Guest posts have had a bad name for over a decade.
Matt Cutts famously declared guest blogging “done” back in 2014. The reason: spammers had taken a legitimate practice — writing for someone else’s publication — and turned it into automated outreach for backlinks. Generic pitches. Outsourced content. Mass-blasted requests to any editor with an email address.
The reputation hasn’t really recovered. And Google’s recent move to extend its spam policies to AI Overviews (more on that below) is tightening the pressure on low-quality guest posts from a new direction.
But there’s a better way. A more modern way to execute guest articles so they actually provide value to the reader while ticking all the boxes AI engines need. We call this ‘citation-grade guest posting’. And this article is a summation of what we’ve learned pitching and placing hundreds of guest articles over the past five years: before AI search entered the game, and at a more concentrated level, after it completely changed search.
Don’t let the old-school SEO agencies tell you optimizing for AI search and Google search is all the same: it’s not.
This guide will cover:
- Why guest articles are one of the highest ROI SEO plays today
- The change taking place in AI SEO and why quality guest articles are key
- What really makes a guest article spammy — and why marketers still love them
- How guest articles in the AI search era are different — and the three essential elements every one needs to have
Why guest articles are one of the highest-ROI plays (especially for B2B SaaS)
A guest article (done properly) does three things at once. Few off-site tactics stack all three, which is why the ROI is hard to beat.
First, the backlink. Contextual, editorial, on a relevant publication. Still one of the most controllable off-site signals you can get.
Second, the brand mention. This is really what drives a brand’s AI visibility. Ahrefs analyzed 75,000 brands and found branded web mentions correlate with AI Overview visibility at 0.664 — over three times stronger than backlinks. A guest article gives you both, on the same page, with editorial credibility behind each.
Third, the part most guest posting advice ignores, a real conversion path. With our SaaS clients, we mine their sales data for the specific pain points buyers raise, then place guest articles targeting those long-tail queries and position the client as the solution. The result is referral traffic from qualified readers and booked demos.
Compare that to publishing the same kind of content on your own blog. On-site informational content used to be the workhorse for top-of-funnel traffic. It’s gotten harder. Around 60% of Google searches now end in zero clicks, and AI Overviews are eating most of the rest. Meanwhile, a guest article on a publication your buyers already read reaches them where they’re paying attention and gets cited by AI engines on the way.
The catch is that this only works when the article is genuinely good. Genuinely useful. And not produced as a wrapper for a backlink.
No Spam: Why focusing on quality guest articles is even more essential
On May 15, 2026, Google quietly updated its search spam policies to clarify that attempts to manipulate its generative AI responses — including AI Overviews and AI Mode — now fall under the same rules that govern traditional search. Search Engine Land has the breakdown.
The targets, based on industry reporting: biased self-promotional listicles, recommendation poisoning, scaled fake comparison content, manufactured mention patterns designed to push your brand into AI answers.
Google didn’t mention guest articles directly, but they didn’t have to. The spammy guest post industry runs on exactly the tactics Google’s telegraphing they’ll devalue.
If you’ve been buying cheap guest posts to game AI Overviews, the runway just got shorter.
But there’s a more interesting angle here, and veteran SEO Steve Toth put it well in a recent LinkedIn post:
“Google has been telling SEOs what not to do since 1997… Every time Google said ‘don’t,’ the cutting-edge SEOs tested it and won the channel… Remember: Google’s job is to protect Google. Your job is to test.”
He’s right.
Every Google “don’t” has historically been a signal about what to do more carefully, not what to abandon. For example, multiple studies (such as this one from Contently) show that the single biggest predictor of whether ChatGPT, Perplexity, and Google’s AI systems will recommend your brand is whether other sites are talking about you. AKA, brand mentions.
Google just told you not to manufacture those mentions. They didn’t say mentions don’t matter. They said spammy mentions are spam. Which has been true the whole time.
The policy update is just more evidence that the spammy tactics are getting squeezed. The authentic guest article, which is written by a human with a real point of view, placed on a publication an editor actually runs, just got more valuable.
Because cheap, spammy guest content is being depreciated. And one day, we predict, it will become altogether useless.
What makes a guest article spammy (and why marketers still use them)
A guest article is spammy when:
- The pitch was sent to anyone with an inbox. Generic outreach, no personalization, no familiarity with the publication.
- The host site isn’t related to your space. For example: a SaaS company guest posting on a recipe blog because the DR is 50.
- The content has no expertise behind it. The “author” is a freelancer who got the assignment that morning, or worse, an LLM with no editor checking the claims.
- The article exists to host the backlink. Strip out the link and there’s nothing left. The writing is scaffolding.
None of this is hidden. Editors have been complaining about these pitches for over a decade. Google has been devaluing the resulting links for almost as long.
So why do marketers still do this?
Because it’s cheap, fast, and look like progress on a spreadsheet. Also, for certain niches, they can work. But not so much in high-competition spaces, like B2B SaaS, where the traffic is highly valuable and content expectations are even higher.
And then there’s the cost factor. A marketplace placement costs a few hundred dollars, ships in two weeks, and produces a live URL with a backlink that an SEO dashboard will dutifully report. Volume metrics are easy to celebrate. Compounding outcomes are hard to measure. The spammy guest post industry has thrived in that gap for years.
But that gap is closing. AI search engines don’t care how many placements you have. They care whether the publications you appear in are credible, whether the article makes a real argument, and whether your brand is mentioned in a way a reader would trust. The marketers who keep buying volume are about to find out their numbers don’t translate.
A new and better way: citation-grade guest articles
Google search would tolerate a vague article. It matched keywords, weighed links, judged quality, and ranked the page. If your positioning was muddy, strong enough signals could still float you to the top. AI search has no such tolerance. To put you in an answer, it has to actually understand what your product is, who it’s for, and what problem it solves. Then it decides when to recommend you.
This is because AI synthesizes multiple sources across the web, forms a view, and hands the buyer a recommendation. It has opinions now. If it frames you wrong, or leaves you out, the buyer typically just takes the answer and moves on. So your job isn’t just to get mentioned; it’s to earn a clear, positive read of your brand from a system you don’t directly control.
A modern “citation-grade” guest article has to do that work. Here are three things to consider for each piece of content:
1. It’s built on facts and real experience
Don’t make your product sound like something it isn’t. Don’t invent a capability, stretch what a study found, or cite a stat that doesn’t say what you want it to.
The article should read like it was written by someone who has actually done the work, because the alternative is obvious to everyone now. Sourceless, generic authority gets skimmed and forgotten.
There’s a mechanical payoff to grounding everything in real evidence. You can absolutely bring a strong opinion (you should) but back it with something concrete: real numbers, real outcomes, original research nobody else has. That’s the material AI engines lift directly into answers, because it’s specific and hard to fake. An opinion with evidence behind it gets cited by name. An opinion floating on its own gets passed over. Give the model something it can quote, and you become the source instead of a footnote.
2. It reflects clear positioning and a unique POV
Even without AI search considerations, positioning a SaaS brand is tough. But it’s also essential.
Much like how a prospect needs to understand your product’s unique value before buying, AI models also need to understand this before recommending you. This essentially boils down to making four things clear:
1) what product category you’re in
2) who you’re for (ie your persona or ICP)
3) what you replace (competitive alternative)
4) what makes you different.
Check out our article on how weak positionings sabotages your AI visibility for more details on this.
If your guest article frames you one way and the rest of the internet frames you another, the model’s picture of you blurs. As a result, your brand gets ignored by LLMs entirely.
Specificity is the unfair advantage, especially for smaller brands. Take Descript. At around 200 people, it’s up against Adobe, yet it shows up consistently in AI answers about podcast editing. This is because that’s the one use case it’s relentlessly clear about, across every page and article. It doesn’t try to sound like an all-purpose creative suite. So when someone asks AI about editing a podcast, Descript is the obvious match.
Then there’s POV. Positioning tells the model what you are and the problems you solve; your point of view tells it what you believe about those problems. LLMs lean on sources that take a clear, repeatable stance. A guest article is your chance to plant that stake on a site the model already trusts: say exactly what you are, who you’re for, and what you actually think about the problem you solve. Do it the same way everywhere, and the signal compounds.
Like all successful marketing, a consistent message is key.
3. It solves a real problem for the reader
Your article should solve a real problem that the audience has. The more specific, the better.
AI engines retrieve answers by matching a reader’s question to content that addresses that exact situation: what the problem is, who has it, what they’re trying to do, where the process breaks down, and how a specific tool fixes that step. An article built around a real problem maps cleanly onto the questions buyers actually ask.
For example: Say the problem is measuring which LinkedIn posts actually drive pipeline. A buyer asks AI, “how do I tell which LinkedIn posts influence revenue?” The article that gets surfaced is the one that lays out that exact situation ( the marketer posting consistently but unable to connect any of it to deals) and then explains how a tool that ties post engagement to CRM events solves it. The product shows up because it’s the answer to the ultra-specific question being asked.
One of the best ways to get ideas for problem-focused guest article content is by looking at how your customers describe their most pressing, top-of-mind problems. Think sales call transcripts and G2 reviews. That voice of customer data is the best source for creating citation-grade guest content that drives demos and signups.
The (citation-grade) guest article comeback
Guest posting earned its spammy reputation. But the version that got mass-produced, isn’t the version that works now.
AI search rewards real expertise, clear positioning, and content that solves an actual problem for the reader. That’s impossible to fake at scale.
Policies will keep changing. Google will tighten its spam rules again. The AI engines will tweak how they retrieve and cite. But despite whatever changes come, AI search will keep rewarding content that’s genuinely useful, clearly positioned, and grounded in real expertise. Because that’s the content that actually answers the question a buyer asked.
Bet on that, and the next policy update isn’t a threat. Because you have a reliable vehicle for expanding your brand’s unique footprint in the eyes of LLMs.
