Lessons from building an e-commerce beast

SAPO

2015

SAPO Biz was an online store builder, a Portuguese take on Shopify. It started with two of us, Bruno Carreira, an engineer, and me, the designer. A small team joined along the way, and it took about nine months. It launched in July 2015, a month after SAPO got a new owner, and it never got the chance to grow. It's the project that taught me how much product management matters, and why you get a first, smaller version out the door before you build everything.

SAPO before it was just a media company

SAPO was born in 1995, when six engineering students at the University of Aveiro built a search engine for Portuguese websites, three years before Google. In 1999, Portugal Telecom bought it. By 2014 it was the most visited website in Portugal, and inside a big telecom company, it was still run by a bunch of geeks. Email, maps, blogs, videos, Codebits, a yearly tech conference for other geeks: what mattered most were the technological challenges.

Half of SAPO was building crazy new things, or its own versions of crazy big things that already existed elsewhere. The other half was media, which is where most, if not all, of the money came from.

Design worked differently too. Photoshop was still how we communicated: you needed to see something before you could start a conversation. But then we built it ourselves, in HTML and CSS. There was no handoff to front-end engineers. There wasn't even a clear line between front end and back end.

Trust takes time

I joined SAPO in 2008. For my first four years I worked mostly on products that had to run everywhere: mobile web, iOS, Android and TV apps. In 2009, I started working with Bruno Carreira, a lead engineer. We got along so well, and worked so well together, that we started doing lots of things as a pair, and people knew it and kept putting us together.

From 2012, a lot of that was e-commerce. First smaller projects, like a restaurant vouchers website. Then Ticketline, Portugal's leading ticketing site, which was a much bigger one.

Along the way we pitched lots of ideas for e-commerce products to Celso Martinho, SAPO's co-founder and, at the time, its leader. We built proofs of concept to show we could do them. None of them became a product. What stuck with him, I think, was that we had initiative.

Looking back, that's how trust got built.

"We think you two should take it"

Then, in 2014, Celso pulled Bruno and me into a room. "We have this challenge. We want to take this on, and we think you two should take it."

The challenge: an online store builder. A designer and an engineer, a whole product, seven months.

Why? SAPO saw a gap. The latest numbers at the time showed that only 1 in 4 people in Portugal had bought anything online in the past year. Across the EU, it was almost half.

And SAPO had something Shopify didn't: everyone in Portugal knew its name. Tech people would pick Shopify, sure. But if you ran a small shop and had never sold online, wouldn't you trust a name you already knew? At least, that was our assumption at the time.

We said "let's do this". Today, that brief looks wonderfully naive to me. Back then it felt natural. Of course we're going to build this. We're builders.

Where do you even start?

If you're going to copy, copy from the best. So we looked at two ends. Shopify was the full beast. Tictail was the opposite: fewer features, a store in five minutes, much more pragmatic (Shopify ended up buying it in 2018). In between, we tried Squarespace, which was mostly a website builder at the time and was just turning to e-commerce. When I tried it, it felt like a good balance. We borrowed what we could from all three. By then it wasn't just Bruno and me anymore: about five engineers had joined, and there was still a whole product to build. The business side, the plans and the free plan, wasn't ours to decide. Almost everything about the product was.

I knew very little about the plumbing of e-commerce

Payments, shipping, taxes: the plumbing of e-commerce. I knew very little about any of it. How do you help a small shop owner set up shipping prices and VAT, when you barely understand either yourself? We had to learn fast.

Other parts were hard but familiar. Creating a product, with variants like size and colour, gets complicated quickly, because every option multiplies. That page I copied on purpose. It was too complex to reinvent under a deadline, so I took Shopify's approach and rebuilt it with our UI.

Most of the people we were building for had never sold online either. So the admin explained itself as it went: a welcome checklist on the first visit, and small links like "What are product attributes and variations?" next to the hard parts.

It was also my first real war room. For months the whole team left our usual desks and sat together, so we could move faster and have the hard conversations right there, face to face. Around month six, we knew seven months wasn't going to happen. We launched in about nine.

A launch, and a new owner

One of my strongest memories is presenting SAPO Biz to Armando Almeida, who ran PT Portugal for about ten months. I was nervous. But I had helped build it from the ground up, so I could demo it from A to Z.

The launch itself had a government minister. It all felt a bit weird. SAPO was a big company, sure, but why was a minister here, helping launch an online store builder? Maybe there were government funds involved. I never found out.

What I didn't see coming was the timing. In June 2015, Altice finished buying Portugal Telecom. Altice saw SAPO as a media brand, not as a place to build software, and stopped funding projects like ours. SAPO Biz launched in July. After that, there was no plan to maintain it or add anything new. It would quietly die.

What I'd do differently

It felt really good to deliver a beast that big. It wasn't a waste of time: it was a challenge I took on, and it simply didn't go any further. But for a product that never got the chance to help many people, it was probably a waste of resources.

Part of that wasn't ours to control. Part of it was. We had no product manager, nobody whose job was to stop us and ask the tough questions. We weren't thinking about a first small version. We were executing, we were on a mission. If we had launched sooner with everything free, and no plans to design at all, we'd have had real shops using it while we did the harder work. Maybe that would have given it a better chance. Maybe not. But we'd have learned more, sooner.

It also taught me something about myself. I can take on hard, messy things and finish them. I deal well with chaos, and I can help a team get things done even when it feels like we've been left alone in a war room.

Only years later did I hear about slicing an initiative: shipping it in thin pieces, each one useful on its own, and learning from each before building the next. Oh, what a lesson that was.

©

2026

Luís Carmona. All rights reserved.

Built using Framer, Claude and some tricks up my sleeve.
Brush your teeth at least twice a day.

©

2026

Luís Carmona. All rights reserved.

Built using Framer, Claude and some tricks up my sleeve.
Brush your teeth at least twice a day.