A software company gets better launch coverage from customer proof, original data and expert commentary than from a press release. Reporters are overloaded with pitches, most of which they consider irrelevant, so the launch that works gives a journalist something to use: a named customer result, a number nobody else has, or a clear opinion about where the market is heading. The release can still exist, but as a record, not as the plan.
Why does a launch press release rarely work for software?
A launch press release rarely works for software because it announces a product, and journalists cover problems, trends and people. A release that says a company has launched version four of a platform gives a reporter no reason to care unless the reporter already covers that product. The reporter needs a reason tied to the audience.
The survey evidence is blunt. According to Muck Rack's State of Journalism 2026 survey, as reported by PR Daily, 47 percent of journalists said PR pitches were seldom or never relevant to their work, and 51 percent said only 1 to 10 percent of their published stories began with a pitch. In the 2024 edition, 79 percent of respondents named lack of relevance as a top reason for rejecting a pitch. The 2025 edition, as cited in a PRSA student piece, found that 86 percent of journalists will disregard a pitch that is irrelevant to their coverage area.
The same coverage reports that journalists still value PR professionals: 73 percent said PR professionals are at least moderately important to their success. That is the encouraging part. Reporters want help, but they want help with stories, not announcements. A software launch that treats the release as the main event is asking them for the wrong thing.
These are survey results and secondary reporting, so I treat them as direction and not as precise measurement. The direction is consistent across years: relevance decides almost everything.
What do journalists say they actually want?
Journalists say they want original data, access to an expert, an embargoed announcement or compelling visuals. That list comes from Cision's 2026 State of the Media Report, as summarized by PR Daily. It is a useful checklist for a software launch because each item can be built from the product and the team.
- Original data: aggregate and anonymized findings from product usage, or a survey of customers, that tell the reader something about their industry.
- Expert access: a founder, engineer or customer who can explain what is changing and why, without a sales pitch.
- An embargoed announcement: a chance to prepare a story before the news goes public.
- Visuals: a clear demo, chart or product image that makes the story easy to show.
Muck Rack's findings on format point the same way. The 2025 survey, as cited by PRSA, reported that 69 percent of journalists prefer pitches under 200 words and 62 percent prefer one-to-one email. A launch plan should therefore produce short, specific pitches, each with its own angle, not one long release sent to everyone.
What should a software company do instead of a release?
A software company should build the launch around five assets: a named customer result, original data, a point of view, a prepared briefing and a launch page that machines and people can read. Each asset does a different job, and together they give a reporter several reasons to cover the story.
What is a named customer result?
A named customer result is a specific outcome a real customer has agreed to share, such as time saved, error rates reduced or a process replaced. It is the strongest asset in software PR because it turns a feature into evidence. I would rather have one customer who will speak on the record than ten anonymous testimonials. Ask for permission early, agree what numbers can be used and let the customer review the wording. Reporters will call the customer, so the customer must be willing.
What is original data in a software launch?
Original data is information the company holds that nobody else can publish. It might be anonymized patterns across customers, such as how long a common process takes, or the result of a short survey. The data should answer a question the audience already has. A finding about how teams actually use a tool tells a reporter something about the industry, not only about the product. Keep the method visible, state the sample size, and say what the data cannot tell you.
What is a point of view?
A point of view is a clear opinion from a named person about where the market is going. It gives a reporter something to quote and something to argue with. The best launch commentary connects the product to a trend without claiming the product is the trend. A founder who can say, in two sentences, what is changing in the industry and why it matters gives a reporter a reason to call back later, long after the launch.
What is a prepared briefing?
A prepared briefing is a short, scheduled conversation with a small number of reporters or analysts, held before the public announcement, in which the company walks through the story and the evidence. It works best with a deck of a few slides, a live or recorded demo, a customer on call and the data in hand. The guide to embargoed press briefings covers them in more detail.
What is a launch page that machines can read?
A launch page that machines can read is a plain, public page that states what the product does, who it is for, what it costs or how pricing works, and what has changed, in short factual sentences. It matters because buyers and journalists increasingly ask AI engines about products before they visit a website, and the engines quote clear pages. The answer engine optimization guide explains how to write pages that engines can lift, and the AI visibility guide explains how to check what engines say about a product after launch.
How should a software launch be sequenced?
A software launch should be sequenced over about six weeks, with the public announcement in the middle and the follow-up work after it. The aim is that the news date is the start of coverage, not the whole of it.
- Weeks 6 to 4 before launch: Secure the customer result, finalize the data and write the point of view. Decide which three to five outlets matter most and why.
- Weeks 4 to 2: Prepare the briefing materials and the launch page. Draft individual pitches under 200 words, each built for one reporter's beat.
- Week 2 to launch day: Hold briefings with an agreed embargo, then publish the announcement, the page and a short release as a record.
- Week 1 after launch: Follow up with reporters who showed interest, offer the customer and the data again and publish the founder's commentary on the company's own site.
- Weeks 2 to 6 after launch: Pitch the data and point of view to trade and industry outlets that were not in the first round, and check how AI engines now describe the product.
Notice that the release has moved to a supporting role. It is published on launch day as a source for anyone who needs the facts, not as the thing the team spent most of the time on.
What mistakes should software companies avoid?
Software companies should avoid four mistakes: pitching features instead of problems, sending one pitch to everyone, hiding the evidence and measuring the wrong things. Each one maps to what journalists say they dislike.
- Feature pitching: A list of capabilities gives a reporter nothing to report. Lead with the problem the product addresses and the proof that it does.
- Mass pitching: A generic pitch signals that the sender has not read the reporter's work. Relevance is the top complaint in the surveys cited above.
- Hidden evidence: Claims such as "the fastest" or "the most secure" without a method invite doubt. Show the test, the sample or the customer.
- Vanity measures: Counting impressions or pickups of the release says little. Count placements in the outlets your buyers read, the quality of customer conversations that follow, and mentions in AI answers over time.
I would add one more caution: do not announce what customers cannot yet use. A reporter who tries a product that is not ready will write about that, and the story will stay on the record.
How does this connect to the rest of a PR program?
This approach connects to the rest of a PR program because the same assets keep working after launch. The customer result becomes a case study, the data becomes a recurring report and the point of view becomes a regular byline. Over time those pieces build the independent record that both journalists and AI engines draw on. The wider guidance is in PR and SaaS businesses and technology companies PR, and the media side is covered in media relations. For sector guides across industries, see the industry PR guides.
Software companies that want help building launch assets and briefing reporters can look at 5W's technology PR practice.
Frequently asked questions
Should a software company skip the press release entirely?
No. A short release is useful as a public record and as a source for reporters and search engines. It should not be the center of the launch plan, because most journalists rarely begin a story from a pitch and most routine announcements are skipped.
What is the best asset for a software launch?
A named customer result is usually the strongest, because it turns a feature claim into evidence that a reporter can check by calling the customer.
How long should a pitch be?
Short. Muck Rack's 2025 survey, as cited by PRSA, reported that 69 percent of journalists prefer pitches under 200 words.
What do journalists want from a launch?
According to Cision's 2026 State of the Media Report, as summarized by PR Daily, they want original data, access to an expert, an embargoed announcement or compelling visuals.
How do we measure a software launch?
Track placements in the outlets your buyers read, the number and quality of inbound conversations, and how AI engines describe the product in the weeks after launch, rather than counting impressions of the release.