Between clients for a month. Been buying links since 2014, ask me anything

You're right about the mechanism and I still think the proxy is better advice.

I've watched a lot of people be told "check the footprint before you buy" and I've watched approximately none of them do it after the third purchase, because by then the guy is answering within an hour and the links are going up and checking feels like paranoia. A rule you follow when you're bored beats a diagnostic you skip when you're busy.

Also — and this is the part I only worked out afterwards — I did sample his sites. In month two. They looked unrelated, because in month two he was selling me his good inventory.

That last paragraph is the strongest thing in the thread and it isn't in your original answer. Sampling at the start of a relationship measures a curated set, by definition.

Every supplier's first delivery is a portfolio piece. Ask a photographer.

Another one if you're still around — how do you decide a site is worth five times what a similar site costs? I keep seeing two sites with the same traffic where one is $200 and one is $1,000 and I can't tell what I'd be paying for.

Usually you're paying for the fact that the expensive one doesn't need your money.

That sounds glib but it's most of it. A site that sells links reluctantly, at a price designed to make you go away, is a site that has other revenue, which means it has readers, which means the link sits in something a person will read. The $200 one has a rate card because links are the business model.

Whether that's worth 5x depends entirely on whether you're building for six months or six years. In 2014 I'd have said always take the cheap one and buy five. I've changed my mind and I can't prove I'm right.

Are the guest posts you buy written by AI now? Does it matter for whether the link works?

Most of them, yes, including some I paid a premium for on the understanding that they weren't.

Does it matter for the link? Not in any way I can measure. The link is a link. What it changes is the site — a site that publishes twelve machine-written guest posts a month stops being a site anyone reads within about a year, and then the link is on a dead property. So it doesn't damage your link, it shortens its useful life, and those look identical for the first eight months and completely different at month twenty.

This is the same shape as the article directory collapse in 2011 and 2012, incidentally. Nothing was penalised on day one. The inventory just quietly became worthless while everyone was still buying it.

and everyone will keep buying it until the invoices stop, same as last time

Bookmarked. The shortened-life framing helps a lot, thank you.

$2,000, one site, no existing links, six month horizon. What do you actually buy?

New client starts next week so this might be the last long one for a while.

With two grand and six months I'd buy eight links and I'd spend more than half the budget on two of them.

• Two at $600 to $700 each. Real sites, real readers, in or adjacent to the niche, bought direct from the owner after a conversation rather than off a list. These are the ones that have to still be there in three years.
• Four at around $150. Ordinary small sites with a side hustle, the kind that answer emails. Nothing clever.
• Two at nothing, from suppliers I already have a relationship with, thrown in because I'm buying the four.

What I would not do is what I'd have done in 2015, which is buy forty links at $50 and let volume do the work. That worked, genuinely, for years. It stopped working somewhere around the 2022 link spam update — not because the cheap links started hurting, but because they stopped counting at all, and forty things that don't count cost the same as two things that do.

The part people get wrong with a small budget is spreading it evenly. Eight identical $250 links is the worst version of this. You want a barbell: a couple of things good enough that you'd mention them to a client, and a tail of cheap stuff so the profile doesn't look like four links appeared out of nowhere and nothing else.

And hold back $300. Something will come up in month four — a site you've wanted for two years will suddenly answer, and if the budget's gone you'll spend the next year thinking about it.

Different angle if you have time: how do you report bought links to a client who has been told by their legal team that they can't buy links?

I've had this twice and handled it badly both times. Curious what the honest version looks like from someone who's done it for a decade.

Question about non english. All your experience is english market i think. Is the same rules for russian or polish serp or is different game?

Different game and I'm not qualified to answer it properly. I've run two Polish campaigns and both were for English-speaking clients targeting Poland, which is a third thing again.

What I noticed, for what little it's worth: links did more, faster, and cheaper, and the sites were happier to sell. That's what English looked like in about 2016. Whether it converges or whether it's structurally different because the SERPs are thinner, I genuinely don't know.

Removed two posts advertising a link package. Thread stays open, keep the sales pitches out of it.