A structured approach to product improvement

I’ve worked across different product teams and environments, and I’ve seen many ways of building and improving digital products. What I’m sharing here comes from that experience, a structured way to improve products and features.

I believe nothing works in isolation. So before making a change, the real question is not how to do it, but how that change impacts the entire system.

Start with the purpose of the page

Let’s talk about a digital product. Let’s say you want to improve the user experience of some pages. Every page of a product, every interaction, and every piece of information has a purpose. Before diving into data, ask yourself: “What is the purpose of this page?”

The goal of a checkout page is to convert intent into a completed purchase, to make it as easy, fast, and trustworthy as possible for users to finalize their transaction. The goal of a product page is to turn curiosity into action.

This is important to define first, because you can’t measure progress without purpose.

That’s why you start with a clear goal and then define your KPIs based on it. KPIs to Monitor (for a checkout page, for example): Checkout completion rate (conversions / sessions); Cart abandonment rate; and so on..

Use data to understand what works and what doesn’t

Run a full UX audit, use analytical tools such as heatmaps, session recordings to observe real user interactions under real conditions, not assumptions or opinions. Take notes on insights like scroll depth, time spent on page, bounce rate, technical performance, and traffic sources.

Understanding these behaviors helps you connect what users do with the purpose of the page.

Benchmark

Now, let’s talk about benchmarks. Once you’ve gathered your internal data and understand how users interact with your product, compare it first against your past performance, then against best-practice thresholds or competitors in your space. That’s how real progress happens. Conduct a comparative analysis of your competitors. Study what they do well, where they fall short, and what opportunities exist for differentiation.

Once you understand the external context, re-evaluate every key page of your product through the lens of data and benchmarks. This is how design becomes strategic and not opinion-driven. This approach ensures that each design choice is backed by evidence, not assumptions.

Turn data into actionable improvements

Here’s the main piece: insight without action and prioritization is noise. Improvement starts by translating insights into designs that actually work. Once you gather observations, categorize them into: Quick wins; Usability issues; Strategic improvement. This helps you focus on the highest-impact changes first.

And here’s the part I care about most..

When you focus on improving something small, don’t just look at the surface. Look at how it connects to the entire system. The same framework you use to improve one small thing can often be reused everywhere else.

The goal is to maximize output with minimum input. At that point, you’re no longer improving a feature you’re improving the system that creates improvement. And once that system works, all you have to do is repeat it.