The Link Scarcity Thesis

Every agency you know quietly moved its link budget into content this year. Ask around and you’ll hear the same three things: outreach is dead, DR is a vanity metric, and nobody answers the emails anymore. So you’re probably wondering why you’d spend another dollar on links at all.

Well, the truth is that the retreat is the exact reason to stay, and I can show you where that comes from rather than just asserting it.

I went back through every interview in the Unscripted SEO archive that has a transcript, looking for what SEOs actually agree on, and link building came up more than anything else. Two guests, working separately and with completely separate datasets, said the same strange thing, and neither of them framed it as a headline.

Yeah, backlink correlation dropped. That’s not the interesting part.

Patrick Stox has Ahrefs-scale data behind him, and he doesn’t soften it.

The correlation with rankings went down overall compared to what we’ve seen traditionally. It was still high for certain things like local companies. But for like local SEO they still mattered.

Patrick Stox

You already believe that part, so here is what follows from it, and this is the bit almost nobody says out loud.

The general trend I see in the industry is people are kind of going away from the links. And that means it’s actually probably gonna be more of a differentiator in the future. Now that they’re more rare, the links actually matter more.

Patrick Stox

The retreat is the mechanism here: links get more valuable because fewer people are building them, which means every competitor of yours who quietly moved that budget into content this year has just made your remaining links worth more. You didn’t do a thing to earn that, they handed it to you.

A second dataset ties backlinks to AI search visibility

Ben Wills ran 100 industries with 10 personas each. That’s 1,100 personas, then LLM prompts, keyword sets, an archive of Reddit, and Wikidata correlations on top.

There was a pretty decent correlation between your backlink strength, your search ranking strength, and your likelihood to show up in the [LLM] results.

Ben Wills, OppAlerts

He puts the caveat on it himself before I could, which is why I trust the number.

This is correlation, not causality, this is what it means.

Ben Wills

Two people, two methods, and the same direction, which is about as close to a finding as this industry ever gets.

You can hear cold outreach link building dying

Seven guests told me templated cold outreach is finished. Christopher Gimmer explained what actually replaced it.

It’s hard to do kind of traditional link building now just because everyone’s inboxes are full of spam. And so I think people are more willing and likely to link out to a really good tool that actually solves a problem versus yet another blog post with a wall of text.

Christopher Gimmer

Rob Bonham put the cost plainly.

Typically, good link building is not very easy. It’s tedious, it requires a lot of outreach.

Rob Bonham

Tedious and expensive is why people quit, and that is exactly why Stox thinks the ones who don’t quit will get paid for it.

Quit leading with DR, screen for link relevance instead

Seven guests independently said third party metrics are the wrong first filter. Bradley Benner was the bluntest about it.

It’s not some stupid third party metric like DA or DR or trust flow or anything else that matters about whether a link is valuable or not. What matters is whether it’s relevant.

Bradley Benner

What they screen for instead, in their own words:

• Kristiyan Yankov looks at the site and the actual page before anything else.
• Jeremy Moser asks whether there’s a real company behind it.
• Ben Senescu asks the question that should end most sales calls: we’ll buy you all these backlinks to increase your domain rating. But is that going to get you any organic traffic?

One guest disagrees, and I’m not going to hide him. Matthew Slaymaker uses domain authority as exactly the weighting proxy the others reject.

I got you a thousand links from Domain Authority one out of a hundred websites. None of that is gonna carry the weight of one link from a 50 out of 100 domain authority website, right?

Matthew Slaymaker

So there is one dissenter in the set, which is worth remembering before you go telling a client the question is settled.

Unlinked mentions are doing the link’s job now

This is the largest cluster in the archive, nine guests. Jason Barnard has been saying it longest.

Whoever 10 years ago was saying, don’t care about mentions, should be kicking themselves today.

Jason Barnard

Bonham says the same thing from the client side: unlinked mentions and visibility on third party publications matter more than ever. Ernesto Ortiz described a footprint he built that generated no links at all. It’s a massive footprint and it’s not a link. It’s just a mention.

Can toxic backlinks still hurt you? These two flat-out disagree

This is the sharpest contradiction I’ve got on tape, and both men do this for a living.

Even if I built now, like, a hundred links for a website from PBNs, I don’t think it’s going to do anything to that website, even negatively.

Kristiyan Yankov, who builds links for SaaS

There’s so many people, even without manual action penalties, that are sat in a partial penalty where they might be held back slightly, could be held back two or three positions across the whole site because they were above a toxicity threshold. And I never knew about it.

James Dooley, speaking to founding host Mark A Preston on Unscripted SEO

Stox doesn’t pick a side. He points out that both of them are arguing about a referee who already forgave everybody.

I think SEOs got lucky with Google. We got Penguin and stuff. And then what was it, 2016 ish? They’re just like, yeah, we forgive all your bad links. They could have held that over us for another like ten, fifteen years, easy. And these AI search systems, they may make that decision. How long will they hold a grudge against you?

Patrick Stox

The amnesty was a gift, not a law. The new systems never promised you one. If you’re building on the assumption that bad links wash out in a couple of years, you’re betting on a policy that exactly one company ever had, and you’re betting it gets renewed by companies that never signed it.

They bought 66 backlinks and nothing moved

Ash Nallawalla is the counterweight to all of this, and I’m keeping him in because the piece is dishonest without him. A senior exec at a large Australian brand insisted competitors had to be buying links. So they bought some.

Roughly six articles times eleven, so around 66 links, a five-figure investment. One month passed, two months, six months, absolutely no change to any ranking.

Ash Nallawalla, speaking to founding host Mark A Preston on Unscripted SEO

His conclusion: for a large company, buying links doesn’t move the dial much. Nobody publishes the campaigns that did nothing, so you should weight this one higher than the case studies that worked.

The $50 PBN that games ChatGPT in a week

Ernesto Ortiz says 100 listicle articles and a PBN will get a brand named as the best inside ChatGPT in under a week, for about fifty dollars, and he also says he won’t do it. Both halves of that matter, because the tactic works right now and it’s priced like a rounding error, which tells you roughly how long it will keep working.

Your link building strategy, in five moves

Move	Because
Screen the page, not the DR	Relevance is the filter that survived
Count mentions as output	Largest agreement in the archive
Kill the templated sequence	Inboxes are full, reply rates are gone
Ship a tool, not a post	People link to things that solve a problem
Assume no amnesty	Only Google ever granted one

It all reduces to one bet: if everyone else is quitting, and the thing still correlates with getting cited by the models, then the right move is to be one of the few people still doing it properly. Stox is right that it becomes a differentiator. It only becomes a differentiator for whoever stays.

That’s the bet I’ve made with my own time, and these are real conversations where the links get earned on tape rather than bought in a spreadsheet. If you want that run for you, that’s podcast-powered link building.

So before you cut the line item, ask yourself three things:

• Who in your space is still building links, and what does it cost you if they’re the only ones?
• When a model picks three sources to cite for your money query, what makes yours one of them?
• If the grudge never gets forgiven this time, which of your links are you going to wish you hadn’t bought?

Everybody else is walking off the field. Go build some links.