- HARO's ownership changed twice in under two years: Cision's paid "Connectively" rebrand (up to $149/month) proved unpopular and shut down permanently in December 2024; Featured.com acquired the brand and relaunched it free in April 2025, monetized via newsletter sponsorships instead of user fees. Qwoted and Source of Sources (SOS, built by HARO's original creator) are worth using alongside it in 2026.
- Propel's behavioral data (roughly 400,000 pitches, ~4,000 reporters, Q1 2024): reporters open about 46% of pitches but respond to only about 3.43% overall: a realistic program runs 30-50 pitches a week across a few platforms, not a single high-effort pitch and a wait.
- The same Propel dataset found pitches under 150 words earning a 5.89% response rate versus just 1.46% for pitches over 500 words: roughly a 4x gap driven by length alone.
- Digital Third Coast's analysis of 450+ media pitch subject lines found the best-performing length clustering around 60 characters, and question-based subject lines scoring as the least effective of five phrasing styles tested.
- This piece assumes the material already exists. If it doesn't yet, how to actually earn a citation from a journalist covers what has to be true before any of the tactics below are worth using.
How to actually earn a citation from a journalist covers the material a pitch needs behind it: original data, a specific angle, a disclosed methodology. This is the other half: the actual mechanics of getting that material in front of a journalist, once it exists. Format, timing, where the requests live now, and what the disclosed response-rate data says actually gets a reply.
Where do journalist requests actually live now?
HARO's ownership has changed twice in a short window, which is worth knowing before building a workflow around outdated advice. Cision rebranded HARO as "Connectively" and introduced paid tiers running up to $149 a month (a change the PR community broadly resisted) before shutting the platform down permanently on December 9, 2024. Featured.com then acquired the HARO brand and relaunched it on April 22, 2025: free again for both journalists and sources, monetized through newsletter sponsorships rather than user fees. Three platforms make up the practical 2026 picture:
- HARO (via Featured.com): back under its original free model as of the April 2025 relaunch, worth checking directly rather than assuming the intervening paid "Connectively" era still applies.
- Qwoted: a searchable database plus a steady feed of media requests, skewing toward finance, technology, healthcare, and business reporters at established outlets. Generally described as a smaller, higher-quality pool of requests than the old HARO firehose, fewer opportunities, but from more credible publications.
- Source of Sources (SOS): built by Peter Shankman, HARO's original creator, launched immediately after the 2024 shutdown. Free, structured as a simple email newsletter, sending up to three daily digests of journalist requests.
Running a pitch program that only checks one of these misses requests the others surface: the fragmentation since HARO's 2024 shutdown means the single-platform habit that used to work well no longer covers the same ground.
What response rate should you actually plan around?
Treating pitching as a numbers game from the outset avoids the common failure mode of writing one exceptional pitch, sending it once, and concluding the whole channel doesn't work. Propel's behavioral analysis of roughly 400,000 pitches sent to about 4,000 reporters in Q1 2024 (aggregated into the 2026 PR Pitch Response Rate Study alongside other datasets) found reporters opening about 46% of pitches, but actually responding to only about 3.43% overall. That's a large, normal gap between a journalist seeing a pitch and a journalist acting on it, not a sign of a bad pitch specifically. One caveat: the underlying behavioral data is from Q1 2024, the most recent disclosed figure of its kind found for this piece, so treat the exact percentage as directional rather than a live, constantly-updated number.
The practical implication: a realistic program sends a consistent volume (commonly cited as 30-50 pitches a week spread across a few of the platforms above) rather than treating each individual pitch as a make-or-break event. A few quality placements a month, from that volume, is a realistic outcome; expecting a high reply rate from a handful of pitches sent sporadically sets up a false read on whether the material or the approach is the problem.
What format does the data say actually works?
Subject line. Digital Third Coast's analysis of 450+ media pitch subject lines found the best-performing length clustering around 60 characters, and (a specific, counterintuitive finding worth knowing) subject lines phrased as a question scored as the least effective of five phrasing styles tested, despite being a common instinct for grabbing attention. The subject line isn't a label for the pitch; it's the first and sometimes only signal a journalist uses to decide whether to open it at all, so it should state the strongest, most specific angle as a direct statement rather than a tease or a question.
Length. Propel's behavioral dataset found pitches under 150 words earning a 5.89% response rate, against just 1.46% for pitches over 500 words: roughly a 4x difference driven by length alone. That's short enough that a pitch has room for exactly one angle, one or two supporting numbers, and one ask, not enough room to also justify why the topic matters in general terms, provide company background, and make a secondary ask. Cutting a pitch to fit inside that length usually means cutting the throat-clearing, not the actual evidence.
Structure. The pitches reported to perform best follow a consistent shape: the specific angle stated first (not the company, not a general topic), the evidence that makes the angle credible (the original number, the disclosed sample), and a single, clear ask (a quote, an interview, first access to the data), in that order, with nothing extraneous between them.
One ask, not several. A pitch asking a journalist to consider a story angle and schedule a call and review an attached press kit is asking for more decisions than a busy inbox will make. One specific, low-friction ask (usually "happy to send the full dataset" or "available for a 15-minute call this week") is easier to act on than an open-ended invitation to explore.
What still matters more than any of this?
None of the format guidance above rescues a pitch built around thin or unoriginal material. How to actually earn a citation from a journalist covers what has to be true first: original data, a disclosed methodology, an angle that updates rather than restates an existing narrative. A perfectly-formatted, perfectly-timed pitch for a recycled stat or an unsourced claim still gets ignored; format determines whether a good pitch gets read, not whether a weak one gets forgiven.
Frequently asked questions
Is HARO still the main way to find journalist requests?
Not exactly: HARO (rebranded "Connectively" under Cision, with paid tiers up to $149/month) shut down permanently in December 2024. Featured.com acquired the HARO brand and relaunched it free in April 2025, monetized through newsletter sponsorships rather than user fees. Qwoted and Source of Sources (SOS, launched by HARO's original creator Peter Shankman right after the shutdown) are worth checking alongside the relaunched HARO/Featured for the same kind of journalist-request access.
What response rate should I actually expect from cold pitching?
Low, and treating it as low from the start avoids wasted effort chasing a fantasy number. Propel's behavioral analysis of roughly 400,000 pitches sent to about 4,000 reporters (Q1 2024, published in the 2026 PR Pitch Response Rate Study) found reporters opening about 46% of pitches but responding to only 3.43% overall: meaning a consistent program of 30-50 pitches a week across a few platforms, not a single perfect pitch, is what a realistic program looks like.
How long should a pitch actually be?
Short. The same Propel dataset found pitches under 150 words earning a 5.89% response rate, against just 1.46% for pitches over 500 words: roughly a 4x difference by length alone. A pitch that states the angle, the evidence, and one clear ask (in that order) outperforms a longer, more thoroughly-justified version most of the time.
What's the one thing a pitch needs before format even matters?
The actual material: original data, a specific angle that updates or complicates an existing narrative, and a disclosed methodology behind whatever number is being pitched. Format and timing only matter once that exists; a well-formatted pitch built around a recycled or unsourced claim still gets ignored. The material side of this is covered in full in how to actually earn a citation from a journalist.
Sources referenced
- Muck Rack: 2026 State of Journalism Report (Jan-Mar 2026, n=1,044 surveyed / 897 analyzed)
- EPR Research & 5W: The PR Pitch Response Rate Study 2026 (aggregates Propel's behavioral analysis of ~400,000 pitches, ~4,000 reporters, Q1 2024)
- Digital Third Coast: How to Write Media Pitch Subject Lines — 450+ Analyzed
- Originality.AI: Featured Acquires HARO — What This Means for Content (Dec 2024 shutdown, Apr 2025 relaunch timeline)
- Source of Sources (SOS) (free journalist-request newsletter, founded by HARO creator Peter Shankman)
This piece is the execution layer paired with how to actually earn a citation from a journalist, which covers the material a pitch needs before format or timing matter at all. Both build on the earned-media case made in why earned media outperforms owned content in AI citations.
