By making the first two sentences prove the story belongs on that specific reporter's beat, and by offering something they cannot get anywhere else: data they can check, a person they can interview, or first access to the news. Most pitches are not rejected for bad writing. They are discarded for being sent to the wrong person, and the two largest 2026 surveys of working journalists both say so with numbers.
Why do most pitches get deleted before anyone judges the writing?
Because they are off-beat, and journalists treat that as disqualifying rather than as a minor annoyance. Muck Rack's State of Journalism 2026, published on 19 March 2026 and based on 897 analyzed responses from 1,044 journalists surveyed between 30 January and 2 March 2026, found that 88 percent of journalists immediately disregard a pitch that misses their beat. Nearly half said they seldom receive pitches that are relevant to them at all.
Cision's 2026 State of the Media Report, released on 13 May 2026 from a survey of 1,899 journalists across 19 markets in North America, EMEA and APAC, puts the same problem from the other direction: 72 percent of journalists say fewer than a quarter of the pitches they receive are relevant.
Read those two findings together. Journalists are not drowning in pitches they have to choose between. They are drowning in pitches that were never for them, and the small number that are relevant face much less competition than the raw volume suggests. That is the opportunity, and it is a targeting problem, not a copywriting problem.
Does PR even matter if journalists complain about pitches this much?
Yes, and more than most founders assume. Cision found that 66 percent of journalists rely on PR-provided content, meaning press releases, pitches and media kits, for story ideas, which made PR the leading source of story leads in that survey. Muck Rack found that 86 percent of journalists said PR pitches inspire at least some of their stories.
So the channel works, and it is not the only door. When you want to make an argument rather than offer a news story, the route is writing an op-ed and offering it to one outlet, which runs on different rules but rewards the same discipline about knowing the publication. Newsrooms are running on fewer people than they used to: 49 percent of the journalists Cision surveyed named shrinking budgets, staff cuts and heavier workloads as their biggest obstacle. A reporter covering a beat that used to have three people on it needs usable material. A pitch that arrives as a finished, checkable story idea is help. A pitch that arrives as homework is a deletion.
What belongs in the first two sentences?
The beat connection and the news, in that order, with nothing in front of them. No "I hope this email finds you well." No paragraph about your company's mission. A reporter decides in a few seconds whether this is theirs, and anything that delays that decision is working against you.
A usable opening names what the reporter covers, states what happened or what you found, and makes clear why it is happening now. If you cannot write that in two sentences, the problem is usually that there is no story yet, which is worth discovering before you send 200 emails rather than after.
Keep the whole thing short. Muck Rack's own guidance, drawn from its survey data, reports that 65 percent of journalists prefer pitches under 200 words, and quotes one journalist's blunter version of the same point: journalists seldom read past 200 words. Two hundred words is roughly four short paragraphs. If your pitch needs more than that, what it needs is a link to a page holding the detail, not more email.
Put the subject line to the same test. It should read like a headline a reader of that publication would recognize, not like a company announcement. "Our new platform transforms compliance" is a sentence no journalist has ever written. "UAE fintechs are failing onboarding checks at three times the regional rate" is.
What makes a pitch worth answering rather than just readable?
Something the reporter cannot get from a competitor or generate themselves. Cision's report names what actually works in practice: original research, expert access and embargoed information. Muck Rack's respondents named the same things, with clear beat relevance first, then access to relevant sources, then original data or research.
That list is short and it is specific, so treat it as a checklist:
- Original data. Numbers from your own operations, survey or product that nobody else has. This is the strongest asset a company has and the most under-used, because it requires deciding to publish something real. If you are sitting on usable data, announcing it properly as research is usually worth more than three product announcements.
- A person worth interviewing. Not "our CEO is available for comment," but a named person with first-hand experience of the thing the reporter is writing about, who will say something specific on the record. This is the mechanism behind almost every time a company gets quoted in media coverage it did not generate itself.
- First access. An embargo or exclusive gives a reporter a reason to prioritize your story over the five others on their desk.
- Ready assets. High-resolution images, a factsheet, correct name spellings and titles. These do not win coverage on their own, but missing them has lost coverage that was already agreed.
Notice what is not on that list: a press release written in the third person with three adjectives per noun. Releases have a role, mostly as a reference document and a record, and we have written about when a press release is and is not worth the money. They are rarely the reason a journalist replies.
Should you use AI to write the pitch?
Not to write the pitch itself. Journalists are using AI heavily in their own work: Muck Rack found 82 percent use at least one AI tool, up from 77 percent the year before, with ChatGPT at 47 percent, Gemini at 22 percent and Claude at 12 percent. That has not made them receptive to AI-written outreach. Cision found that 53 percent of journalists oppose AI-generated pitches, citing accuracy and personalization.
The reason is practical rather than ideological. An AI-written pitch reads as though it could have been sent to 400 people, because it usually was, and a pitch that could have been sent to anyone fails the relevance test that 88 percent of journalists apply first. Use AI for the parts that are genuinely research: finding who covers what, reading a reporter's recent work, checking whether your claim holds up. Write the three paragraphs that go in the email yourself.
The same caution applies to accuracy. Twenty-six percent of Muck Rack's respondents named unchecked AI as a concern, up from 18 percent a year earlier, and accuracy and misinformation were the top newsroom challenge in Cision's survey. Sending a journalist a number you have not verified at its original source is the fastest way to be permanently ignored.
How do you find the right journalist in the first place?
By reading what they have published in the last two or three months, not by buying a list. A media database tells you a reporter's title. Their last ten bylines tell you what they actually cover, what angle they take and whether they have already written the story you are about to pitch. That is twenty minutes per reporter, which is why most people skip it, which is why 72 percent of pitches miss.
LinkedIn is the one platform worth working alongside email. Cision found 62 percent of journalists use it professionally and 33 percent rank it as the single most valuable platform for their work, and Muck Rack found it to be the most trusted social platform among journalists at 58 percent. That does not mean pitching in the DMs. It means a reporter who has seen a founder posting something substantive in their field recognizes the name when the email arrives.
Keep the list small and keep records, which is the whole method behind building a media list journalists actually respond to. Fifteen reporters you have genuinely read will outperform 300 addresses every time, and the coverage that results is earned media rather than paid placement, which is why it carries weight with readers and with the AI systems that now summarize the news. This matters even more when you are pitching coverage in a trade publication, where the reporter is a specialist who can tell in a sentence whether you actually read the beat.
The direct answer, for anyone who read this far
Fix targeting before you fix wording. Pick a short list of journalists whose recent work you have actually read, open with the beat connection and the news in two sentences, and attach one thing they cannot get elsewhere: your own data, a named person who will talk, or first access. Write it yourself, verify every number at its original source, and follow up once. If you want help building that list and the data story behind it, tell us what you are announcing.
