There’s a memorable scene in “The Sorcerer’s Apprentice” from Disney’s Fantasia in which Mickey Mouse cuts up an enchanted broom into hundreds of pieces, only to be overwhelmed by hundreds of smaller versions of the same broom. It’s an allegory that marketers might take to heart when looking at customer segmentation. In many cases, segmentation creates a bigger problem by creating smaller versions of the same problem—namely, how to have a more personalized relationship with customers.

You can’t fault marketers for not wanting to know more about their customers. Last year, three out of four companies (seventy-eight percent) had either implemented or planned to implement a customer data platform (CDP) to support customer-centric initiatives. And in a world where Amazon, Netflix, and Spotify (among others) are redefining the expectations of Personalization, CMOs understand that the pressure is on to deliver richer and more personalized digital experiences. In fact, research shows that Digital Transformations are significantly more likely to result in double-digit growth when they’re spearheaded by CMOs. Yet connecting the dots between customer data and great customer experiences remains a challenge for nearly every business.

Sweeping the Problem Under the Rug

We have more data than ever before, and more powerful analytics tools too, so why are marketers still struggling with customer churn and Marketing campaigns that miss their mark? The culprit, in many cases, is complexity: it’s hard to have a one-to-one relationship with your customers when you’re treating them as one drop in a larger bucket. Segmentation was supposed to solve the “personality” problem by customizing the customer experience, but it only provides the illusion of personalization. We’ve all had the experience of interacting with a brand that seems to know us, only to receive a completely off-target offer from them that shatters the illusion.

Couldn’t more, and more refined, segments address this issue? Not in any meaningful way. The new reality is that businesses need to see their customers as a segment of one. The more that marketers try to blend customers into demographic generalizations or shared data affinities, the less personal their interactions become. And these interactions should align with changing developments in the consumer’s life, or what we often call the customer journey. As an example, when a consumer starts a family (or starts a new job), their interests will often shift, and the Marketing conversation needs to shift with them to remain relevant.

Your Data Won’t Work If Your Workflow Doesn’t

A segment of one sounds good, you’re thinking, but how do I manage thousands (or even millions) of personalized Marketing interactions when we struggle to execute campaigns across a handful of segments? By automating and orchestrating those interactions to recommend relevant content. The problem for most marketers isn’t that they don’t know enough about their customers, but that they don’t do enough with what they know.

Let’s say you’re a digital media company, and improving content engagement through subscriptions is your primary goal. Delivering content based on job description or industry might increase engagement slightly, which can give you the illusion that your customer segmentation is working. But when you begin looking at how each user interacts with your content—what they read, when they read it, what they share and with whom—individualized patterns emerge that allow you to tailor your content experiences at a much more personal level. For example, the fact that Jane Smith is a CIO in the Healthcare industry would immediately suggest Healthcare technology content, but her recent interest in Retail industry news on her mobile device after 5:00 pm could signal an imminent career shift that makes Retail technology content much more relevant to her right now.

Now, your customers don’t expect you to manually micro-manage each interaction you have with them. But they do expect you to notice and adapt to their interactions with your brand. If they visit your website and look for the same kind of product or information each time, they expect to see shortcuts to that information over time that personalize the web experience. If they start shopping for organic food in your stores, you can probably stop sending them coupons for Raman noodles. These kinds of adaptations transcend segmentation and get to the core of the personalized experience.

The good news is, you don’t have to guess what your customers are interested in right now. The trick is picking the right CDP to harness customer data that’s hiding in plain sight. Customers send you millions of signals throughout their online activity. You’re already creating a composite picture of each customer that tells you exactly what they want, when, and how they want it.

Given that there are thousands of signals being created, it would be impossible for any person to know which signals are important at any given time and what action to take based on each signal for every individual. It takes Machine Learning algorithms to provide insights and recommendations that allow for an infinite number of possible next steps.

With Machine Learning it is easier than ever to connect what you know with what you do every day: send emails, display digital ads, engage with new content offers, etc. And the best way to do that is by automating your working flow and orchestration with a Customer Data Platform. Name brands such as Netflix have been reaping ROI from this new Marketing paradigm for years. Think about how the popular streaming service delivers Hyper-Personalized content. Netflix harnesses customer signals gathered from viewing and search habits to make recommendations. Marketers can do this too. By offering content that most pertains to the signals that have been received over the past along with the signals that are being sent at any particular moment. This allows you to constantly provide the most relevant content to your customers.

Good Marketing isn’t magic; it’s the result of the right customer intelligence combined with the right technology. So the next time someone brings up customer segmentation, ask yourself: Do we really want to take the Mickey Mouse approach to our problem, or do we want to use Machine Learning and Advanced Marketing technology to deliver the kind of customer experiences that the world’s most successful digital businesses are using? As for which technology is the best fit for that job, well, that’s an animated discussion for another day.

Read more: Performance-To-Pay for Scaling Partner Marketing Programs