For the past few years, AI has promised to revolutionize marketing. The attraction? Better results, booming pipelines, and conversions that practically run themselves. But here’s the real question: has AI actually delivered on all that hype? The answer is it depends.
This year has been a real turning point for companies trying to achieve ROI from AI in marketing. Business leaders had high hopes: non-stop leads, smarter targeting, and content so personalized it feels custom-made for every prospect.
But a recent blog post from Writer, “AI ROI Calculator: From Generative to Agentic AI Success in 2025,” sheds some light on why those expectations haven’t always been met. The article points out a tough reality: an incredible 95% of AI projects fell short of delivering the ROI leaders were looking for, mainly because organizations relied on old-school metrics. As Matthew Olson, the writer, puts it, the problem isn’t the technology itself, it’s how most companies calculate business value.
From what I’ve seen, the underwhelming ROI from AI has just as much to do with how it’s rolled out and what it’s actually designed to achieve. The rush to embrace flashy, content-generating AI led many companies to focus on squeezing out costs, rather than creating new value.
Over the past several months, I’ve been advising clients and prospects about a better way forward. Here are three big ways AI can, and should, move the needle:
- Make the customer experience shine through real hyper-personalization of both the content and the journey.
- Deliver much deeper, sharper insights into the customer base for your team.
- Steer product and service development toward what really matters in the market.
The trouble is, even when companies chase these goals, AI often isn’t used the right way. That’s led to a lot of disappointment with Agentic AI, but things are changing. More and more organizations are stepping back and rethinking exactly how to use AI, and those who get really clear about their goals and objectives are starting to see real results.
When AI is laser-focused on goals like understanding customer preferences, tracking buying behavior, and measuring what customers truly value, the payoff can be huge, at least for the clients we’ve worked with. But it’s not always smooth sailing. These projects can also reveal when a company just isn’t ready for AI, or when no one’s quite clear on the best ways to use it.
I’ve seen leaders invest a fortune in AI, only to have their teams use it for something different than they intended. Other times, teams ask for more support or bigger budgets, but don’t effectively communicate what they actually want to achieve, so requests get denied. And sometimes, the tools and goals are aligned perfectly, but the people running them haven’t had the right training. Unsurprisingly, none of these scenarios leads to better business outcomes.
If AI has let you down, maybe it’s time to dig into what’s really causing the disconnect, and reset your approach if needed. If this sounds familiar, and you want some help figuring it out, let’s connect.
