Meet the Team: Ferruccio Balestreri and Enzo Avigo

Enzo Avigo and Ferrucci Balestreri had a simple query that they couldn’t answer with SQL: “What if behavioral data were easy to use?” That question drove them to found their startup June, where they worked to pioneer a simpler way to understand user behavior. And now, it brings them to Amplitude.

Read on for insights from their journey, their takes on AI and product development, and what they think the future holds for even simpler analytics.

Your mission at June, which you’re now continuing at Amplitude, was to make it easier to understand how people use products. What first got you excited about that?

Enzo: There’s what people say, and what they actually do. That’s true in real life and just as true online. People often say one thing but do another, sometimes just to be nice, or because of recency bias, or for many other reasons. That’s why behavioral data is such a powerful input when building products. It’s not always the most important signal, but it’s one you can always rely on.

Before starting June, I worked as a product manager. I always loved using behavioral data to guide decisions, but doing so meant learning SQL or navigating complex analytics tools. Over time, I became the go-to person on my team to pull insights. That made me wonder: what if everyone could access that data without the pain? What if behavioral data was as easy to use as creating a wiki page in Notion?

That idea never left me. I’ve been obsessed with it ever since. In a world where more people understand how their products are actually used, they can make better decisions. And when better products get built, people end up living better lives. It really is that simple.

Ferruccio: When deciding what to focus on, I asked myself a simple question: who would I genuinely enjoy spending time listening and talking to? The answer was that I like engaging with creative people who are trying to make something new. I found the idea of being able to shape the tools your whole industry uses to figure out what to build next very exciting.Software has been one of the most dynamic fields of the past few decades, and it's thrilling to build products that help shape such a rapidly evolving industry. Now, especially with the rise of AI, it feels like having a front-row seat to innovation across every domain.

It’s kind of funny that making products easier to use can be so hard. What are some of your top tips or advice for that?

Ferruccio: The way I think about building products is about understanding the problem you’re trying to solve and trying out different solutions. For the first, there’s no replacement for talking to customers and observing what they’re doing. For coming up with solutions, once you know what problem you want to solve, a question I find is really helpful is, “Who spent the most money trying to figure this out?” Then copy that as a starting point!

The way anyone can get to really unique and effective solutions is by finding references across different fields and industries where people had to solve that same problem. Then use some analytics to figure out whether it worked or not :)

Enzo: “Simplicity is the ultimate sophistication.” I can’t remember who said it, but it stuck with me.

Building a simple product is a bit like teaching. A regular teacher gives you the slides and expects you to figure it out. A great teacher rewrites the whole curriculum to make it land. The same thing applies to product. To make something feel simple, you need a deep understanding of the product and the user.

The mistake people often make is assuming that simplicity means the same thing for everyone. It doesn’t. A technical user can work from a few lines of documentation. An office manager might need a fully no-code experience to get the same result. Just writing things clearly is not enough for everyone to succeed.

One of my favorite frameworks for this is the knowledge gap. It helps you look at where your users are today versus where you want them to be. Then you figure out the shortest path to help them close that gap. I wrote more about that here.

What do you think about the intersection of AI and analytics? How is it helping, or how do you use AI yourself for analytics or product development?

Enzo: Analytics has gone through a huge simplification curve over the years. It started with SQL, then moved to visual tools that helped non-technical people run basic reports like retention or active users. Later, templates made it easier to reuse proven setups. Now with AI, you can ask questions in plain English and get results instantly.

AI is breaking the tradeoff between complexity and time. It can serve long-tail use cases that teams would never have the time or resources to handle manually. But most analytics is still reactive. You need to be in the mindset to ask a question, open the tool, and look for the answer.

What excites me is the shift toward proactive analytics. AI can help surface the right insight at the right time, with the right context. Imagine something like Apple’s suggestions for which app to open next. Or something even better, like having a personal analyst that works for you 24/7 in the background, always one step ahead.

Ferruccio: For product development, AI has been really helpful. We’ve been early adopters of Cursor and Lovable, and to be honest, I think I’m never going back to a non-AI-assisted coding editor.

For analytics and AI, I think there aren’t yet any applications that are truly game-changing. Hopefully, over the next year, we can change that! I’m very fascinated by the ideas around self-improving products that we’re working on at Amplitude, and I’m excited to see how far we can get.

What should customer-centric teams watch out for with AI?

Ferruccio: I think the biggest challenge for customer-centric teams using AI is ensuring they’re actually solving problems customers deeply care about. Many companies rush to build AI features that don’t genuinely improve the user’s experience or create lasting value.

The real magic happens when AI removes work, reduces friction, and makes life easier. Not adding complexity or demanding extra effort from users. Staying humble, rigorously testing assumptions, and continuously validating the actual benefit to customers are essential to making AI genuinely valuable.

Enzo: AI makes it tempting to treat everyone the same. You might start adding dynamic fields or templates that look customized, but in reality, just pull public data and insert it into a message. It feels personal, but there’s no real value or resonance for the person on the receiving end.

A good example is dynamic fields in sales emails. Many teams tried this and saw no improvement in conversion rates. In some cases, it made things worse. You think you’re personalizing, but you’re really just automating a fake version of it.

AI actually gives us a chance to do the opposite. We can create experiences that are deeply tailored to each person’s needs and preferences. But that only happens if we stay problem-centered in how we apply AI. Whether it’s a workflow or a product decision, the key is to start with what the person really needs and use AI to support that, not to replace it.

What sealed the deal for you about joining up with Amplitude? What are you looking forward to here?

Enzo: It was the people and the shared vision. Every conversation felt aligned. We were all pulling in the same direction and imagining a similar future.

But what really sealed it was meeting the team. The conversations felt natural. There was a real sense that we could be in the trenches together for the next few years. That made the decision easy.

Ferruccio: I think the conversations with the team were the most important thing for me. I was very impressed with everyone I met, and everyone was genuinely excited about working in the office again.