Michelle Dunst

Let’s start with you. Who are you, and what lens do you bring to understanding people online?

Hi, I’m Michelle. I’m one of those people who’s probably best described as what the Gen Z-ers call permanently online. Not in a doomscroll-y way (only sometimes), but in an always-paying-attention way. I follow conversations, behaviors, and the delightfully weird corners of the internet so my clients don’t have to.

I’ve spent the last 15+ years working in social and digital insights, but my lens goes beyond traditional social listening. I’m constantly pushing how evolving engagement, conversation, and digital behavior standards can move marketing and communications forward, not just document what already happened.

I started in the early 2010s, back when social insight meant long forum threads, blogs, and very text-heavy Twitter debates. I used Radian6 back then - anyone remember that one? Shout out! Today, insight lives everywhere: video comments, private communities, AI search, podcasts, Substack, and places most people aren’t looking yet.

My lens is shaped by a mix of experience and instinct. I keep up with trends obsessively, follow signals across platforms, and pull data from wherever it exists, because the most important insight rarely comes from the obvious place. I balance analytical rigor with curiosity, strategic thinking, and a healthy stubbornness. I’m relentless about finding the insight that actually shifts thinking, informs strategy, or unlocks an unmet need.

In my current role, I focus on building what’s next in digital insights - new ways of understanding how people learn, seek information, and engage in an increasingly fragmented, AI-influenced world. 

‍

What’s a working theory you have right now about how people behave online?

Calling it a “theory” almost feels generous at this point. It’s more like a universal truth we’re all pretending we can keep up with. Between nonstop feeds, AI-generated answers, and conversations splintered across every platform imaginable, we’ve hit peak information whiplash.

The whole idea that online and offline behavior are separate worlds? That is so 2025.  We’re all living in the Venn diagram. We scroll, we search, we double-tap, and now we ask AI to tell us what to think before we think it. The audience journey includes machines now, between the AI summaries, rankings, and answers shaping what people see next are part of the conversation itself.

What excites me most is the second layer. The comments, replies, and community threads where people drop the performance and talk like real humans again. In healthcare, that means the unfiltered patient POV, the HCPs comparing notes in unexpected communities (Reddit continues to blow up beyond expectations), the sentiment hidden in video comments, the migrations between platforms, and the signals that tell us what audiences expect long before brands shift strategy.

So if this is still a “theory,” I guess my punchline is: the future of insights is less about predicting the next 12 months and more about decoding the next 12 minutes. No pressure!

‍

What’s an insight you surfaced that you still think about? What one stuck with you?

Years ago, I was working with a healthcare brand team preparing for launch in metastatic cancer. To do that well, I immersed myself in social listening to truly understand the patient community. That work led me to a highly trusted advocacy organization on Facebook, one that had already built a close-knit forum where women connected, shared, and supported each other. They were trying to spark engagement, too, but one post went viral in the community for all the wrong reasons.

The organization asked something like, “What are you most looking forward to this year?” It was meant to inspire hope, but instead it revealed a massive gap between intention and reality. The responses were raw, honest, and unforgettable. Women talked about trying to get out of bed without pain, battling exhaustion, protecting precious family moments in limited time, or navigating the unfiltered emotional and physical toll of their disease. One person mentioned planning her funeral playlist. I’ll never forget it.

It was such a lightbulb moment for me, especially so early in my career. Engagement is great. BUT resonated engagement is everything. It’s the “we got you” feeling every brand hopes to earn, but few actually do. Because if you don’t deeply understand your audience first - their language, emotions, digital habits, and community dynamics - you can’t anticipate what they need. And when you can’t anticipate their needs, even the cleverest strategy falls flat.

‍

What’s the weirdest rabbit hole your work has ever sent you down? And what did it teach you?

I won’t even call it weird, because it was something more remarkable. The coolest rabbit hole I ever fell into started while I was doing social listening in the migraine space, specifically around cluster headaches and frequent, debilitating migraine experiences. I was mapping patient conversations, identifying influential voices, analyzing emerging creators, and tracking the topics that mattered most to patients who felt deeply unheard by both their doctors and the broader scientific community.

That’s when I stumbled on Migraineistan: a patient-built corner of the internet that felt like a different stratosphere, one you couldn’t find through hashtags and standard queries alone. Migraineistan.com was built upon beautiful artwork, and metaphors, and first-person storytelling about the emotional and physical toll of migraine. It wasn’t just descriptive - it was expressive. It stopped me in my tracks. I learned more about the lived experience of migraine in one sitting than I had in months of combing through patient posts.

It taught me that health insights are cultural, emotional, and often beautifully human.. It also reinforced something I bring to my work every day -  if you want to understand audiences, especially in healthcare, you have to understand the communities (or in this case, worlds!) they build when they think no one is watching.

‍

What skills or mindsets do you think the next generation of analysts will need?

Curiosity, cultural fluency, and AI-confidence. The best analysts trust their hunches, ask smarter questions, and explore signals others scroll past.

And for mindset - AI isn’t replacing curiosity.It’s powering it. It’s your friend, helper, and accelerator: a tool to test ideas faster, analyze new formats like video and audio, and scale insight without losing depth. The analysts who win don’t fear AI, but instead they collaborate with it.

What’s a niche community, account, or corner of the internet you’re watching right now? And why?

I spend a lot of time watching how questions drive behavior online. We see patients and HCPs go to Reddit, TikTok comments, and Facebook communities to ask things. What are symptoms? What works? What doesn’t? How do other people manage this?

Those same questions are now being asked in chatbots first, like a ChatGPT. The answers people get there often get shared back into social platforms, where communities respond, validate, correct, or challenge it. That creates a new flow of conversation that influences decisions quickly.

I pay attention to this because it helps client teams and my analyst teams better understand real patient and HCP needs, language, and expectations, knowing AI is now part of how people learn and gut-check information. It pushes me to think about how we guide clients to meet audiences with accurate, relevant content in the places they already trust.

Last non-work thing you read that shaped your thinking?

I recently read a piece on X that basically made the case for being a generalist in a world obsessed with specialization. It talked about how we’ve been taught to pick one thing and stick with it, even though the most interesting work today happens at the intersection of lots of different interests.

That really landed for me.Every interest you pick up adds another layer to how you see the world, and over time those layers start connecting in ways that feel pretty powerful.

It was a good reminder that the goal isn’t to know everything, but to know enough across different domains to ask better questions and see patterns earlier. That mindset has definitely stuck with me. My motto is always, “know enough to be dangerous”. 

‍

‍