Guy in my inbox selling DR52 links, site gets 30 visits a month. Is that a real link?

$90 for a homepage link on a DR52. He's got a spreadsheet, forty of them, all DR40 plus.

Problem is I actually looked and the site gets about 30 visits a month. He says traffic doesn't matter for link juice, DR is the metric that counts, and honestly the DR52 does look huge next to what I'm paying elsewhere.

Is he right? Feels like there's a catch but I can't work out what it is!

DR is a number a tool vendor made up. That isn't a criticism of the tool, it's just what it is — a log-scaled estimate of how many links point at the domain, produced by a company whose crawler is not the one that decides your rankings.

Which means it is buildable. And once something is buildable and people will pay for the number, someone builds it. That's what a DR52 with 30 visitors is: it has been fed. Usually a couple of hundred links off other sites in the same seller's inventory, plus whatever's cheap that month, and the whole cluster props each other up. In 2013 the same guy was selling PageRank 4 blogroll links and it was the same conversation, right down to "traffic doesn't matter for link juice".

The catch you can't quite see is this: the only thing DR measures is the property the seller is manufacturing. It's a price tag, not a valuation.

Ask him for search traffic instead and watch what happens to the conversation.

Depends what you're running, but here's the two-minute version of the check, and I do this before I buy anything myself.

Pull the referring domains and look at who links to the linkers. If forty of them are sites you can tell were built by the same person — same three hosts, same age bracket, same faintly-off article titles — the DR is internal to that little economy and it doesn't leave the building.

Then look at when the traffic died. A site that had 4,000 visits a month until September and 30 now is a different purchase to one that never had any. The first one is a real site that got hit and might come back. The second was never a site.

At $90 with forty of them on a spreadsheet, I'd bet on the second.

Count the outbound links on the homepage. FWIW anything over about 25 and you're buying a slot on a billboard, not a link.

Also check whether the existing outbounds are all commercial anchors in five unrelated niches. Casino, insurance, a Turkish escort site and a plumber in Leeds is not an editorial homepage.

31 outbounds. One of them is a casino. So it's junk then?

It's junk. You worked it out yourself, which is the useful part.

Adding one thing about timing, because the last month makes this worse than usual.

A lot of the inventory being cold-mailed around right now is sites that were quietly earning until the September update and are now worth nothing to their owners, which is precisely why the spreadsheets have got longer since about the 20th of September. You are being offered the wreckage, at wreckage prices, described as a bargain.

I'd rather buy one link on a site with 3,000 real visits a month at three times the price than forty of these. I have not tested that as a controlled comparison and I'm not going to pretend I have — but every placement I can point at that moved something in the last two years was on a page a human being had a reason to be on.

Passed. Told him I wanted traffic screenshots and he stopped replying within the hour, which I think answers it.