Hi everyone, this is Neo.

While keeping an eye on the SEO industry over the past few days, I came across a Search Engine Journal webinar with a headline that stopped me cold: How We Earned 1,000+ Links For GEO/SEO. One client, two years, and more than a thousand earned media citations — from The Wall Street Journal, Fortune, Axios, Fox Business, and Reuters.

Honestly, my first instinct was skepticism. Everyone in the industry knows links have gotten harder to earn and cheaper links have lost their value. When an agency claims a thousand-plus citations from top-tier media, you’d be right to raise an eyebrow.

But after reading the full recap and digging into their published case studies, I’m convinced this is the real deal. It’s a far cry from the doom-and-gloom “link building is dead, rush to GEO” takes flooding LinkedIn. This is a method built on actual client results, with numbers you can check yourself.

So here’s the full breakdown: why traditional link building stopped working, what “research-driven link building” actually looks like, the five data story angles that make journalists come to you, two complete case studies, and — most importantly — my take on how independent sellers can make this work in 2026.

Why Traditional Link Building Suddenly Stopped Working

Kevin Rowe, CEO of PureLinq, opened the session with a story he said he’d never told publicly.

Around 2020, PureLinq nearly went under. “Overnight I lost half of our business because link building just stopped working,” he said. So he spent two years rethinking everything about off-page SEO.

He broke the failure down into three forces. Let me expand on each with what I’ve seen in the field:

First: the value of links has been diluted. Guest posts, directory links, and the other volume plays that used to move the needle barely register anymore. Search engines have long since learned to recognize “links for the sake of links.”

Second: AI content has flooded the link economy. It’s now trivial to spin up content sites with AI and interlink them at scale. Those links aren’t just worthless — they can actively put you at risk.

Third, and most important: user behavior changed. Kevin pointed out that many SEO teams watch their click-through rate fall in Search Console while their rankings hold steady. That’s the weirdest signal of all, because your rankings didn’t drop — your audience did. Users are asking AI instead of scrolling the SERP. Your rankings hold, the clicks don’t come, and no amount of link-building effort can fix that.

Put those three together and the conclusion is unavoidable: the era of link building as a standalone tactic is over. You can’t treat “getting links” as an isolated action anymore.

Kevin’s Answer: One Original Study, Four Payouts

So if not link building as a tactic, then what? Kevin’s answer: original data research.

His core argument is that one genuinely valuable piece of research can earn four things at once:

1. Media links — journalists cite your data and naturally link to you.
2. Brand mentions — unlinked but valuable exposure.
3. AI citations — ChatGPT and AI Overviews use your data as a source of fact.
4. Social distribution — data-driven stories travel well on social platforms.

This is the same logic I’ve hammered on in earlier posts: in the AI era, a link isn’t really a “link” anymore — it’s a citation. The more your data gets cited, the more authoritative you look to both search engines and AI engines.

But Kevin took this further than most: he turned it into a pipeline. Not one lucky report, but a repeatable system covering which data is worth doing, what angle to tell, how often to publish, and how to get journalists pitching you.

The Most Valuable Slide: 5 Data Story Angles

Kevin said the most valuable slide in his deck was the one answering: “what makes a dataset worth a journalist’s time?” He narrowed it down to five angles. The full deck lives behind the webinar recording, but between his examples and my own experience, here’s the framework:

Angle one: data that breaks common sense. Journalists love “everyone thinks X, but the data says Y.” His student loan study — built entirely on free federal data on student debt — produced a counterintuitive finding that landed full feature stories on Fox News and Fox Business. And none of the source data was hard to get. “It’s really low hanging fruit and I think almost anybody can do it,” Kevin said.

Angle two: data tied to the news cycle. When a story is breaking, whoever can supply relevant data first becomes the media’s best friend. For an ecommerce client, PureLinq ran seasonal studies — holiday shopping stats, Mother’s Day gift trends, Easter candy data — publishing right before each seasonal window when journalists needed exactly that content.

Angle three: data with a local dimension. In the Q&A, Kevin was explicit: “Go local with news and data.” Find something happening at the city or state level, then find data tied to it. His team once surveyed homeowners with irrigation systems to learn what blocks outdoor water efficiency — a hyper-local study local media has a natural reason to cover.

Angle four: data with rankings or lists. Score and rank an industry. His team’s office real estate study ranked US financial districts by foreclosure risk — and earned Axios Seattle, Barron’s, and WSJ coverage. List-style data is inherently viral: those who made the list want to share it, and those who didn’t want to know why.

Angle five: data that can be updated on a schedule. A dataset you can only use once is a one-off news story. A dataset you can refresh monthly or quarterly becomes a machine that keeps producing coverage. The Wall Street Journal eventually asked PureLinq to update one of their datasets — the journalists had started depending on them.

Two Case Studies: Research Hub vs. Light-Touch Supplement

Kevin walked through two very different ways to run this play, and both are worth studying.

Case one: rebuild the blog as a research hub.

The client was a national B2B brand with no physical locations, losing visibility to local competitors. Its blog ran on commodity content — “how to collect debt,” “top three financing tips” — the kind of stuff any competitor could have published word for word. (Kevin’s action item: flag every post from your last quarter that a competitor could have published verbatim. Those are the posts to kill.)

The fix was drastic: delete the blog, rebuild it as a research hub, and publish original studies monthly.

Month one produced exactly one link. The client complained; Kevin talked him into staying the course. Placements climbed month over month after that — until journalists stopped waiting for his pitches and started emailing to ask when the next dataset would be ready. The campaign ended with 1,000+ citations across WSJ, Fortune, Axios, Fox Business, and Reuters.

Case two: keep the blog, add research on the side.

The second client was an online university in a heavily regulated category. Tearing down the blog wasn’t an option. So they kept it, added a handful of studies alongside, and put their own professors in front of the media — one-to-one pitching plus a run of podcast bookings.

One detail Kevin emphasized: a credentialed expert on the byline works fastest. One client’s study, authored by a plant science professor at Utah State University, now ranks first for its target terms and gets cited as the primary data point in AI Overviews.

The contrast between the two cases is the real lesson: this approach doesn’t require a heavy rebuild. A light-touch version works too.

Why This Play Pays Off Even More in the AI Era

The sharpest question of the session: if AI Overviews answer the query right on the results page, do links and rankings even matter anymore?

Kevin’s answer was blunt: “We are not in the zero click world. 100% not.”

His numbers back it up: the research pages pull traffic from AI Overviews on their own, and several hold position one for commercial keywords — not just informational ones.

Two more observations worth writing down:

1. ChatGPT clicks convert at a higher rate than average. Even in ecommerce. And of all the AI platforms his team tracks, ChatGPT is the largest measurable source of direct clicks — “it’s not even close to Google search with AI Overviews,” he said.

2. Links still matter in the AI era — but the logic changed. When AI systems cite your research data, you’re accumulating “machine-verifiable authority.” Same logic I covered in my post about AI recommending your brand while someone else takes the click: in AI-driven traffic allocation, what matters is whether your brand is worth citing.

Neo’s Take: How to Make This Work for Your Independent Site

Enough about their methodology — here’s what I think.

First, this play is actually an advantage for Chinese sellers. Why? Because overseas journalists genuinely need data studies, and a team willing to grind out solid English-language research has a real edge. Plenty of agencies are just wrapping free public data in a new skin. If your data is real and your angle is fresh, you have every chance to compete.

Second, don’t get burned by AI slop. Kevin’s quote: “AI slop is the biggest problem in research right now. If you send out slop to journalists and they catch you, and your data is wrong, they will probably stop working with you.” Free data plus an LLM makes it trivial to mass-produce “studies” — but wrong data kills your media credibility for good. Slow and real beats fast and fake.

Third, cadence beats one-off hits. PureLinq publishes roughly one study a month and keeps it up for over a year. Links are a compounding game: month one gets you one link, month twelve might get you a hundred. What most sellers lack isn’t ideas — it’s the patience to keep publishing. Pick one dataset you can refresh on a schedule and make it your recurring column.

Fourth, start small. You don’t need a full research hub on day one. Start with a single localized study. Sell gardening tools? Survey lawn-care habits across US states. Sell pet supplies? Build a “how much Americans spend on pets” ranking. Your data sources can be government open data, anonymized order data from your own store, or a small survey. The goal is to run the loop once and land your first media citation.

Fifth, bylines are a hidden lever. If you have an advisor with a PhD, or can get an industry expert to co-sign your study, use it. Expert byline + real data + regular updates is the hardest-hitting “trust asset” you can build in the AI era.

Let me leave you with this: link building isn’t dead — the action of “getting links” is. In 2026, a link is really about being seen, being cited, and being trusted. And original data research is, in my view, the single best lever I’ve seen that moves all three at once.

I’m Neo, and I write about independent site SEO. If you’re figuring out how to earn quality links for your site, drop your approach in the comments — I’d love to hear how you’re playing it.