Pew Research Center’s analysis of newsroom employment found that U.S. newsroom jobs fell by roughly a quarter between 2008 and 2020, with newspapers absorbing the worst of it. That single number should reset how you think about pitching.
Fewer reporters are covering more territory. The person who once had a beat now has three. Their inbox volume did not shrink when their newsroom did, and the calculus they apply to an unknown sender got harsher in proportion. This is why the instinct to build a bigger list is exactly backward. When each recipient has less time, the only thing that buys attention is evidence that you read their work before you wrote to them, and you cannot fake that across 500 names.
So the goal is not a media list. The goal is a small, verified, defensible list where every row survives a test. Here is how to build a media list that produces replies, column by column and source by source.
Why the 500-name list keeps failing
The big list fails for a structural reason, not a tactical one.
When you build a media list at scale, size dictates everything downstream, and it dictates it in the wrong direction.
A list of 500 forces generic copy. You cannot write 500 personalized pitches, so you write one and merge a first name into it. Reporters recognize merged copy on sight, because they receive dozens a day, and the pattern is unmistakable: a compliment about their work that could apply to anyone, a paragraph of company description, a quote from a founder nobody asked about. The list size determined the copy, and the copy determined the outcome.
The big list also poisons your sending reputation. Purchased and scraped databases carry dead addresses, role accounts, and spam traps. Bounce enough and your domain reputation degrades, which means the twelve pitches you actually cared about land in a promotions tab.
And a big list hides the diagnosis. If you send 500 pitches and get one reply, you learn nothing. Was the story weak? Was the targeting wrong? Was the subject line bad? With fifteen well-targeted pitches and two replies, you can read each non-response against what you know about that reporter and adjust. Small lists are instruments. Big lists are noise generators.
There is one legitimate use for a large list, and it is announcement distribution rather than pitching. Wire distribution and a broad “for your awareness” send have their place for funding news and public company disclosures. That is a separate function from earning coverage, and conflating the two is how PR budgets get spent on volume that produces nothing.
What belongs in the 12 columns?
The spreadsheet does the thinking, so the columns matter.
Outlet name, and outlet tier in your own judgment: national, trade, regional, or creator-owned. Tier is not prestige ranking, it is a targeting label that tells you what angle each contact needs.
Reporter name, spelled the way they spell it, including accents and middle initials. Getting a name wrong in a first email is unrecoverable.
Title and stated beat, copied from their own bio rather than from a database field. Reporters describe their own beat with more precision than any third party does.
Email address, with a source note recording where you found it.
Secondary channel: their handle on whichever platform they actually use, their LinkedIn, or their newsletter. This is your fallback and sometimes your better first move.
Most recent relevant story: the URL. Not their most recent story overall. The most recent one that touches your topic.
Date of that story. This column is the one that makes the whole list honest, and I will come back to it.
The angle they took, in your own words, in under fifteen words. “Skeptical of AI vendor claims, focuses on procurement cost.” That note is what your pitch has to answer.
Publishing cadence and any known deadline pattern. A weekly columnist and a daily breaking-news reporter need different timing and different pitches.
Pitch preference notes, harvested from their bio, their newsletter footer, or their public statements about pitching. Many reporters state their preferences plainly and almost nobody reads them.
Prior relationship: have you or anyone at your company ever spoken to this person, been quoted by them, or met them at an event. Fill this in by asking your team, not by guessing.
Status and last contacted date, which turns the list from a research document into a working system.
Twelve columns. If a row has blanks in the story URL, the date, or the angle column, that row is not finished and should not receive a pitch.
Where do you actually find the names?
Sourcing order matters, because the first pass shapes everything after it.
Start with the coverage that already exists. Search for the last twelve months of stories about your exact topic, your named competitors, and the problem you solve. Read them. The bylines on those stories are your first candidates, and they are better candidates than any database query because you have proof of interest in hand before you record the name.
Then work the citation trail. Every good story quotes sources. Find the analysts, academics, and operators who get quoted in your category, then find who quotes them. Reporters return to the same expert pool, and the pool tells you who is writing the stories.
Third, work trade press hard. Trade publications are undervalued by founders who want a national tech outlet and overvalued by nobody. Trade reporters know the category, they have space to run a substantive piece, and their coverage gets read by your actual buyers. They also get pitched far less, which means a specific, informed pitch has a much better chance of landing.
Fourth, add creator-owned media. Independent newsletters, podcasts, and video channels covering your category are now a real tier. Many were staff reporters two years ago and took their audience with them. They tend to publish faster, they are easier to reach, and their archives get read for years. Their footers usually state exactly how they want to be pitched.
Fifth, check the outlets’ own staff pages and newsletter mastheads for the editors above your reporters. You do not usually pitch the editor first, but you need to know who they are and what section owns your topic.
Sixth, mine the competitor coverage you would have killed for. Take the three stories about your competitors that you were most jealous of and find every person involved: the reporter, the editor of that section, and any freelancer credited. Freelancers are the most underused contacts in press outreach. They pitch multiple outlets, they are hungry for angles, and a freelancer who likes your story becomes an advocate inside newsrooms you cannot reach on your own. Their contact details are usually on a personal site, along with a list of where they have published.
A note on sourcing platforms. Cision folded the old HARO service into Connectively and then shut that service down at the end of 2024, which left a gap that a number of alternatives moved into. Query services are a useful supplement for reactive commentary, but they are not list building. They put you in front of a reporter who already knows what story they are writing. Building a media list is the proactive half of the job, and it is the half that lets you originate a story instead of decorating someone else’s.
Apply the Proof-of-Beat Test
Here is the rule that keeps a list honest, and the one thing I would ask any agency to prove before I paid them for outreach. Call it the Proof-of-Beat Test.
A reporter earns a row only if they published a story on your specific topic within the last 90 days, and you have the URL and the date in the spreadsheet.
Not their outlet covers your topic. Not they used to cover it. Not their title contains a relevant word. They published, on this topic, recently, and you can point at it.
The test kills most of the names people want to include, which is the point. It removes the reporter who moved to a different beat in March. It removes the outlet-level guessing where you pitch a general assignment desk and hope. It removes the aspirational national name who has never written a sentence about your category and is not going to start with your funding announcement.
Ninety days is the working window for most categories. Stretch it to 180 for slow-moving beats like regulatory or scientific coverage, where a reporter may publish quarterly. Tighten it to 45 for fast categories where beats shift constantly. Whatever window you pick, write it at the top of the sheet and apply it without exceptions, because the exceptions are always the names someone in leadership wants included for reasons that have nothing to do with coverage.
The second half of the test is the angle column. Once a reporter passes, read their piece and write down the position they took. If a reporter published a piece questioning whether your category delivers on its claims, a pitch that assumes enthusiasm will fail. A pitch that opens by acknowledging their skepticism and offering data has a real chance. That is what the angle column buys you, and it is why the test requires reading rather than counting.
Verify the contact, not just the byline
A verified byline with an unverified address is half a row.
Work the sources in order. Outlet staff pages often list addresses outright. Byline pages sometimes carry a contact link. Reporters’ social bios and personal sites frequently include an email, occasionally with pitching instructions attached. Newsletter footers almost always do. If none of those produce an address, most outlets use a consistent format that you can infer from any published staff address at the same domain.
Run every address through a validation tool before you send. On a list of fifteen, one bounce is seven percent of your campaign gone, and hard bounces accumulate against your domain reputation in ways that hurt every future send.
Record where you found each address in the source-note column. When the list gets rebuilt in six months, that note tells you whether to recheck a staff page or a personal site, and it tells whoever inherits the list that the contact was sourced rather than purchased.
Where an outlet publishes a tip line or a submission form and states that it is the preferred route, use it, and note that in the pitch preference column. Some newsrooms route form submissions to the desk that actually handles them, which beats guessing at an individual inbox. Ignoring a stated preference to email someone directly is a small act of disrespect that costs you the read.
Two hygiene rules that are not optional. Never put a journalist into a marketing automation sequence with tracking pixels and drip logic. Reporters can tell, several have written about it, and the resulting story is not the one you wanted. And keep a suppression list of everyone who has asked to be removed, permanently, across every campaign and every tool.
How many names is enough?
Between twelve and twenty-five for a single announcement in a single category.
That number surprises people who came from demand generation, where volume math works. Press does not run on volume math. It runs on fit, and fit collapses fast as the list grows. Rows twenty-six through five hundred are, by construction, the reporters who fit your story less well than the first twenty-five.
If you cannot find twelve reporters who pass the Proof-of-Beat Test, you have learned something important and it is not about the list. Either your topic is not being covered right now, in which case the story needs a different frame that connects it to what is being covered, or you are searching with your vocabulary instead of theirs. Both are fixable, and both are cheaper to fix before you send than after.
If you find more than forty, split the list by angle rather than pitching all forty the same thing. National business reporters need the market story. Trade reporters need the operational detail. Regional reporters need the local employment or community angle. Same news, three different first paragraphs, three separate sends.
Keep it alive or throw it away
A media list decays faster than any other asset in marketing. Reporters move outlets, change beats, go freelance, get laid off, start newsletters, and take parental leave. A list built in January and used in June is a list of guesses.
Put a last-verified date on every row and reverify quarterly. Verification means checking that the person still holds that beat at that outlet and that their most recent relevant story is still recent. It takes about an hour for twenty rows.
Log every interaction in the status column, including the ones that went nowhere. A reporter who replied “not for me, but send me anything on the regulatory side” has handed you a targeting instruction worth more than ten new names. That kind of intelligence lives in the list or it lives nowhere.
Watch for the moves. Reporters announce beat changes and new jobs on whichever social platform they live on, and following the twenty-five people on your list is a cheaper maintenance system than any quarterly audit. A reporter who just started covering your category at a bigger outlet is the warmest contact you will ever have, because they need sources fast and they remember who helped them early.
Add a dormant tab rather than deleting anyone. People come back to beats, and a reporter who ignored you at one outlet may be exactly right at their next one.
What a working media list looks like in practice
Fifteen rows. Every row has a name spelled correctly, a beat copied from the reporter’s own bio, a verified email with a source note, a story URL published inside the last ninety days, that story’s date, and a fifteen-word note on the angle the reporter took. Four are trade, six are national or regional business, three are independent newsletters, two are podcasts. Three rows carry a prior relationship note. Every row has a last-verified date within the quarter.
Fifteen pitches go out over two days, each one opening with a specific reference to the story in the URL column and a first paragraph written against the angle note. Replies come back from two or three, a couple more read it and file it away for later, and the rest say nothing.
That is a successful campaign, and it took about four hours of research to make possible.
Build a media list the small way: source from published coverage rather than from a database, apply the Proof-of-Beat Test with a stated window, fill all twelve columns before anyone gets an email, verify every address, keep it between twelve and twenty-five names, and reverify every quarter. The list stays short, the pitches stay specific, and the replies come from people who were always going to be interested if someone had bothered to read them first.