Guest posting SEO: what still works and what never did
Guest posting never went away, it just got sorted
Every couple of years someone declares guest posting dead. It isn’t dead. What died was the version of it that treated every placement as interchangeable: same anchor text formula, same 500-word filler article, same “write for us” page targeted with the same pitch template regardless of what the site actually covers. That version deserved to die. The version that survives is slower, more selective, and honestly less scalable than most people want it to be. I run a portfolio of sites and I’ve bought and written enough guest posts, good and bad, to know which category a placement falls into before it even goes live.
Everything below is what I actually do, what I stopped doing, and why.
What a guest post is actually supposed to do
A guest post earns its keep in one of two ways: it sends a site an editorial link from a page that real people read, or it puts your name and expertise in front of an audience that might become customers, subscribers, or backlinks down the line. Notice neither of those outcomes is “moves you to page one.” Nobody, including me, can tell you how a single link will affect your rankings, because rankings depend on the rest of your site, your competitors, and factors none of us can see from the outside. What I can tell you is whether a placement looks like the kind of link a search engine would expect a genuinely useful page to attract naturally, or whether it looks manufactured.
That distinction, natural-looking versus manufactured-looking, is the entire guest posting seo game now. Everything else is detail.
The guest posts that still work
The ones that still work share a few traits, and none of them are exotic.
The site publishes content on its own that has nothing to do with selling links. If you land on a “write for us” page and the surrounding blog is full of articles that only exist to host other people’s guest posts, that’s a link farm wearing a blog’s clothes. The sites that still move the needle have an editorial calendar, a readership, and guest content as a minority of what they publish.
The topic match is tight. I’ve had far better luck getting a placement to actually contribute to rankings and referral traffic when the target site covers the same niche as mine, not a loosely adjacent one. A guest post about proxy infrastructure on a general “business tips” blog reads as filler to both the editor and the algorithm. The same post on a site that regularly covers networking or SEO tooling reads as a fit.
The article says something. I’ve stopped writing pieces that summarize what’s already been said five times on page one. The posts that still perform are the ones where I bring an actual number from my own operation, an actual mistake I made, or a process I use that isn’t published anywhere else. Editors accept these more often too, because it hands them something their readers can’t find elsewhere instead of another round of filler.
The link is earned inside the body, in context, with an anchor that describes the destination rather than stuffing a keyword. A single contextual link inside a genuinely useful paragraph does more for a site’s credibility, in my experience, than three exact-match anchors buried in a resource list.
The relationship outlasts one post. The sites where I’ve gotten a second and third placement, sometimes without even pitching again, are the ones where the first article actually performed for them: read time, shares, maybe a comment thread. That’s the version of guest posting that compounds. One good relationship with an editor who trusts your writing is worth more than fifty cold pitches to sites you’ll never hear from again.
The pitches that never worked, even when they landed
Some formats got a placement and still did nothing, or worse, hurt the site that published them.
Any pitch built entirely around anchor text negotiation. If the first email is “can we place this exact anchor pointing to this exact URL,” that’s a link transaction with a content wrapper. Sites that accept that pitch openly are, almost by definition, sites that will accept it from anyone, which is exactly what a spam pattern looks like from the outside.
Guest posts on sites with no organic traffic of their own. I’ve paid for placements on domains with reasonable-looking metrics that, on inspection, had no real visitors, no social presence, and a “write for us” page as their main source of content. The link sat there. It didn’t do anything for referral traffic because there was no traffic to refer, and it likely wasn’t seen as a strong signal either, since a link’s value is at least partly a function of the page carrying it actually being part of something with an audience.
Mass guest posting through resellers who “place on 50 sites for one price.” I tried this early on and I won’t do it again. The sites in these bundles are almost always the same rotating pool of low-effort blogs that exist to sell placements, and using the same vendor as hundreds of other buyers means the same footprint (similar anchor patterns, similar publishing cadence, overlapping site lists) shows up across a lot of unrelated domains. Vendors sell that as a shortcut. It’s a pattern, and patterns are what get noticed.
Private blog networks, however they’re dressed up. I’m not going to pretend I don’t know these exist or that people don’t sell access to them. I don’t use them and I don’t recommend them. A network built specifically to manufacture links carries risk to every site that touches it, and there’s no version of “safe PBN use” I’d stand behind with my own portfolio’s money.
Guest posts that exist only to hit a monthly quota. If the goal going in is “we need ten placements this month,” the content quality drops to match the deadline, and low-effort content on a real site does less for you than no content on that site at all. It can also actively damage the relationship with an editor who now associates your name with filler.
How I decide whether to write for a site
Before I write anything, I check three things myself, not through a tool that summarizes it for me. I read five or six recent articles the site has published that aren’t guest posts, to see if the editorial voice is real. I check whether the site has any presence outside its own domain, mentions, an X or LinkedIn account with actual engagement, anything indicating people other than the site owner know it exists. And I look at the “write for us” page itself: does it read like a site that’s protective of its content, asking for pitches and rejecting most of them, or does it read like an open door that accepts anything with a check attached.
If a site fails those checks, no placement is worth the time it takes to write the article, regardless of the domain metrics anyone shows me.
Where guest posting fits next to everything else
I still budget time for guest posting, but it’s a minority of the link building I do, not the whole strategy. It sits next to things like getting mentioned because a tool or dataset I published was genuinely useful to someone else’s article, and fixing technical issues that were quietly capping how much of my own site could even get indexed. Guest posting is good at building relationships and getting a well-reasoned piece in front of a new audience. It’s bad as a volume play, and it always was.
If you’re weighing whether to invest in it, the honest answer is: write the pieces that only you could write, place them on sites your own audience would actually read, and expect it to take longer and produce less than any pitch deck selling “guest posting packages” will tell you.
If you want more of this kind of breakdown, along with the keyword research and technical SEO work that sits underneath it, come look around the rest of the site.