What Does a Spanish Fiat Have to Do with Robotics? More Than You'd Think

SEAT built Fiats under license for decades before eventually getting absorbed into Volkswagen's empire. That licensing relationship is interesting. They weren't just assembling kits. They were adapting, modifying, sometimes improving. The four-door 800 wasn't just a Fiat with extra doors bolted on. It required structural changes, different weight distribution, new interior layouts.

We see this in robotics all the time now, though we call it "integration" or "system customisation." A company licenses a robot arm from one manufacturer, a vision system from another, grippers from a third, and then some integrator (or their own engineering team) has to make it all work together. Add another door.

The difference is that SEAT was transparent about it. Everyone knew they were building Fiats. In robotics, there's this tendency to obscure the lineage. Startups especially love to claim they've built something entirely novel when really they've taken existing components and, well, added more doors.

I'm not saying that's bad. Actually, I think it's underrated. The skill isn't always in inventing something new. Sometimes it's in seeing that an existing design can serve a different purpose if you just modify it intelligently.

What We've Lost

Here's where I get a bit grumpy, so bear with me.

Post-war Spain needed SEAT because they needed to rebuild, and they needed to do it fast and with limited resources. The licensing model made sense. You take proven technology, adapt it to local conditions, and get moving. There was no shame in it.

Somewhere along the way, the robotics industry decided that everything has to be "revolutionary" or "groundbreaking" or whatever buzzword the marketing department cooked up that quarter. We've lost the appreciation for solid, unglamorous adaptation work.

I worked on a project in 2014 (I think, maybe 2015) where we essentially took a palletising cell designed for one customer and spent six months adapting it for a completely different application. Different products, different throughput requirements, different safety constraints. The customer wanted to call it a "next-generation system." It wasn't. It was a really good adaptation. And it worked beautifully for eight years until they finally upgraded.

That's the SEAT 800 approach. Take something proven, understand it deeply, and modify it for the actual need.

The Volkswagen Question

SEAT eventually became part of Volkswagen. They stopped building licensed Fiats and started building cars on VW platforms. Whether that was good or bad for SEAT specifically, I don't know enough to say.

But it raises a question I think about regarding consolidation in robotics. When the big players absorb the smaller ones (and they always do, eventually), do we lose something? The smaller companies often had that SEAT mentality, that scrappy adaptation capability. The big players tend to want everything standardised, everything on their platform.

It's too early to say how the current wave of acquisitions will shake out. We've seen some good ones and some disasters. I called my old colleague at ABB last month and he's... let's say cautiously pessimistic about how some of the integration efforts are going. But that's probably a topic for another day.

The Point, If There Is One

Look, I'm not suggesting we all go study the history of Spanish automotive manufacturing. But I do think there's value in remembering that adaptation, intelligent modification of existing technology, is a legitimate engineering discipline. It's not just what you do when you can't afford to innovate.

Sometimes adding another door is exactly the right solution. The trick is knowing when.

James Chen · 25 Jun · 5 min