corporate and into consulting, I’ve had the opportunity to get ‘under the hood’ of a number of new and evolving platforms. Martech stacks are being rebuilt and roadmaps are being rewritten. AI is being embedded everywhere. There’s no avoiding it. Every CRM platform demo now starts the same way. A sleek dashboard. A predictive insight. An auto-generated journey. A subject line written in seconds. And then the line: “This is powered by AI.” There’s one consistent theme, AI is the headline.

Don’t get me wrong, I like the concept. I’m not anti-AI. The technology is impressive. The speed gains are real. The interfaces are more accessible than ever. I do think expectations are running ahead of reality. Because when you move beyond the demo layer and look at what’s happening beneath the surface, something becomes clear. A lot of what is now being labelled as “AI” is what we used to call data modelling, predictive analytics and automation logic. While being an important step, it’s not exactly new. The evolution of turning ‘the difficult and complex’ into being more accessible is also being branded under the “AI umbrella.”

Acceleration Is Not Reinvention

In many cases, AI is not introducing fundamentally new CRM capabilities. It’s making what was formerly complicated simple and available to be utilised by marketers with both technical and non-technical skillsets. What previously required SQL, reference tables, AMPscript, CSS, etc, can now be executed in minutes via a prompt or a configuration layer. Fifteen years ago, building a sophisticated lifecycle segmentation or automated product recommendation engine required technical depth and time. Today, a junior team member can assemble something similar in a fraction of the time.

While this is progress, it’s acceleration, not reinvention. The capability itself often existed. It just required technical skill to unlock it. AI reduces the barrier to entry. It doesn’t magically invent entirely new strategic levers. Why does this distinction matter? If we confuse accessibility with transformation, we risk misallocating both expectation and investment in our CRM program.

The Real Impact: Resource management and optimisation

Where AI is genuinely transformative in its current state is resource management Anyone who has built or scaled a CRM function knows this challenge well; implementing a strong platform is only half the battle. The other half is finding the right people to use it. For years, experienced CRM operators with the right blend of technical capability (SQL, scripting, automation logic), commercial acumen and lifecycle strategy have been difficult to recruit. The talent pool has always been narrower than the demand. Onboarding a sophisticated CRM platform often meant simultaneously competing for scarce technical CRM talent. “AI enhancements” within modern CRM platforms are quietly changing that equation. They compress the time between idea and execution and open the ability for even complex campaigns and customer journeys to be created by a much wider pool of CRM talent (where we have also seen the rise of the “Instagram CRM world best email marketer.”)

What previously required deep technical skill can now be configured through guided workflows, prompts or automation layers. That expands the available talent base. It reduces dependency on a handful of technically advanced operators. It allows organisations to unlock value from their platforms without waiting months to hire the perfect specialist. Junior team members can build more quickly and experiment more often, thereby constantly (and efficiently) optimising CRM initiatives. Senior team members can spend more time on commercial modelling and cross-functional strategy. Teams can test more variations without expanding headcount. That is meaningful, but it’s not the same as strategic revolution.

AI removes friction. It does not remove judgement. CRM, at its core, is still about judgement. What do I mean by judgement? Which customers do we prioritise, which incentives do we protect, where do we trade margin for volume and when do we hold price versus chase sales? No AI tool (at least not yet) understands your corporate pressure, your cash position, your inventory exposure and your brand positioning simultaneously. We humans still do.

Why I Don’t “Fear” AI in CRM

There’s a narrative emerging that AI will replace CRM teams. At this stage, I’m not convinced. If anything, AI may increase the value of CRM teams; simplifying complex and time-consuming tasks so that teams can focus on more strategic and customer focused tasks. When execution becomes easier, the bottleneck moves; it moves from “can we build this” to “should we build this?” Senior CRM professionals become more important, not less, because someone still needs to:

• Validate AI-generated recommendations and content

• Understand unintended commercial consequences

• Detect bias in data patterns

• Interpret anomalies

• Troubleshoot flawed outputs

• Align campaigns to broader business objectives

Meanwhile, junior team members gain leverage. They execute faster. They learn more quickly. They spend less time stuck in technical build and more time in analytical thinking. AI doesn’t remove CRM capability, it amplifies it (if the organisation knows how to use it properly.)

The Bigger Risk: The creation of the AI Silo?

We’ve spent the last decade trying to shift organisations from department-centric thinking to customer-centric thinking. Now are we at risk in recreating silos (just with smarter tools inside them?) CRM AI wants engagement and revenue growth. Pricing AI wants margin protection. Stock AI wants inventory efficiency. Media AI wants acquisition efficiency. Service AI wants cost reduction and containment. Individually, each system is smart, collectively, they may be misaligned. For example, if the CRM AI recommends a 20% discount to drive conversion, the pricing AI flags margin erosion, the stock AI predicts inventory risk and the service AI anticipates complaint spikes. Which objective wins? Local (department based) optimisation does not equal enterprise optimisation. As AI systems become more autonomous, these conflicts will only intensify.