After more than 20 years chasing the holy grail of engagement and behavior change, I finally had a Eureka moment.

I believe I’ve discovered why so many engagement strategies fail despite having more data than ever before.

And if I’m right, it could fundamentally change how we think about patient engagement, behavior change, and personalization.

But first, a story.

My journey to that realization started more than two decades ago.

After a five-year apprenticeship in Japan running a small advertising agency, I returned to the UK during the dot-com boom. Following a few failed startup ventures, it was a chance encounter through my appearance on the UK’s version of MasterChef, that I met one of the smartest marketers I’ve ever had the privilege of learning from: Grant Harrison.

Grant was widely credited with helping launch the Tesco Clubcard, the first and arguably most successful loyalty program ever created. Later, at Virgin LifeCare, he helped pioneer one of the first health rewards programs.

More importantly, he taught me the value of truly understanding the customer.

The idea that we could apply the same principles used to influence purchasing decisions and loyalty to something far more meaningful – helping people make healthier choices – didn’t just captivate me. It became an obsession. One that would send me down more rabbit holes than I care to admit and consume the better part of the next 20 years.

That obsession took me from population health and consumer data to ethnography, anthropology, behavioral economics, digital therapeutics, AI, and everything in between.

Along the way, I worked with Humana to explore how consumer behavior influences health outcomes. I partnered with Experian to build predictive health risk models using consumer and health data. I even spent time trying – rather unsuccessfully – to teach IBM Watson how to become a psychotherapist, incorporating recursive thinking and emotional sequencing to guide people towards better choices.

The technologies changed, the datasets got bigger, the algorithms got smarter. But one question refused to go away: Why do people know what to do, but don’t do what they know?

Why do patients stop taking medications that could save their lives? Why do some people successfully change their behavior while others struggle despite the best intentions?

For years, I thought the answer was engagement.

Like much of the industry, I was focused on helping people do the right thing.

How do we get patients to eat more fruit and vegetables? How do we get people to exercise more? How do we increase participation, activation, adherence, retention, and engagement?

It seemed like the right question.

Until I realized something fundamental: Despite decades of innovation, billions of dollars of investment, and countless engagement platforms, we’re still struggling with many of the same challenges.

We’ve spent years optimizing click-through rates, open rates, logins, time in app, daily active users, activation rates, completion rates, retention rates, Net Promoter Scores, and numerous other engagement metrics.

Yet for all that effort, many of the underlying behaviors remain stubbornly unchanged. Not because engagement doesn’t matter. But because engagement is a lagging indicator. It tells us what happened, but it tells us almost nothing about why.

The breakthrough came when I stopped asking: “How do we get people to engage?” – and started asking: “How well do we actually understand them?”

Their motivations, beliefs, fears, emotional state. Their progression from stress to self-doubt to avoidance and ultimately disengagement.

The hidden emotional journey that often begins long before a patient misses a dose, skips an appointment, or abandons treatment. The invisible factors shaping every decision they make. The things traditional data can never tell us.

That’s when it hit me.

For 20 years, we’ve been trying to optimize for engagement. We designed campaigns to increase participation. We built platforms to increase activation. Gamified apps to improve retention. Introduced incentives to drive adherence.

What we should have been building is a deeper understanding of the human beings behind those metrics.

Because people aren’t numbers. They’re stories.

And when you truly understand people, everything else becomes easier.

Personalization becomes more meaningful. Communication becomes more relevant. Behavior change becomes more achievable. Outcomes become more predictable.

That realization became the foundation for a new framework I’ve been developing called Digital Behavioral Science.

It’s built around a simple premise: Traditional data tells us what happened. Human understanding helps us understand why.

And when we understand why, we can intervene earlier, personalize more effectively, and help people make better decisions before disengagement occurs.

And understanding that “why” may be the most important opportunity in healthcare today.

Which has led me to a provocative conclusion:

AI’s greatest value may not be helping us understand the world better. It may be helping us understand ourselves better.

I’ve captured the full framework in a new white paper. If you’d like a copy, email me at: imagine@bluedoor.us

I’d love to hear whether this resonates with your own experience.