By starting from the stories reporters have already published rather than from a database export, and by keeping the list short enough that you can genuinely know every person on it. The two largest 2026 surveys of working journalists agree that the problem with most outreach is not the writing. It is the list. Muck Rack's 2026 State of Journalism report, based on 897 journalist responses collected between January 30 and March 2, 2026, found that 88% immediately disregard pitches that miss their beat. Cision's 2026 State of the Media report, from 1,899 journalists surveyed in January and February 2026, found that 72% say fewer than a quarter of the pitches they receive are relevant to their work.
Read those two numbers together. Roughly three in four pitches are irrelevant to the person receiving them, and nearly nine in ten of those get deleted on sight. A media list is the document that decides whether your pitch is in the relevant quarter or the irrelevant three quarters. This article is about building that document. How to write the pitch itself is a separate question.
Why do most media lists fail before the first email goes out?
Because they are built from job titles instead of from bylines. A database can tell you that someone is a technology editor at a national paper. It cannot tell you that she has written exclusively about chips and export controls for six months, has never covered a seed-stage startup, and published a skeptical piece on your exact category two weeks ago. All three of those facts decide whether she opens your email, and none of them appear in a contact record.
The same Muck Rack survey found that nearly half of journalists say they seldom receive pitches that match their coverage. That is a remarkable thing for a profession to say about an industry that exists to send it stories. It means the typical list is not slightly off. It is mostly off.
How many journalists should be on the list?
Fewer than you think, and fewer than any vendor will suggest. For a single announcement from a startup, a working list of fifteen to twenty-five named reporters is usually enough. For a standing list that covers a company's whole beat over a year, forty to sixty is a realistic ceiling for a team that intends to actually read what those people write.
The arithmetic is simple. Reading a reporter's last ten pieces takes about twenty minutes. Doing that for fifty people is roughly two working days. Doing it for three hundred is six weeks, which nobody does, which is how three hundred becomes a spray list. Cision's finding that 66% of journalists rely on PR-provided content for story ideas tells you the demand is real. Muck Rack's finding that 86% say at least some of their stories originate from PR pitches says the same thing. The reporters want the stories. They do not want to be the three hundredth address on a list built in an afternoon.
A good discipline is to tier the list. The top tier is the five to ten reporters who cover your category every week and whose coverage would actually move something for the business. The second tier covers adjacent beats where your story is a stretch but a plausible one. Anyone you cannot place in either tier should not be on the list.
Where do you find the right reporters?
In the coverage itself. Search for the last three months of articles on your category, your competitors and the problem you solve. Note who wrote them, which outlet, and what angle they took. Then read what else each of those people has published. This is slow and unglamorous and it is the entire method. A reporter who wrote about your competitor's funding round last month is a far better entry than a reporter whose title says she should care.
Trade publications in your sector often matter more than national outlets for a first-time story, and their reporters are usually easier to reach. What earned coverage from a trade title is worth, compared to a paid placement, is a question founders consistently underrate.
LinkedIn is worth adding to this research. Cision found that 62% of journalists use it professionally and 33% rank it as the single most valuable platform for their work. Muck Rack found that 58% of journalists trust LinkedIn to treat journalistic content fairly, the highest of any platform in its survey. Reporters often describe on LinkedIn what they are working on and what they want to hear about. That is not an invitation to pitch in direct messages. It is a research source for the list, and it is also why a founder who posts something substantive in their own field gets recognized when the email arrives.
Source-request platforms such as Qwoted and the Muck Rack and Cision journalist request feeds are a third input. A reporter who posts a request for sources on your topic has told you her beat, her deadline and her angle in one message. Responding to those requests is a tactic in its own right, but the names you collect from them belong on the list too.
What should each entry record?
More than an email address. A useful record for each reporter holds the outlet and beat, the last three relevant pieces with dates and a one-line note on the angle, the stories they have already covered that you should not repitch, their stated preferences if they have published any, and a log of every contact with them and what came of it.
That last column is the one most teams skip and the one that pays most. If a reporter ignored two pitches and answered a third, the record tells you what the third one had that the others did not. If she asked you to stop sending her product news, the record stops a colleague from sending her product news. A list without a contact log is a list that lets you make the same mistake twice.
Cision's report notes that journalists prioritize original research, expert access and embargoed information in pitches. Noting which of those a given reporter has used before tells you what to lead with.
How often does a media list go stale?
Faster than almost any other marketing asset, and the pace is accelerating. Pew Research Center's analysis of U.S. newsroom employment found that total newsroom jobs fell from about 114,000 in 2008 to about 85,000 in 2020, a 26% drop, with newspaper newsrooms down 57% over the same period. The people on your list are working in an industry that has been shrinking for nearly two decades.
The churn has not stopped. Press Gazette's rolling tracker of journalism job cuts, updated October 7, 2026, counted at least 3,434 journalism job cuts across the UK and US in 2025, at least 3,875 in 2024, and about 6,000 in 2023. Its 2026 entries include The Guardian ending the use of up to 315 casual staff, announced October 6, and the BBC's plan to cut about 2,000 jobs, around one in ten of its staff. Every one of those cuts changes a beat, closes a desk, or moves a reporter to a new outlet where the old address bounces.
The practical consequence is that a list built in January is unreliable by the summer. Before any campaign, re-check each top-tier entry: is the reporter still at the outlet, still on the beat, and still publishing? It takes a few minutes per person and it is the difference between a pitch that lands and one that goes to an inbox nobody reads anymore.
Sources: Muck Rack's 2026 State of Journalism report, Cision's 2026 State of the Media report, Pew Research Center's newsroom employment analysis and Press Gazette's job cuts tracker, all linked in the article.
Should you buy a media database?
A database is a starting point, not a list. The large subscription tools are good at telling you who exists and reasonably good at keeping email addresses current. They are poor at telling you what someone actually covers right now, and no database can tell you whether a reporter is already tired of your category. If you can afford one and you send to more than a handful of outlets a year, it saves time on contact details. If you cannot, the method above works without it, because the hard part was never the address.
Where a database becomes dangerous is when its export becomes the list. Pulling every contact tagged with your industry and hitting send is exactly how a reporter ends up in the 72% of pitches Cision says are irrelevant. The export is the raw material. The reading is the work.
The direct answer, for anyone who read this far
Build the list from bylines, not titles. Keep it to the reporters you have actually read, tiered by how directly they cover your category, and record every contact and its outcome so the list learns. Re-verify the top tier before every campaign, because the industry it describes is shrinking and reorganizing every month. A list of twenty people you know will outperform a list of three hundred you exported, and the coverage it produces will be earned rather than bought.
If you want a media list built this way for your sector, including the GCC outlets that rarely appear in international databases, request a quotation and we will show you how we would approach it.
