Introduction

When I graduated college and for the first several years of my career, I had no professional direction. I was taking any opportunity I could find to gain experience. My only real experience at the time had been writing for my friend’s movie blog, doing IT support in the business school computer lab, and a brief internship building Facebook communities and measuring social media data for a startup that ultimately didn’t succeed.

With that little bit of experience and an Economics degree, I caught on as a marketing data analyst for an HR Tech startup and recruiting company. From there, I became a data engineer, and eventually a data product manager, launching easily the most fulfilling and exciting phase of my career thus far.

Each of those combined experiences taught me lessons that have made me a better product manager, but none have helped me launch successful products more than these three:

1. Treat Analytics Like Part of the Product

My career has always been connected to data, and I couldn’t have asked for a better introduction to what would become my style of product management. Your gut and instincts will always play a role in decision-making, but pairing them with hard data is what gets results.

Building an effective analytics feedback loop requires several key steps:

• Find out what you can measure today.: Start small—what’s already being tracked, even if it’s not perfect.
• Create a meaningful set of metrics.: Boil down the noise to focus on numbers that matter to your product’s success.
• Push for better measurement.: Advocate for tools, processes, or resources to improve what you track.
• Refine and expand your metrics.: As your product grows, so should the depth of your insights.

Just as you wouldn’t stop improving your product after launch, your analytics deserve the same iterative energy. Begin with fundamental metrics such as:

• Average Session Duration: How long are users engaging with your product?
• Retention Rates: Are users coming back consistently?
• Monthly and Daily Active Users: How many people are actively engaging with your product?

The data doesn’t have to be perfect—it just needs to move you forward. Make decisions with it, and then improve your ability to measure and understand as you go. Analytics aren’t an afterthought; they’re an integral part of your product’s DNA.

2. Technical Execution is Crucial

Once you know what to build, you still have to make it happen. And here’s the hard truth: execution is where good ideas go to die. Simplicity is your best friend—build a small foundation, then layer on complexity as you go.

Too many projects stall because they chase perfection instead of progress. You don’t need to reinvent the wheel to get started; you just need to get the car moving. Here’s how:

• Start with POCs: Get a bare-bones version of your idea up and running. Is it viable? Does it solve even a fraction of the problem?
• Get feedback early: Don’t hide your work behind endless iterations. The sooner you can test with users, the sooner you’ll know what to fix or scrap.
• Cut complexity: Every feature, every piece of code, and every dependency introduces risk. Ask yourself: Does this actually need to be here right now?

Let me share an example from building my first generative AI feature. I partnered with my data team to create a prototype that would generate personalized career insights. We started with basic datasets and rough outputs, prioritizing speed over perfection.

Initially, the results were mixed—some insights were remarkably accurate while others missed the mark. Through rapid iteration, we refined the models, improved the prompts, and enhanced our data sources. The feature evolved from its rough beginnings to become our most engaged-with offering, driving user retention and inspiring future innovations.

Execution isn’t about perfection; it’s about progress. Build momentum. Create small wins. Because at the end of the day, nobody cares about what you almost launched.

3. Ship It

While execution focuses on how you build, shipping is about getting your product into users’ hands. The key distinction? You can’t wait for perfect—your creation needs to get out into the world and start providing value.

When I say “ship it,” I mean taking a strategic approach to releases while maintaining momentum. Here’s what I’ve learned:

• Build incremental versions. You don’t need a masterpiece on day one. A scrappy, working version is better than an idea that never sees the light of day.
• Accept imperfection. Some of my best work started as mediocre ideas I wouldn’t even want my name attached to. That’s the process—build, fix, build again.
• Focus on validation. Every release should answer a question: Did we solve a user problem? Did we hit the metric we were targeting?

For instance, when launching a new feature dashboard, we started with just three core metrics instead of the twenty we’d planned. This allowed us to validate the basic concept and gather user feedback before investing in a more comprehensive solution.

Conclusion: Build Something

Looking back, my career was anything but a straight line. I wasn’t someone who woke up one day and decided to be a product manager. But every odd job, every project, and every failure taught me something that I bring to the table today.

That’s what product management is at its core: you learn by building. You build analytics loops to measure your success. You build proofs of concept to test ideas. You build products and ship them, even when they’re not perfect.

So, if you’re waiting for the perfect role or the perfect opportunity to jump into product management, don’t. Start where you are. Build something. Learn from it. Iterate on it. Because no one cares where you started—they care about what you’ve shipped and what you’re capable of building next.

Now go, and make something happen.