Most marketplace founders start from the same place.
You’ve got inventory – listings, products, profiles – and you assume the SEO work is to get all of it indexed and then out-publish whoever’s ranking above you. I ran that playbook for clients for the better part of twenty years. On a marketplace it usually buys you the same thing: a pile of thin, near-identical pages that go stale the second an item sells, and rankings you’re basically renting on stock you don’t even own.
So when I started building my own thing, I had to get honest about which marketplace SEO strategy I actually trusted. Not the version that’s easy to pitch in a deck. The one I’d put my own money behind. What I landed on caught me off guard: the smartest move was to stop playing the marketplace game the way everyone else plays it.
A bit about me
I’m Marcin. I run GrowthTurn, an SEO agency, and I’ve spent roughly two decades watching search strategies win and lose across all sorts of industries. Almost all of that was for other people’s businesses.
At some point I wanted to point it at something of my own. There’s a particular kind of honesty that only turns up when it’s your money, your roadmap, and your own late nights on the line, with no client to hand the blame to if you’re wrong. So treat this less as a case study I’m reporting on and more as one I’m still in the middle of, out in the open, with the messy bits on show.
What we’re building
The product is Carseto, a platform for classic-car collectors in Europe. The quick description would be “search classic cars across Europe in one place,” and that’s how most people bump into it for the first time. But the search is the front door. It isn’t the house.
The actual thing we’re building is a home base – somewhere a collector keeps their collection, logs each car, keeps an eye on what it’s worth, and shows it off to other collectors if they feel like it. We call it the Garage. I keep coming back to the same handful of products when I think about it. Strava is, technically, GPS tracking, and yet nobody thinks of it that way; it’s where runners and cyclists go to be runners and cyclists. Letterboxd can tell you a film’s runtime, but that’s plainly not why anyone’s there – it’s where film obsessives keep a record of themselves.
That’s not a branding point. It quietly sets the whole SEO strategy, because a search engine and a home base rank, defend themselves, and grow in completely different ways.
Search is the front door, not the house
This is the part that’s actually useful if you run a marketplace yourself.
Search is how you get people through the door, not the reason they stay. What pulls a stranger in is something useful, right away, no strings attached: someone wants to know what their car is worth, or wants to see what’s for sale across borders without trawling through fifteen national sites. That’s the job our cross-border classic-cars search across 44 countries does. It earns the first visit and hands people something before they’ve signed up for anything. But I don’t kid myself that it’s what makes us hard to copy. It’s bait, not a moat.
The rankings that last on a marketplace almost never sit on listing pages, and this is the one most marketplace owners work out too late. Individual listings are thin, they repeat each other, and they vanish the moment the car’s gone, so you spend your life rebuilding the same positions on sand. The pages that actually hold up over years are the aggregate ones: category, make, model, the reference pages a listing feeds into rather than the listing itself. If you’re going to pour effort in anywhere, pour it in there.
Restraint counts for more than people expect. A marketplace with filters can spin up a near-endless number of URL combinations, and the path of least resistance is to let Google crawl the lot. That’s exactly how you bury your good pages under a heap of half-empty ones and burn crawl budget on nothing. Being fussy about what genuinely deserves to be indexed, and having the nerve to keep the rest out, tends to move the needle on a big site more than another month of publishing. I’m not going to spell out our own rules here, but the gist is that a smaller set of strong pages will out-rank a bigger set of weak ones.
The bit I’d underline hardest: own your data instead of renting content. The most durable SEO asset a marketplace can have is data nobody else has. Numbeo gets cited because it sits on numbers you can’t get anywhere else. A blog post can be lifted by lunchtime; a dataset only you hold keeps earning links and citations long after you’ve published it. We’re building our own data for that reason. The specifics stay ours, but the logic holds for any marketplace sitting on information its competitors can’t reach.
Stack those up and it hangs together on its own. Search is what brings people in; the network and the data are what keep them there and make the whole thing a pain to clone.
The hard parts
I’ll be straight with you – none of this is as neat in real life as it reads on the page.
The discipline I struggle with most is just not building. Founders and SEO people tend to share the same flaw, which is that we love shipping things. Hand me a roadmap and I’ll happily spend a year adding features and call it momentum. So I’ve given myself a blunt rule: no new feature work until the product has shown real, paying demand. Get the distribution right first, be clever later. I don’t enjoy the rule, which is probably the point of it.
Language is the other one. “Classic car” doesn’t reduce to a single keyword. It’s oldtimer in German, auto d’epoca in Italian, klasyki in Polish, and every market searches differently and even argues about what counts as a classic in the first place. So a pan-European play isn’t really one SEO project translated six ways; it’s closer to six overlapping projects that happen to run on the same code. I started with Poland, partly because it’s home and partly because I know that market in my bones in a way I don’t yet know the rest.
And then the thing I honestly can’t call yet: whether a community actually forms. Search visibility you can more or less engineer. Belonging you can’t. The best you can do is set the table and wait to see who sits down. That part isn’t a growth tactic. It’s a bet, and the jury’s still out.
What I’d carry into any marketplace
If I had to boil it down to one thing for any marketplace SEO strategy, it’s that the sequence matters. Work out the positioning, then the architecture, then the content, in that order. Decide what you’re really building – a search engine, a marketplace, or a place people come back to – before you optimise a single page, because that one call quietly decides which pages were ever worth ranking.
That’s the sort of thinking we bring to marketplaces and big content sites at GrowthTurn: the strategy sitting above the tactics. If you’re chewing on the same questions for your own platform, that’s a conversation we enjoy.
And if you want to watch how it actually plays out, we’re building Carseto in the open, warts and all.