Favors.dev
There is a moment in every first launch where somebody suggests a press release, and it sounds like the adult thing to do. It has the shape of marketing. It has a price, a deadline and a deliverable. What it very rarely has is a reader.
I am not going to tell you press releases are dead, because that is lazy and it is also wrong. They do one narrow job extremely well. The problem is that the job they do is almost never the job an unknown startup is hiring them for. So here is the honest version: what the money buys, the three cases where it is the right call, the one modern reason a release is not entirely useless, and the stack I run instead.
Is a startup press release worth it?
For most early-stage startups, no. A wire release distributes your own words. It does not produce journalism, and an unknown company announcing a product is not news to anyone who has not already heard of it. The confusion comes from the word coverage. Coverage means a person with an audience decided your company was worth their byline. A release is you paying to put a document somewhere, which is a different transaction with a different outcome.
The tell is in what you receive afterward. You get a distribution report: how many outlets picked up the syndicated copy, how many views the release page got, which feeds carried it. None of those are readers who came looking for you, and none of them are an article somebody chose to write. A distribution report measures the delivery of your message, never its reception.
Coverage is when somebody else decides you are interesting. A release is when you pay for the paper. Only one of those is for sale.
What does a wire release actually buy?
Syndicated reprints of your own text, a permanent dated record, and a report. The prices are public, and they are not trivial for a bootstrapper. As of September 2026, EIN Presswire lists $149 for a single release and $999 for a bundle of fifteen. eReleases, which distributes through PR Newswire, lists packages at $399, $499 and $699, with surcharges per extra hundred words and per additional industry target list. Writing help costs more again. Check current pricing before you buy, because these move.
What you should not buy it for is search. Google's spam policies list, as an example of link spam, “links with optimized anchor text in articles, guest posts, or press releases” distributed on other sites. Read that as written. It does not say releases are forbidden. It says the keyword-rich links inside one are precisely what the policy is aimed at, which removes the entire SEO argument wire vendors have been making for fifteen years.
The same document contains a detail almost nobody quotes, and it cuts the other way. Where Google describes site reputation abuse, the practice of parasitic third-party content riding on a strong domain's rankings, it explicitly exempts wire service and press release service sites from that policy. Syndicated releases are not treated as an abuse of the host site. They are understood for what they are: a notice board. Nobody is penalized, and nobody is impressed either.
How does it compare to the alternatives?
Every announcement channel trades money, time and control differently, and the wire is the only one where you spend the most money for the least control over the outcome. Line them up honestly and the choice usually makes itself. The costs below are what each move takes out of a solo founder's week, not an agency's retainer.
| The move | What it costs | What you actually get | When it fits |
|---|---|---|---|
| Paid wire releaseEIN Presswire, eReleases, PR Newswire resellers | Roughly $149 to $699 per release, as of September 2026 | Syndicated copies of your own words on aggregator sites, a distribution report, and a permanent dated record that your company said a thing | Funding news, regulated disclosures, and anything an investor or a lawyer expects to find on a wire |
| Story-first pitch to a named writerOne reporter, one angle | Free. Two to four hours of research and writing per pitch | A real chance at an article written by somebody else, with a real audience and an editorial link | You have a genuinely new fact, a number nobody else has, or a story that is interesting without your product in it |
| Source request platformsHARO, Qwoted, Source of Sources | Free to low. Cap it at twenty minutes a day | Quotes inside other people's articles, attributed to you by name and title | You know one narrow thing better than most people and can prove it in three sentences |
| Your own announcement postBlog, changelog, email list | Free. An afternoon | The canonical version of the news on a domain you own, indexable and linkable forever | Always. This is the page every other channel should point at |
| Launch directoriesProduct Hunt, Peerlist, niche lists | Free to modest, depending on the venue | Listings, profile pages and referral traffic from people actively browsing for new products | Any launch or relaunch, because the audience arrives already curious |
| Peer coverageFounders writing about founders | Points, earned by helping somebody else first | A real article on a real site, verified before anything settles, read by people who are also building | You want the outcome a release only implies, without the lottery attached to it |
The row that surprises founders is the fourth one. Your own announcement post is free, permanent, entirely under your control, and it is the asset every other row links back to. Skipping it in favour of a paid release is renting a billboard that points at an empty lot. That is the same portfolio logic behind marketing a SaaS on no budget: spend first on the things you keep.
When is a release genuinely the right call?
When somebody other than you will go looking for a wire record. That is the whole test, and it is satisfied in three situations.
- Funding, acquisition and corporate events. Investors, acquirers, analysts and business journalists genuinely do check the wire for these. A funding announcement with no wire record looks odd to the people whose job is to notice. This is the strongest case by a distance, and it is also the case where your investors will often handle the release for you.
- Compliance and disclosure. If a regulator, exchange, grant body or contract requires public notice on a specified timeline, the wire is the mechanism that exists for exactly that. You are not marketing. You are filing.
- You already have an audience that reacts. A company with existing press relationships uses a release as a coordination artifact: the reporters were briefed days ago, and the release is the timestamp everybody publishes against. That is a real workflow. It presumes the relationships exist first, which is the part an early startup does not have yet.
Notice what is absent from that list: launching a product, hitting a user milestone, shipping a feature, rebranding, or wanting to look established. Those are the reasons founders actually buy releases, and none of them pass the test.
Do press releases help with AI search?
This is the one modern argument for them that is not nonsense, and it is still weaker than the people making it suggest. The argument goes like this. Language models and knowledge graphs need to resolve your company as an entity, a consistent thing with a name, a description, a founding date and a category. Dated third-party pages stating those facts identically help that resolution. A syndicated release is, technically, a lot of dated third-party pages stating your facts identically.
The entity sidebar
If entity clarity is what you are after, buy it directly instead of renting it. Publish Organization structured data on your own site with your legal name, logo, founding date and social profiles, keep that description byte-identical everywhere it appears, and make sure the same facts show up on the profile pages other people control. That is the mechanism a release only approximates, it costs nothing, and it does not expire.
The reason to stay sceptical of the release version is where citations actually come from. The 5W State of AI Citations 2026 analysis found that engines draw on very different source pools, with ChatGPT leaning heavily on Wikipedia, Reddit and review sites, and Perplexity favouring attributed, named-author content. Wire aggregator pages are not what those systems reach for. Reviews, community threads, profile pages and articles with a human byline are.
Keep the traffic side in perspective too. AI referrals are growing fast in relative terms, with 1.13 billion referrals to the top 1,000 sites in June 2025, up 357% year over year, the large majority of that from ChatGPT. In absolute terms it is still a small slice of how people arrive anywhere. Being cited is worth engineering for. Paying a wire in the hope of being cited is not.
What earns real coverage instead?
A story that is interesting without your product in it, sent to one named person who covers that beat. Michael Seibel's guidance on getting press for your startup in the Y Combinator library makes the point plainly: work out the narrative first, build a small list of writers who cover it, and approach them individually. Nothing in that process involves a wire.
The practical version for a solo founder is smaller than it sounds. You need one fact nobody else has, and it is usually your own data: what happened when you changed pricing, what a hundred user interviews revealed about a category, how much a thing genuinely costs to run. Writers need material, and a founder with a spreadsheet is more useful to them than a founder with an adjective.
The two channels that convert that material into published words are the ones I have written up separately: pitching against live journalist requests, covered in the current map of HARO alternatives, and the slower authority work in how to get backlinks for a startup. Both cost time instead of money, and both produce the thing a release cannot: somebody else's byline over your name.
What is the DIY announcement stack?
Five steps, one day of work, and it has outperformed a wire release on every launch I have run. Run them in this order, because each one feeds the next.
Publish the canonical post first
One page on your own domain with the full announcement: what changed, who it is for, what it costs, and a screenshot that shows the thing working. Every other channel links here. Skip this and you are asking journalists and directories to link to a homepage that does not mention the news.
Tell the people who already said yes
Email the list, message the beta users, post in the two communities where you are a known member rather than a stranger. This is the only part of an announcement with a reliable response rate, and it costs nothing.
Submit to the venues where browsing is the point
Launch directories exist because people go there specifically to find new products. A listing beats a syndicated release for the same reason a shop beats a billboard: intent. Pick five that fit your category and write each submission separately.
Pitch two writers, not two hundred
Find the two people who have covered your exact category in the last ninety days. Read what they wrote. Send a short note about the story rather than about you. Two researched pitches beat a blast to a purchased media list every time.
Turn the announcement into evidence
A month later the news is not news, but the artifacts remain: the post, the listings, the quotes, the profile pages. Those are what a cold visitor and an answer engine both read when they try to work out whether you are real.
There is a sixth option that behaves like the thing a release only pretends to be. On Favors.dev, a founder marketing co-op, the featured-article action in the favors queue is another founder writing a real article about your product on a site that publishes, with the placement verified before any points move. It is not a national outlet and I would not pretend otherwise. It is an article carrying a byline that is not yours, which is more than $399 on a wire has ever bought me.
How do you write one that is not embarrassing?
If you have decided you need one anyway, the fix is to write it as news rather than as an advertisement wearing a suit. Most bad releases fail in the same four places.
- The headline states the news, not the benefit. Company X launches Y for Z. That is it. Every adjective you add moves it further from something a desk would run and closer to something a reader skips.
- The first paragraph survives alone. Assume nobody reads the second. Who, what, when, and why anyone outside your company should care, in about forty words.
- The quote says something a human would say. The standard founder quote is three clauses of nothing. Replace it with one concrete sentence about why you built the thing, ideally carrying a number or a specific customer problem. If you would be embarrassed to say it out loud, cut it.
- The links are plain and few. One to the canonical announcement on your own site, one to your homepage. No keyword-stuffed anchors, for the reason quoted further up this page.
Then set your expectations at the right level. The release is a record. The work of convincing anyone still happens on your own pages, in other people's articles, and in the places your buyers already spend their time.
Frequently asked questions
Are press releases worth it for startups?
For most early startups, no. A paid wire release distributes your own words to aggregator sites and news feeds, which is not the same as a journalist deciding your company is worth writing about. It reliably produces syndicated reprints and a distribution report, and it does not reliably produce coverage, customers or ranking improvements. It becomes worth it in a small set of cases: a funding round or acquisition where investors and press expect a wire record, a regulated or compliance-driven disclosure, or an established company with an audience already primed to react to the news.
How much does a startup press release cost?
Published package prices from the well-known distributors sit between roughly $149 and $699 per release as of September 2026. EIN Presswire lists $149 for a single release and $999 for a bundle of fifteen, while eReleases, which distributes through PR Newswire, lists packages at $399, $499 and $699 with per-word and per-target-list surcharges on top. Writing services, additional industry lists and extra words all add cost, so the sticker price is usually the floor rather than the total.
Do press release links help SEO?
No, and chasing them can hurt. Google's spam policies name links with optimized anchor text in press releases distributed on other sites as an example of link spam, alongside guest posts and advertorials. The practical read is that the links inside a syndicated release are not a ranking asset, and building a release specifically to place keyword-rich anchors is exactly the pattern the policy describes. The mention itself can still have value as a dated public record. The link is not the reason to do it.
What should a startup do instead of a press release?
Publish the canonical announcement on your own site, tell the people who already opted in, submit to the launch venues where people browse for new products on purpose, and pitch two researched writers rather than a bought media list. That stack costs a day of work and produces assets you keep. If you want the guaranteed-coverage outcome a release only implies, arrange it with a peer instead of a wire, where the article is written by a person with an audience and verified before anything changes hands.
When should a startup send a press release?
Send one when somebody other than you will go looking for a wire record. Funding announcements, acquisitions, executive appointments at a company of real size, and legally required disclosures all belong on a wire, because investors, acquirers and compliance teams expect to find them there. A product launch by an unknown company almost never belongs on one, because nobody is searching the wire for it and the release does none of the work of convincing a reader that the product is good.