Unscripted SEO. Host: Jeremy Rivera. Guest: Chris Panteli, co-founder of Linkifi, a done-for-you digital PR link building agency in the UK that earns tier one media coverage for its clients.
Listen to the episode
Also on Castos.
Recorded 20 August 2025; the full episode is audio only.
Ask ten SEOs whether a link needs to be topically relevant and you’ll get eleven answers. Chris Panteli sidesteps the whole fight. His agency, Linkifi, builds links the way a PR firm does: pitch journalists, land coverage, take whatever link the publication gives you. You don’t get to pick the page, the anchor or the topic. And he’s fine with that, because of what you get in return.
“There’s almost no chance that Google is going to devalue the BBC or a link from the BBC anytime soon. So it’s like a zero risk game, digital PR.”
Chris Panteli, Linkifi
That one line is the spine of the episode. Everything else, from the helpful content update to what ChatGPT cites, hangs off it.
Key takeaways
- After the helpful content update, the Linkifi clients that came through were the ones with a real brand behind the site.
- Digital PR is almost exclusively a homepage link. You give up control of topic and anchor, and you get branded authority that filters down to every other page.
- For one specific money page in a competitive niche, reverse engineer the competitors’ links to that page: count them, bucket them by type, check the anchor distribution.
- Google says it devalues bad links rather than punishing them. That cuts both ways, and cheap tier two links can still pay off for a while.
- Three questions set your link strategy: your tolerance for risk, what your industry is doing, and your budget.
- LLMs show their sources. Tier one media, YouTube transcripts, Reddit threads and listicles are what they cite, so you can see exactly which ones to get into.
The brands that survive have an army of fans
Jeremy opens with a complaint about his own industry: somewhere along the way SEOs became Google reverse engineers, chasing an authority score a tool invented, and forgot that the original idea in 1998 was simply authority moving through hyperlinks. Chris agrees and goes further. The value of earned media is bigger than the value of the link.
“We’re in the era of brand building and the brands will be the ones that survive. The brands that have an army of fans will be the ones that survive the longest.”
Chris Panteli, Linkifi
He saw it happen in his own client list. Post-HCU, Linkifi had clients “from every walk of life,” and in his words, “the ones that came through and are still around are the ones that had like a real brand.” Jeremy ties that to the click data that surfaced in the Google antitrust documents, and to anchor text: stop building links that say Nashville custom pool designer and start building ones that say Joe’s Pools. Chris’s read is that a trusted brand is simply easier for Google to recognize and feel confident surfacing, especially as it tries to lean less on links.
If you want the counterpoint on the disavow side of the same era, Cyrus Shepard ran the disavow test that did nothing on this show and landed in the same place on brand signals.
Why digital PR is a homepage game
Jeremy brings a question he’d been kicking around with Matt Brooks: does the topical relevance of the linking page actually matter to Google, or have SEOs over-engineered it? A news article about a city can mention a lawyer or a pool builder in passing. Is that link worth less?
Chris’s answer depends on which kind of link you’re building. “For digital PR, we’re all about homepage authority building. So digital PR is almost exclusively a homepage link.” You’re at the behest of the journalists, what they’re writing, the publications, so relevance is mostly out of your hands anyway. What you get is branded anchors pointing at the homepage, and authority that filters down to the rest of the site.
A money page is a different job. If you’re in iGaming, or you’re a local real estate business with one page that drives most of your leads, and a competitor outranks you on it, here’s the method he borrows from Matt Diggity:
- Pull the competitor’s links to that specific page, not to the domain.
- Compare counts. If you have zero and they have four, one or two might close the gap.
- Bucket each link by type. Guest post, exact match anchor, something else.
- Check the anchor text distribution before you copy any of it.
“Links is just a part of the algorithm,” he says, so you can’t prove any single link is why they rank. But it’s the part you can see. Digital PR sits alongside that work as one piece of a holistic strategy, not a replacement for it.
Penguin, link sludge and the zero-risk link
This is where the conversation gets candid. Jeremy argues that Penguin turned link building into a head game. Before it, a bad link profile got you slapped down and sent crawling back through the disavow tool. After it, Google stopped telling you anything and asked you to assume it was quietly devaluing the junk. He describes a real case of what he calls link sludge: an operator buying roughly 30 domains a month, spinning the same content across different WordPress themes and pointing bad anchor text at every competitor, except one. That one was the client paying for it.
Chris doesn’t pretend Google’s system is clean. If Google really does ignore bad links instead of punishing them, then you don’t need to panic about a negative SEO attack. But the same logic means someone can still buy thousands of cheap tier two links, point them at one money page with one anchor, and profit until it stops working. He knows someone who did exactly that and made it pay.
Which is his whole case for earned media. Nobody at Google is going to decide the BBC was a spammy referring domain. He also has a good jab at Google’s guidelines telling you not to manipulate the algorithm at all, since producing content is a manipulation of the algorithm. Jeremy’s reply: sure, it’s a gift. “Here you go, Google. Show this in your AI overview. We just wrote it for fun.”
So how should you decide? Chris boils it down to three things: “Risk, tolerance for risk, what is going on in your specific industry and like what your budget is.” If a best-slot-machine keyword makes you a hundred grand a month, of course you’ll keep paying for links. If you want to be a trusted entity in Google and in the large language models, that’s a different kind of link. There’s more on why that second kind is getting scarcer, and more valuable, in links are getting rarer, which is exactly why they are about to matter more.
What the LLMs actually cite
Here’s the part of the episode you can act on this week. Google hid its sources. ChatGPT shows them. Chris lists what keeps turning up: “tier one media is a big trusted source, heavily utilized by the large language models,” then YouTube videos and transcripts, Reddit threads, and listicles.
And the listicles aren’t always the New Yorker. His example is a Chicago food blog listing the ten best chefs in the city. If a mediocre chef gets onto that list, the model recommends him more. Whether that happens can come down to who owns the blog and whether they’ll amend the list for a payment. Jeremy’s version is the carnival tent: you can hoodwink people inside, but there had better be a mermaid in there, even if she’s an old lady in a costume. Chris agrees. If you’re the worst chef in Chicago and you buy your way onto the best list, you better be making at least okay food.
Ann Smarty goes deep on the same mechanic in LLM consensus, Reddit and brand control.
Are these tools fit for purpose?
The back third turns skeptical. Jeremy rattles off AI overviews recommending gasoline in recipes and dangerous dive depths, and asks why tools this unreliable are out of beta at all. Chris points out that Google never used to put a disclaimer on its first page, and that the models now do. He cites Charles Floate’s view that the improvements will flatten, and raises the training data problem: if everything is AI-written, what does the next model learn from? That exchange is why he’s quoted in SEO Arcade’s roundup, What SEOs Are Actually Handing to the Robots.
His answer to it is practical. Podcasts and YouTube are fresh, original content, and the models are hungry for exactly that. Jeremy calls it information gain, Michael McDougald’s term: nothing in this conversation would have come out of a keyword research tool. Cahill Camden makes the case for podcasts as a trust channel from the other side of the mic.
“It’s brand, brand, brand, brand. So yeah, I think we’re in the era of brand.”
Chris Panteli, Linkifi
Questions this episode answers
Does a link need to come from a topically relevant page?
For a digital PR link, relevance is mostly out of your hands and Chris doesn’t sweat it; the value is branded homepage authority. For a specific money page in a competitive niche, relevance and anchor type matter more, so study what the pages above you have.
Can a link from a major publication hurt you?
Chris’s view is that there is almost no chance Google devalues a link from the BBC, which is why he calls digital PR a zero risk game.
Where do ChatGPT and other LLMs get their recommendations?
From sources you can see: tier one media, YouTube videos and transcripts, Reddit threads and listicles, including small niche blogs.
Where to find Chris Panteli
Chris is on LinkedIn as Christopher Panteli. Linkifi’s free digital PR cheat sheet walks you through landing your first few media links yourself, using journalist request platforms like HARO and Source of Sources. He also co-hosts the Market Movers podcast.
About this recap
Every quotation on this page is verbatim from the episode transcript and is Chris Panteli speaking, unless it is attributed to Jeremy. The episode was recorded in August 2025, so event and model references reflect that date.
Cite this episode
Rivera, Jeremy. Chris Panteli: Digital PR, Brand Signals and the Zero-Risk Link. Unscripted SEO. Guest: Chris Panteli, co-founder of Linkifi. Available at unscriptedseo.com/chris-panteli-digital-pr-zero-risk-link/
