That might seem obvious. But it's definitely not easy.
Signal acquired
SourceMike Costanzo
Strength15+ yrs
InsightThe opportunity sits in plain sight, consistently ignored, because the cost structure made it uninvestable.The race is won before the product ships.Defensibility isn't built. It accumulates.In a non-consumption market, customers don't try you and leave. They never arrive.Positioning isn't a marketing function. It's the mechanism by which latent demand becomes active demand.The channel and the message are one motion.

I started in bands: making posters, booking shows, selling tickets, building an audience together. Songs, then albums, were the first products I shipped. I learned early that good work takes more than an idea. It takes a team that believes in it and keeps showing up.
At Rangle, I built the UX and product design practice from zero, hiring 15+ people as the company grew from 13 to 150+. I was a designer who could code, obsessed with the craft. Together, we developed the practices and standards that helped teams ship excellent client work, one product at a time.

At Onist and My Pain Sensei, I saw that teams can ship features and still lose sight of the people they serve. I brought product and engineering closer to users through continuous discovery; at My Pain Sensei, I built a remote product and engineering function from zero. Reconnect the team to the problem, and direction gets clearer.
Since 2023, I've led teams building AI products: first a personalized shopping experience for Nike, then a dozen GenAI products for clients in banking, insurance, and accounting. The technology changes constantly but the process doesn't: learn by making, moving quickly, and staying close to the customer.
I still run discovery and prototype in Claude Code and Cursor. Staying close to the work helps me coach teams through uncertainty, test ideas sooner, and close the gaps between product and engineering. The aim is to build a team that can keep moving after the prototype—and products people trust.
I'm always searching for the signal in the noise

Case study
How to find product-market fit quickly, and take the stress out of moving.

Case study
Turning a failing product into a successful exit in less than 2 years.

Case study
Pushing the boundaries of personalized shopping with a Gen-AI-enabled POC for one of the world's leading brands.

essay · ↗
Recent advances in generative and agentic AI make one specific thing possible that wasn't before: effectively serving customers whose demand was always real but whose economics never worked at venture scale.

video · ↗
A conference workshop and webinar, presented with @arsenysmoogly, on lessons learned from years of building and maintaining design systems.

article · ↗
Learning, experimenting, and iterating solutions based on understanding and mapping customer pain, not shipping outputs for their own sake.

article · ↗
"A lot of times, people don't know what they want until you show it to them" is an often-quoted piece of Steve Jobs wisdom that has given license to some bad thinking. What it actually meant.
Highlights
Smaller projects, experiments, and details from over the years. Select an image or video to explore the project.
Swipe to explore