Most guest posting programmes fail for a boring reason: the team is buying inventory and calling it outreach. They pay a fee, send a draft nobody edits, and get a link on a site with a blog nobody reads. Six months later the links are still there, the rankings have not moved, and nobody can explain why.
Guest posting for SEO still works. What has changed is how narrow the window is between a placement that carries real weight and one that is either ignored or actively counted against you. This article covers where that line sits, how to find publications worth pitching, how to write a pitch that gets a reply, and how to tell within a few weeks whether the programme is worth continuing.
Ask two experienced SEOs about guest posting and you can get opposite answers. One will tell you it is the most reliable link acquisition channel they have. The other will tell you it is a tax on people who have not read Google’s documentation. Both are describing something real, because the term covers two completely different activities that happen to produce the same HTML.
Google’s spam policies for web search define link spam as creating links to or from a site primarily for the purpose of manipulating search rankings. The list of examples includes advertorials or native advertising where payment is received for articles containing links that pass ranking credit, and links with optimised anchor text in articles, guest posts, or press releases distributed on other sites.
Read that carefully, because the qualifiers matter. The policy does not say guest posts are prohibited. It targets two specific conditions: payment in exchange for a link that passes ranking credit, and optimised anchor text distributed across placements. Google also states plainly that buying and selling links is a normal part of the web economy for advertising and sponsorship, and that such arrangements do not violate policy provided the links are qualified with a rel=”nofollow” or rel=”sponsored” attribute.
So the policy position is coherent: pay for placement if you want, but disclose it through the link attribute. The moment you pay for a following link and dress it up as an editorial contribution, you are outside the guidelines regardless of how good the article is.
The practical test is simple. Ask whether the publication would have run your piece if you had offered nothing but the piece itself. If an editor read your pitch, decided it was interesting for their readers, and scheduled it, that is a placement. If a rate card determines the outcome, that is a purchase, and it should carry a sponsored attribute.
This distinction is not academic. Sites that sell followed links at scale tend to accumulate a recognisable footprint: hundreds of contributor posts across unrelated verticals links, thin author bios, sudden publishing volume, and outbound commercial anchors in nearly every post. When a network like that gets devalued, every link on it goes quiet at once including yours.
Before evaluating any specific site, it helps to know what you are actually looking for. Three attributes do most of the work.
Someone on the other side should push back on your draft. An editor link who asks you to cut the promotional paragraph, tighten the intro, or add a real example is a signal that the publication protects its own standards. A site that publishes whatever arrives, unchanged, within 48 hours is telling you exactly how much its endorsement is worth.
You can test this before you commit. Send a pitch with two angles and ask which fits their audience better. A real editor will have an opinion. A link vendor will say both are fine and ask for payment details.
Relevance operates at two levels. Site-level relevance asks whether the publication covers your subject area at all. Page-level relevance asks whether the specific article your link sits in is about something adjacent to the page you are linking to. A fintech tool linked from a personal finance publication in an article about budgeting apps has both. The same tool linked from a general business blog in a listicle about office furniture has neither, no matter what the domain metrics say.
A publication with an audience behaves differently from one built for link inventory. Look for comments, newsletter signups, social sharing that predates your involvement, named staff writers, and archive pages showing consistent publishing over years rather than a spike in the last eight months. None of these are ranking factors. They are evidence that the site exists for a reason other than selling links.
Prospecting is where most programmes quietly go wrong, because the easiest lists to build are the worst ones. Anything that appears on the first page for “guest post opportunities” has been worked over by thousands of people already.
The strongest list is the one you can build without a tool. Ask your sales team which publications customers mention. Check which sites drive referral traffic in your analytics. Look at the newsletters your team subscribes to. These sites are relevant by definition, and a pitch from someone who clearly reads the publication stands out immediately.
Operators are still useful, but combine them with a topic rather than with the words “guest post” alone. Try these patterns in your niche:
Run each list through the evaluation signals above before adding anything to outreach. Prospecting volume is cheap; outreach volume is not.
Find someone who writes well in your space and is published in several places, then look at where else their byline appears. Practitioners who contribute regularly tend to write for publications with real editors, because those are the outlets worth having on a CV. Following three or four such trails will usually surface a better list than any scraped database.
Editors at any publication worth targeting receive dozens of identical pitches weekly. Most open with a compliment about the blog, mention a vague interest in collaboration, and offer three generic titles. Yours needs to do the opposite.
Keep the first email short enough to read on a phone. One sentence establishing that you actually read the publication, referencing a specific piece and what you thought of it. One sentence with a single, specific angle and why their readers would care. One sentence on why you are the person to write it the experience, data, or access that makes the piece possible.
Do not attach a draft. Do not offer to pay. Do not mention SEO, backlinks, or domain authority at any point. Editors who care about their publication find that framing insulting, and editors who do not care will bring up money themselves.
A commissioned post is not a favour you have been granted. It is a commitment to deliver something the publication would be happy to have written itself.
A link inside the body of the article, placed where a reader genuinely benefits from following it, carries more weight and looks more natural than a link buried in an author bio. But it has to survive editorial scrutiny. If the only way to justify the link is to bend a paragraph towards it, cut the link and take the bio placement instead.
One or two outbound links to your own material in a 1,500-word piece is defensible. Five is a red flag to the editor and to anyone later reviewing your link profile.
Google’s policy language specifically names optimised anchor text in guest posts as an example of link spam. This is the single easiest way to turn an otherwise legitimate placement into a liability. Use your brand name, the article title, or a natural descriptive phrase. If you find yourself negotiating with an editor over exact anchor wording, you have drifted into buying links.
Guest posting produces slow, compounding results, which makes it easy to keep funding something that stopped working months ago. Track leading indicators rather than waiting for rankings.
Referral traffic deserves particular attention. It is the one metric a link vendor cannot fake, and it correlates with the thing you actually want placement on pages real people read.
⦁ Chasing domain metrics in isolation. Third-party authority scores are estimates built from link graphs. A high score on an irrelevant site is worth less than a modest score on the publication your customers read.
⦁ Scaling before the process works. Ten placements a month through a process that has not produced a single referral visit multiplies a problem rather than solving it.
⦁ Recycling one article across many sites. Republished or lightly spun content puts the host site at risk and marks your placements with an obvious footprint.
⦁ Ignoring the link attribute. If you paid for the placement, a sponsored attribute is the correct outcome, not a failure. Argue about it and you are documenting intent.
⦁ No follow-through. A placement that generates a reader question you never answer is a relationship you have wasted.
Yes, when the placement is genuinely editorial. Google’s spam policies target paid links that pass ranking credit and optimize anchor text distributed across articles, not the practice of contributing writing to publications. The tactic has not been devalued; the low-effort version of it has.
There is no safe number, because volume is not what creates risk. Twenty placements earned across relevant publications with varied anchors and real editorial review are less risky than three bought from a network. Set your pace by how many quality pitches you can genuinely develop, which for most teams is a handful per month.
Often, yes. A nofollow link from a publication your audience reads still delivers referral traffic, brand exposure, and the kind of visibility that leads to unprompted citations later. If the only value you can see in a placement is the following link, that is a signal the placement was not worth pursuing in the first place.
Treat it as a different transaction. You can still proceed, but the link should carry a sponsored or nofollow attribute, and you should evaluate the placement on referral traffic and brand exposure rather than ranking value. If they refuse to qualify the link, walk away they are selling something Google’s documentation explicitly identifies as a policy violation.
Is it worth guest posting on a site with low traffic?
Sometimes. A small, specialised publication read by a few thousand people in your exact industry can be more valuable than a large general site, both for relevance and for the readers it puts you in front of. What matters is whether the audience is real and relevant, not how large it is.
The useful way to think about guest posting for SEO is that you are not acquiring a link, you are earning a publication’s willingness to put your name in front of their readers. Links follow from that, and they hold their value because the placement was never contingent on payment.
If you are starting a programme, resist the urge to build a large prospect list first. Pick five publications your customers genuinely read, study what they publish, and develop one strong pitch for each. The acceptance rate on that approach will teach you more in a month than a scraped list of two hundred sites will teach you in a year and the placements you land will still be working for you long after the bought ones stop counting.
Your customers are searching. Make sure they find you first. Let’s build a strategy that actually moves the needle—more visibility, more traffic, more revenue.