Stop Building a Résumé, Start Shipping a Product: The Case for Treating Your Portfolio Like a Startup

The Résumé Trap

Here's an uncomfortable question: when's the last time you changed anything on your portfolio?

If you're like most developers I talk to, the answer is "when I changed jobs," which means roughly every two years. Your portfolio sits in a repo, unmeasured and unloved, built the way we build résumés — a static list of accomplishments, optimized for nobody, visited by even fewer people.

I did this too. The old version of aleksdelab.com was a handsome single page: my photo, a bio, six project cards, links to GitHub and LinkedIn. It looked professional. It was also completely inert. I had no idea who visited, what they clicked, or whether a single person ever made it past the hero section.

Then it hit me: I'd never treat a client's product this way. So why was my own site exempt?

Your Visitors Are Users

The mental shift is simple but radical: a portfolio has users, and those users have jobs to do. A recruiter is trying to figure out in 90 seconds whether you can do the work. A potential collaborator is trying to figure out if you're interesting. A fellow dev is trying to figure out if your writeup will save them time.

Once I started thinking of visitors as users, the résumé design fell apart on contact. Résumés answer "what have I done?" Products answer "what can you do here, and why should you care?" Those are different questions with different designs.

Instrument Everything

The first thing I did was add analytics — not creepy tracking, just honest measurement. Plausible, a few custom events, done in an afternoon.

What I found was humbling:

• 71% of visitors bounced on the hero section. My carefully written bio was a wall they never climbed.
• Project cards got a 9% click-through rate. Nine percent. I'd spent weeks choosing the perfect screenshot for each one.
• The single most-read page was a two-year-old blog post about a bug I'd fixed. Not the About page. Not the projects. A debugging story.

Before I had numbers, every design decision felt like taste. After, they felt like hypotheses. That's the whole difference between a résumé and a product.

A/B Testing Your Own Blurb

With instrumentation in place, I started running small experiments. Nothing fancy — my traffic doesn't support statistical significance, so I treat these as directional signals over weeks, not days.

Test 1: Project blurbs. The old cards led with technology ("A React + TypeScript dashboard for..."). I rewrote them to lead with outcome ("Cut invoice processing time by 60% for a small logistics team"). Click-through went from 9% to 17% over the following month. People don't click on stack lists; they click on results.

Test 2: The hero. I replaced my bio with one sentence about what the site is for: "Experiments, side projects, and notes from the dev lab." Bounce rate dropped eleven points. Turns out clarity beats credentials.

Test 3: CTAs. Adding a plain "Read the writeup →" link under each project card outperformed the "View on GitHub" button by nearly 2x. Recruiters skim; readers read. Serve both.

Iterate in Public

Here's the part that feels most contrarian: make the iteration itself content. Every experiment I run on this site becomes a lab note. The analytics findings, the failed tests, the embarrassing before screenshots — all of it gets published.

This does three things at once. It gives the site a pulse, so returning visitors see a living project instead of a digital taxidermy. It demonstrates the exact skills a portfolio is supposed to demonstrate — you can't fake a writeup about how you improved your own conversion rate. And it gives recruiters and hiring managers something most portfolios never have: evidence of how you think, not just what you shipped.

The old aleksdelab.com proved I could build things. The new one proves I can improve things. In any product role, the second skill is the one that gets you hired.

The Before and After

To make this concrete, here's the scoreboard from my own rebuild:

None of these numbers are going to impress a growth team at a Series B startup. But for a personal site that most devs never measure at all, an 11-point bounce drop and a 5x return rate feel like proof of concept.

Your Homework

You don't need a redesign. You need a loop. Add analytics this weekend. Look at one number that surprises you. Change one thing because of it. Write down what happened.

That's it. That's the whole methodology. Your portfolio isn't a monument to your career — it's a product, you're its product manager, and version 1.1 is due Friday.