How much does a press release cost in 2026?

The price of a press release in 2026 runs from about $50 to about $9,500. That is a two-hundred-fold spread for a product that is, at every price point, the same act: your words, dated, reprinted on domains you do not own.

Most articles answering this question are written by a vendor and arrive at the conclusion that the vendor's price is correct. This one is also written by a vendor. So here are the published rates first, from every service, without commentary, and you can decide what the commentary is worth afterwards.

Published rates in 2026

Detail worth having: PRWeb's middle tier is $189. EIN's bundles fall to $49, $41, $37 and $33 per release at 6, 12, 24 and 48. eReleases sits at $399, $499 and $699 across three tiers and has partnered with PR Newswire since 2002, with human editorial review on every release. Cited's middle tier is $149, with add-ons from $89 to $349.

Add-ons move enterprise numbers a great deal. Extra words, images and AP syndication are all billed separately on the wires that sell them that way, which is why the $350 figure for PR Newswire and the $9,500 figure are both accurate.

Why the spread is so wide

Three things are being priced, and they are bundled differently at each tier.

Distribution is the cheapest part and the one every vendor advertises. Syndication feeds are close to a commodity. This is why the bottom of the market keeps falling and why a $50 release and a $500 release can reach a similar number of domains.

Editorial review is a real cost and appears around the $400 mark. eReleases has a human read every release. That is worth something if you are unsure whether your announcement is wire-legal, and worth nothing if you already know it is.

Named mastheads are where the money goes at the top. AP News, USA Today, Yahoo Finance networks. These are priced as add-ons almost everywhere because they cost the distributor more. On Cited, AP News is $199, USA Today is $179 and the Yahoo network is $349, and all three are included in Famous at $399.

The reason the top of the market reaches $9,500 is not that the release travels further. It is that enterprise buyers need guarantees, contracts, embargo handling and account management, and those are people-costs rather than distribution costs.

What you are actually buying at any price

This is the part that determines whether the spend was worth it, and it does not vary much across the table above.

You are not buying coverage. A journalist is not going to write about you because a release crossed a feed. A wire is syndication: the original copy, dated, reprinted. Coverage is a reporter who chose your story, and that is bought with a targeted pitch or not at all.

You are not buying readers. Wire releases are read by close to nobody in the wild, at every price. This is normal and it is not the point.

You are buying a dated public record. A page that says a specific thing about a named person, on a domain with standing, that search engines and AI assistants can retrieve when someone asks about you by name. That is a real asset and it is worth real money. It is just a much narrower asset than the marketing implies.

The evidence on what a release does for AI visibility

Since most people now buying a release are thinking about ChatGPT rather than journalists, the 2026 testing is worth quoting directly.

Across 100 category queries on ChatGPT, Perplexity and Gemini, covering ten verticals, zero press release domains appeared. A 0% citation rate. Those answers came from review platforms and editorial listicles: G2, NerdWallet, Clutch, Wirecutter. No release at any price wins a category question.

On brand and name queries the result inverted. Releases on Yahoo Finance's /news/ path drew confirmed citations from ChatGPT and Gemini, while identical content on a /press-releases/ path scored "rarely" or lower. Same words, same vendor, different URL, different outcome.

Separately, Muck Rack's work found 82% of AI citations came from earned media and 94% from non-paid sources, which is the number that should keep any wire vendor honest. A release seeds; downstream coverage gets cited. Third-party news distribution showed a 34% citation rate, a 325% lift over brand-owned content, which is the case for paying anything at all.

The practical conclusion: price is close to irrelevant to this outcome, and URL path is close to decisive. Ask any vendor which paths their placements land on before comparing their price to anyone else's.

What a release should cost you

A rough guide, by situation rather than by vendor.

• One announcement, want a dated record on your name. $79 to $150. Anything more is buying reach you have no use for.
• Want a specific masthead for the screenshot. $180 to $400, because that is what the named networks cost wholesale. Check the URL path first.
• Unsure whether your news is wire-legal. Around $400 for human editorial review, or write it properly and save the money.
• Real news four or more times a year. A bundle at $33 to $49 per release, and only if you are certain about the cadence.
• Public company, regulatory disclosure, embargo. Enterprise wire. The price is what it is and this article is not for you.
• No news. Nothing. A job title is not news, and a release about nothing is a page that helps no one.

Cadence changes the arithmetic

One figure from the 2026 retrieval research reframes the whole buying decision. Roughly 70% of the content behind an AI answer for a given query turns over. The pool a model draws from is not an archive; it moves.

A single release is a page that still exists a year later and has drifted out of what the systems reach for. That is not link rot, and the page has not been deleted. It has simply stopped being what a retrieval system picks up when someone asks.

The implication is uncomfortable for anyone selling one-off releases, including us. If being findable matters to you continuously, a release a quarter tied to something real will do more than one expensive release will. If it matters once, because you are about to be googled by an investor or a client, buy once and time it.

What this does not justify is a bundle bought against news you have not made. Four releases a year about nothing is worse than one release about something. Editors reject the empty ones, and retrieval systems ignore what editors let through.

The free work most people skip

There is a category of findability work that costs nothing, that every vendor omits from their comparison table, and that determines how much value any release delivers.

AI systems identify a person by triangulating across sources: your own site's structured data, third-party records such as Wikidata, Crunchbase and LinkedIn, and directory listings. When those agree and link to each other, a system can resolve you quickly and confidently. When they conflict or do not exist, it infers from scattered signals, and low-confidence entities are cited less often than structured ones.

Three jobs, one afternoon, zero dollars: Person schema on your site with a sameAs array pointing at every profile that names you; identical name, role and company strings everywhere; and a Wikidata entry if you clear the notability bar.

Do that before buying a $399 release rather than after. The release gives a system something to cite. The entity record is how it knows the citation is about you and not about the other person with your name.

How to time a release so the money works harder

A release is a dated record, which means the date is part of what you are buying.

Ship it when the fact is fresh, because a wire will question a launch announced four months late, and a reader who finds it will too. Avoid saturated news cycles, where a release competes with real news for the same attention and loses. And put it out before the moment you need it, not during: if the point is that an investor googles you next month, the page needs time to be indexed and picked up.

The other half of timing is having something to say. A launch, a hire, a funding round, a book, a study, a new practice, a partnership, a milestone with a number attached. A job title is not news. Neither is an opinion, a rebrand nobody asked about, or a "thought leadership" statement, and wires reject all three at a rate that surprises first-time buyers.

What "as seen in" is worth

A large share of what people are really buying at the $180-to-$400 end is a logo they can put on a landing page. It is worth being precise about what that logo represents, because the gap between a real placement and a decorative badge is where this market gets its bad reputation.

A wire placement on a named masthead is a live URL on that domain, carrying your dated copy, labelled as a press release. It is real in the sense that matters: you can click it, a stranger can click it, and it says what it says. It is not a journalist at that publication choosing to write about you, and it is not editorial endorsement. Anyone presenting it as either is misrepresenting it.

The test is simple and it is the one to apply to any competitor's "as seen in" row as well. Click every logo. If the logo does not link to a live page naming the person, it is decoration. If the page 404s, it was decoration and now it is not even that. A logo pack bought as a graphic is worth nothing and carries a real risk of looking like fraud to anyone who checks.

Priced against that standard, $179 for a USA Today placement or $199 for AP News is a defensible number for what it is: a page on a domain that a stranger already believes, which will still be there when someone looks. What it is not is press coverage, and no price buys that on a wire.

The one number to measure afterwards

Whatever you spend, there is a single outcome worth tracking, and it is not traffic.

Search your own name in quotes and see what returns. Ask an assistant what it can say about you and see whether it can name a fact. Do both before you buy and again a month after. If the answer changes from nothing to a dated page on a domain with standing, the spend worked, at $79 or at $399. If it does not change, no amount of reach made up for the release landing somewhere that nothing retrieves.

That measurement takes two minutes and it is the only one that maps to what you were actually buying. Traffic will be near zero at every price, which is normal. Rankings will not move on nofollow syndication. Whether a system can now answer a question about you by name is the thing that either happened or did not.

The costs nobody lists

Two that catch people out.

Rejection. Wires reject releases, and the reasons are usually mundane: missing city, no website, promotional language, a claim that cannot be supported. You get the credits back and you have lost a week. Whether you get your money back depends on the vendor.

Link rot. Placements disappear. Pages rotate off, subdirectories get deindexed, canonicals move. Almost nobody re-checks a placement report at 30 days, which means almost nobody knows their real hit rate. If you buy one release a year, check yours; if you buy a bundle, check every time.

Before spending anything with anyone, it is worth knowing what you are fixing. Ask a model what it can say about your name — it is free and takes a few seconds. If it comes back blank, the $79 tier addresses that as well as the $399 one. If it comes back complete, you have just saved yourself the cost of a release.

Questions

How much does a press release cost in 2026?

Published rates run from about €49.95 flat at Pressonify to $9,500 at PR Newswire. Cited starts at $79, PRWeb Basic is $99, EIN Presswire is $149 for a single release, eReleases starts at $399 and Newswire.com sits around $599. Most single releases for individuals fall between $79 and $400.

Why do press release prices vary so much?

Three things are priced separately: syndication, which is close to a commodity; human editorial review, which appears around the $400 mark; and named mastheads like AP News, USA Today and the Yahoo network, which cost the distributor more and are usually sold as add-ons. Enterprise pricing also covers contracts, embargo handling and account management.

Is a press release worth the money?

It is worth it if you want a dated public record on your name that search engines and assistants can retrieve. It is not worth it if you expect coverage, readers or rankings. A wire is syndication, not journalism, at every price point.

Will a press release get me cited by ChatGPT?

On name queries, possibly; on category queries, no. Testing across 100 category queries on ChatGPT, Perplexity and Gemini returned zero press release domains. On brand queries, releases on editorial URL paths such as Yahoo Finance's /news/ drew confirmed citations while /press-releases/ paths did not.

What hidden costs should I expect?

Rejection, where the wire refuses a release for a missing field or promotional language and you lose a week, and link rot, where placements disappear within weeks without anyone telling you. Re-check every URL in the placement report at 30 days.