A few weeks into working at Watchfinder, the thing that has struck me most about search in this category is how many of the normal ecommerce SEO assumptions don’t quite hold. Most of what I know about optimising an online shop assumes you sell the same product many times over. In pre-owned watches, nearly every item is a one-off. That changes more than you’d think.
I’m still learning the category, so treat this as a list of the problems I can see and how I’d think about each of them in general terms, rather than anything about how we do it.
People search by reference number
Watch buyers are a knowledgeable bunch. Alongside the searches you’d expect, brand plus model, there’s a huge amount of search for specific reference numbers: the code that identifies an exact version of a model, down to the dial colour or case size. Someone typing a reference number into Google knows precisely what they want, and that’s about as close to purchase intent as search gets.
The implication is that reference numbers need to be visible and consistent on the page, in titles and headings, and written the way people actually type them. Watch references come with dots, dashes, letter suffixes and spaces, and enthusiasts don’t type them consistently. I’d look at the search data to see which variants people actually use, rather than assuming the manufacturer’s official formatting is the one that matters.
The stock disappears
This is the big one. A normal shop has a product page that lives for years, gathering links and history. A pre-owned watch listing exists until the watch sells, which might be days. Then what?
The options, roughly:
- Delete it and let it 404. Clean, but you throw away whatever that page had earned, and anyone arriving from an old link hits a dead end.
- Redirect it. To the model page, the brand page, or the closest equivalent in stock. Sensible if the redirect target is genuinely relevant; less so if everything just bounces to the home page.
- Keep it live, marked as sold. Useful for people researching prices, and it keeps the page’s value. The risk is a site full of things you can’t buy, which frustrates visitors and dilutes the site if it’s done at scale.
I don’t think there’s one right answer. It depends on how much search value an individual listing ever builds up, which for a page that lives for a week usually isn’t much. The more durable win is probably at the level above: strong, permanent pages for each model and reference, which always exist regardless of what’s in stock that day, and which the individual listings feed into.
A crowded field
The search results for watch terms are busy. There are the brands themselves, authorised dealers, other pre-owned specialists, auction houses, marketplaces, forums and an army of independent sellers, plus a fair amount of grey-market activity at the edges. The brands are protective of their names, sensibly, and you have to be careful how you use trademarks in both organic and paid search.
Forums deserve a special mention because they rank surprisingly well for detailed queries. Enthusiasts discuss individual references at length, and Google rewards that depth. It’s a reminder that thin listings on their own won’t compete for the more informational searches.
Content earns its place
Which brings me to content. In a category where buyers do weeks of research, there’s a strong case for publishing the material they’re looking for: buying guides, model histories, explanations of the differences between references, honest advice on condition and servicing. Done properly, that’s useful to the buyer, it builds the authority that a site full of short-lived listings struggles to build on its own, and it fits naturally with the trust questions I wrote about in When a Certificate Is Worth More Than a Campaign.
Done badly, it’s another pile of generic articles. The test I’d apply is whether a genuine enthusiast would learn something from it, and watch enthusiasts are a demanding audience.
Where this leaves me
Most of this is common sense once you see the shape of the problem. The main adjustment for me has been letting go of the idea that the product page is the centre of the SEO world. Here it’s the page most likely to be gone next week, so I expect most of my effort to go into the model and reference pages instead.