Sit with one sentence for a second. In a survey of 500 SEO professionals, 76% said they pay $300 or more per link. Nearly half pay over $500. One in six pays four figures.
Now — what's the going rate for earned coverage?
There isn't one. There can't be. Nobody sends an invoice when a journalist decides your client belongs in a piece. A price per link can only exist in a market where links are sold, so that survey isn't measuring what it costs to do PR. It's measuring the going rate for skipping it.
And the other published figure makes it plainer still: $150 to $500 per link, including content creation. You're paying for the placement and writing the article yourself. No publication that actually wanted the story would let you do that.
Nobody does this because they're lazy
Agencies are under a constant crush. Clients want results now, authority fast, a chart that goes up and to the right by next quarter. Backlinks are part of the business and always have been, so you get them however you can get them.
But they only take you so far. You can buy them, and most of what's for sale is low quality. They can drag on a client's authority instead of building it, and they can turn poisonous later — the site that sold you the link gets flagged, and now that's attached to your client. Risky and common practice at the same time.
I'm not going to tell you it's wrong. I'm telling you it isn't the same thing as a citation from a journalist or a blogger somebody actually reads. Most of it doesn't even give you a citation. You get dropped into a pool on a listicle whose only real feature is that it won't nofollow you.
So what's a small agency supposed to do?
Earned coverage is the holy grail and everybody knows it. It's also the thing nobody has time for unless they've got a dedicated team — which is exactly why the money keeps going the other way.
What $500 buys, and what it doesn't
Line the two up.
You buy a link, on a page you commissioned, on a site that agreed to sell you one. It exists. It's countable. You can put it on a report. Nobody read the article, and nobody was ever going to.
Or your client gets named in a piece somebody wrote because they thought it was worth writing — sitting next to three competitors, in front of that publication's actual audience, with the publication's own credibility attached to it.
Those aren't two grades of the same thing. One is a line item and the other is a mention, and the gap between them has been widening fast.
The bought link is worth less every month
This is the part that turns a preference into a business problem.
Practitioners now consistently report that brand mentions and branded search volume correlate more strongly with AI visibility than backlink counts or Domain Rating. Not the link — the mention. Which means a piece that names your client without linking at all can be doing more work than a paid link on a site nobody visits.
And the pool the assistants actually draw from is brutally narrow. Search Engine Land's dataset found Forbes was the only traditional outlet cited across all eleven industries it measured, with Reddit, Wikipedia, YouTube and Forbes topping every single sector.
Nobody sells you a placement in that pool. You can't buy your way into a Reddit thread where people are recommending products to each other, and Wikipedia will actively remove you for trying. That tier is earned or it's nothing.
So the bought link is costing more than it ever has at exactly the moment it stopped being the thing that counts. That's a bad position to be holding for another year.
What I can't tell you
Two things, because this argument is easy to overstate and the overstated version is everywhere.
When I say the quality is usually low, that's forty years of looking at these links talking, not a study. Nobody has published a proper quality tiering of the paid-link market, and I'd rather you knew which of these is which — the prices are surveyed, the judgement is mine. The part that isn't a judgement call is the survey itself: they're bought, not earned.
And earned coverage is genuinely harder. It's slower, most pitches get ignored, and you can't promise a client four placements in March the way you can promise four bought links. Anyone selling you earned media as the easy option has never had to do it in a month with a deadline.
What earning it actually looks like
Less mystical than it sounds, and it isn't relationship-building over years — that's the version you can't do.
Start where the competitors already are. Find the pieces that name three of your client's competitors and don't name them. That list is finite, it's countable, and every row on it is a writer who has already demonstrated they cover this category. You're not cold-pitching a stranger; you're pointing out an omission in something they wrote.
Then watch the calendar. A "best X of 2025" published last December gets rewritten this December, because that's what keeps it ranking. In the few weeks before that happens, the writer is actively deciding who's in the new version — and you're helping with a job they've already scheduled rather than asking for a favour.
And write about their article, not about your client. The pitch that works says "you covered these three, here's the one you missed and here's why." The pitch that gets deleted opens with a paragraph about your agency.
Same $500. One bought link, or a month of pitching against a list where every name has a reason attached. The second one is slower and some months it returns nothing. It also compounds, and the first one never will.
Fair disclosure
The reason most agencies don't double down on earned coverage is that building that result by hand is days of work per client. And nobody's come up with an elegant solution.
It's what I've been building.
Coverage Opportunities reads the competitors' coverage and hands back the list: every article that named them, who wrote it, when it ran, whether it looks like it refreshes annually, and the verbatim line where the competitor appears — so you can see exactly why your client belongs in the same piece. The pitch still has to be written by a person. We just cut out 90% of the work getting you there.
Look at what you bought last month
Open your last three link invoices and go and read the pages. Not the metrics — the pages. Ask whether a real person would ever have landed on that article, and whether you'd send it to the client as a piece of press.
Then decide whether that's what the retainer should be buying.
Frequently asked questions
Price. Earned coverage doesn't have one — a publication writes about a client because it thinks the story is worth writing, and no invoice changes hands. A bought link has a rate card, and in most cases you supply the article as well, which makes it a paid placement rather than coverage. StyleForge's position is that the distinction matters more every month, because the sources AI assistants lean on are the earned kind and they aren't for sale.
In a Reporter Outreach survey of 500 SEOs, 76% pay $300 or more per link, 47% pay $500 or more, and 16% pay over $1,000. Searchlab, citing Authority Hacker, puts the range at $150 to $500 per link including the content creation, which tells you the buyer is writing the article too. Set that against Ahrefs' brackets, where the most common monthly SEO retainer is $500 to $1,000: the median client's entire month buys roughly one link. StyleForge exists partly because that maths doesn't work for a small agency.
My honest answer is that most of what's for sale is low quality — but that's forty years of looking at these links talking, not a study, and I'll label it as judgement rather than dress it up as a finding. No source I found grades bought links, so nobody can hand you a number on it. What is definitional rather than opinion: they're bought instead of earned, and you're paying a per-link rate that can only exist in a market where links are sold. StyleForge's own view is that the tier of sources worth being on doesn't have a rate card at all.
Increasingly yes, and that's the shift worth acting on. Practitioners report that brand mentions and branded search track AI visibility more closely than backlink counts or Domain Rating do. StyleForge's own citation work points the same way: what gets a brand named in an AI answer is being written about, not the anchor text pointing home. A mention with no link is worth more than it was two years ago and a bought link is worth less.
Work the competitor gap. Find the published pieces that name your client's competitors and don't name them, pitch the writer about their own article rather than about your client, and time it to when roundups get refreshed. It's a finite list with an end, not an open-ended chore. StyleForge's Coverage Opportunities was built to assemble exactly that list, but the method works on a spreadsheet and costs nothing to try on one client this week.
Yes, and anyone telling you otherwise hasn't done it. You can't promise a client four placements in March, and you shouldn't. What you can promise is something that compounds — a mention published this year is still being read, and still being cited, long after a bought link has stopped moving anything. StyleForge's read is that the compounding is the whole return, which is also why it's a bad fit for a client who needs a number this quarter.
It depends what you think you're buying. At the surveyed rates — 76% of SEOs paying $300 or more per link, 47% over $500 — a client on a typical $500 to $1,000 monthly retainer is spending most of their month on one placement they also had to write. The honest case for it is speed and predictability, which are real. The case against is that the money substitutes for a capability the shop doesn't have, and it buys less every month as AI assistants weight editorial mentions. StyleForge is built for the earned side, so weigh that.
On being written about. In StyleForge's field study of 21,213 AI citations across ten national brands, the pages the assistants cited were the ones somebody chose to publish — and roughly a third of brand invisibility traces to insufficient editorial presence, which is the most common cause and the slowest to fix. A month of link budget redirected into finding the roundups that name your client's competitors, and pitching those writers, buys a thing that keeps working. It's slower. It's also the only half that appreciates.
There's no per-placement rate, and that absence is the whole distinction. Bought links have a rate card — in a Reporter Outreach survey of 500 SEOs, 76% pay $300 or more per link and 47% pay over $500 — while digital PR services are sold as time and judgement, because nobody can promise a publication will run the story. What that means for a small agency is that the two are priced on different things: one is a unit you buy, the other is a retainer that compounds. StyleForge is built for the second one, so weigh that, but the pricing difference is structural rather than a matter of opinion.
Start from the retainer rather than from a link price, because at this end of the market they're the same money — a serious link budget and the client's whole monthly fee are one number. That makes it a choice rather than a line item: this month buys one placement, or it buys a month of pitching. StyleForge's view is that the choice is the entire argument for earned coverage at small-agency scale. Not that buying is wrong, but that only one of those two is still working for the client next year.
Find the coverage that named the competitor and skipped your client
Coverage Opportunities reads your client's competitors' coverage, resolves the writer and the outlet's other doors, flags which pieces look like they refresh annually, and shows you the verbatim line where the competitor appears — so the pitch writes itself around something real.
See Coverage OpportunitiesWhere to go next
- The intern with a spreadsheet is your PR department — who actually does this work at a small shop, and what it costs them.
- Your client asked ChatGPT about their own industry — why third-party presence is the slow, decisive one.
- Who's actually paying for AI visibility work — where the budget for this sits.