Pitch emails to journalists get opened a little under half the time and answered about three times in a hundred. Propel’s Media Barometer, which tracks pitches sent through its PR platform, reported a 46% average open rate and a 3.43% response rate across roughly 400,000 pitches in Q1 2024, after a 48.12% open rate and 2.91% response rate in Q1 2023. For general email, Mailchimp lists a 35.63% average open rate across all its users. Every one of those open figures carries an asterisk, because Apple Mail Privacy Protection has been registering “opens” that no human performed since September 2021. So the real answer is this: press release open rates look healthy on paper, replies are scarce, and the gap between the two is where most PR reporting goes wrong.
The Benchmarks: What Pitch Open Rates Look Like Across Datasets
Start with the source that publishes the most PR-specific numbers. Propel runs a quarterly Media Barometer built from behavioral data on its platform, meaning pitches its customers sent and what the recipients did with them. In the Q1 2023 edition, reported by MediaPost, the average open rate was 48.12%, up from 39.11% in Q4 2022. The response rate for the same quarter was 2.91%, below the 3.16% average for full-year 2022.
By Q1 2024 the open rate had settled at 46%, unchanged from the prior quarter, while the response rate climbed to 3.43%, which Propel called the highest it had recorded. That dataset covered about 400,000 pitches. MediaPost’s write-up added a timing detail worth remembering: 17.83% of pitches were opened within a minute and 75.37% the same day. Propel’s Q3 2023 release put the response rate at 3.33% across more than 400,000 pitches.
For context outside PR, Mailchimp’s email marketing benchmarks list a 35.63% average open rate across all users, with Business and Finance at 31.35% and Non-Profits at 40.04%. Mailchimp notes the data was last updated in December 2023 and that open accuracy may be affected by Apple’s privacy changes. Marketing newsletters and one-to-one pitches are different animals, so use the Mailchimp figure as a floor for comparison, not a target.
Notice what is missing. No major vendor publishes an audited, cross-platform benchmark for press release open rates, and the most recent Propel quarterly reports sit behind download forms. Every number here comes from one platform’s customers, who skew toward teams that pay for pitching software and probably build cleaner lists than average. Treat the 40% to 50% band as a description of that population, not a law of nature.
Why Is Your Open Rate Probably Lying to You?
On June 7, 2021, Apple announced Mail Privacy Protection at WWDC, and the feature went live with iOS 15 on September 20, 2021. According to Litmus, Apple Mail downloads each message and caches its images, tracking pixel included, through a proxy server. Your email platform sees the pixel load and logs an open. When the reporter later reads the message, or never does, the images come from Apple’s cache and you see nothing new.
The feature is opt-in, but adoption was fast. MediaPost reported in January 2022, citing Litmus Email Analytics data drawn from over a billion opens, that MPP accounted for 40.6% of all email opens by December 2021, up from 25.9% in November. Before the launch, Litmus had measured Apple’s Mail apps at 49.8% of opens by the end of August 2021.
For a PR team, the practical effect is simple. A journalist with MPP enabled on an iPhone generates an open on every pitch you send, including the ones that go straight to trash. BuzzStream made the same point in its 2025 subject line study, noting that Apple MPP and Gmail image caching may inflate open rates. That is why the jump from 39% to 48% between Q4 2022 and Q1 2023 in Propel’s data should not be read as reporters caring more overnight. Some of it may be real. Some of it is machines.
There is a second distortion worth naming. Many newsrooms run security scanners that fetch links and images before a message lands, which can also trigger tracking. None of the public benchmarks separate those events, so no one can tell you what share of a 46% open rate came from people.
What Journalists Say Happens After the Open
Survey data fills in the part pixels cannot. Muck Rack’s State of Journalism 2026 surveyed 1,044 journalists between January 30 and March 2, 2026, keeping 897 responses after quality checks. Asked how often they respond to PR pitches, 47% said seldom, 7% said never, 25% said about half the time, 15% usually and 6% always. Asked how many pitches arrived on their last normal workday, 43% said one to five and 14% said more than 20.
The same report found that 86% of journalists say at least some of their stories start with a pitch, which is the good news. The bad news is that 88% immediately disregard pitches that miss their beat, 71% delete pitches that read as overly promotional and 50% delete anything that looks like a mass email. Sixty-two percent prefer a one-to-one email, 69% want pitches under 200 words and 78% want them before noon.
Cision’s 2025 State of the Media report, released May 7, 2025 and based on more than 3,000 journalists across 19 markets, lands in the same place. Cision found that 86% of journalists would instantly decline a pitch that does not match their beat or audience, and 72% called press releases the most useful resource PR teams provide.
Put those surveys next to the Propel numbers and a picture forms. Reporters open a lot, often on a phone, often within minutes. They decide in seconds. The open tells you the subject line cleared the inbox. Whether the pitch fit the beat shows up only in the reply.
Opens and Replies Measure Different Things
The cleanest proof sits in Propel’s own topic data. In Q1 2023, pitches announcing AI products drew a 91% open rate and a 1.95% response rate, according to MediaPost’s coverage of that Barometer. Solar energy pitches opened at 94.74% and drew replies 0.87% of the time. Both topics were hot, so curiosity drove opens. Neither topic produced replies above the 2.91% average for all pitches that quarter.
By Q3 2023, Propel reported that the “AI” keyword had fallen to a 57.66% open rate with a 1.72% response rate, while pitches mentioning “generative AI” opened at 61.62% and drew a 9.4% response rate. A narrower, more specific angle beat the buzzword on the metric that matters.
This is the core lesson for anyone reporting on press release open rates to a boss or client. Opens reward topical heat and clever subject lines. Replies reward relevance and a real story. If your dashboard celebrates a 70% open rate on a list that produced zero replies, the dashboard is measuring how interesting your subject line sounded, and that is all.
At Instant Press we treat opens as a smoke alarm. If they fall off a cliff, something is wrong with the list, the sending domain or the spam filters. If they stay high, that tells us almost nothing about whether the story will run.
Score Every Pitch With the PROOF Ladder
Because opens are noisy, you need a scoring model that ranks signals by how much human intent they require. We call it the PROOF Ladder, five rungs that run from weakest evidence to strongest.
P is for Pixel open. It is the bottom rung, worth one point, and it proves only that the email arrived and something loaded its images. R is for Reply. Any human response, even a polite pass, earns three points, because a person read the pitch and chose to answer. O is for Opportunity, worth five points, when the reporter asks for something: data, a source, an image, a call. The second O is for On-record interview, worth eight points, because a journalist spent their time on your spokesperson. F is for Filed story, worth twelve points, when coverage publishes.
Score each campaign by summing the points and dividing by pitches sent. A campaign of 100 pitches with 46 opens and nothing else scores 0.46. A campaign with 30 opens, four replies, two asset requests and one filed story scores (30 + 12 + 10 + 12) divided by 100, or 0.64, even though its open rate looks worse. The weights are a judgment call, not a statistical finding, so adjust them to your goals. What matters is that pixels can never outvote people.
The ladder also tells you where a campaign broke. High P with no R points to a relevance problem, which Muck Rack’s 88% figure says is the most common reason pitches die. R without O usually means the story was fine but you offered nothing a reporter could use, which is why Muck Rack respondents rank access to sources and original data near the top of what they value. O without F often means timing or editorial priorities, which you cannot always control.
Which Subject Line and Length Choices Move the Numbers?
The largest public subject line study for PR comes from BuzzStream, which analyzed more than 6 million digital PR subject lines in a study published June 25, 2025. BuzzStream used opens as its main metric and flagged the MPP caveat up front, so read the gaps between segments rather than the absolute numbers.
Exclusives opened at 54.16% versus 37.16% for pitches that did not offer one, yet only 13.38% of pitches mentioned an exclusive. Subject lines that named the target publication opened at 50.81% versus 39.25%, though only about 1% of senders did it. Title case beat sentence case, 40.75% to 35.03% on opens and 1.46% to 1.01% on replies. Including the recipient’s name did not help, coming in at 38.45%.
Propel’s data adds the length picture. Subject lines of one to five words earned a 4.73% response rate in Q1 2024, and pitch bodies of 51 to 150 words earned 7.51%, more than double the 3.43% average. In Q3 2023, bodies between 501 and 1,000 words drew 1.65% while making up 34.67% of all pitches sent. A third of the industry was writing pitches about seven times longer than the length that performs best.
Those findings point in a consistent direction. Short bodies, specific angles and genuine exclusivity move replies. Pasting the full press release into the email body does not.
One caution on reading these studies side by side. BuzzStream’s sample is digital PR outreach, much of it data-led content pitched to a wide set of writers, while Propel’s sample is weighted toward corporate and agency media relations. The two datasets use different platforms, different senders and different definitions of a reply. Neither publishes confidence intervals, and neither controls for pitch quality, so a longer pitch may perform worse partly because longer pitches tend to come from teams sending broader, less targeted lists. Use the direction of each finding, test it on your own lists and keep the version that wins for you.
Set Honest Targets for Your 2026 Campaigns
Begin with the baseline the data supports. For a targeted list, an open rate between 40% and 50% is normal for the Propel population, and a reply rate near 3% is average. A campaign that clears 5% on replies is doing well by any public measure. If you send 200 pitches and get six replies, you are at the industry average, not failing.
Be careful with small numbers. A 20-pitch exclusive offer that earns two replies posts a 10% reply rate, but one extra reply or one fewer swings it by five points. Judge campaigns of fewer than 100 pitches on the quality of what came back, and compare quarters only when list size and beat mix stay similar. The same rule applies to opens. A tech beat list heavy with iPhone users will show higher pixel opens than a trade press list that reads on desktop Outlook, and neither difference says anything about interest.
Next, separate your reporting. Put opens on a deliverability line and judge them only against your own history. Put replies, opportunities and filed stories on the performance line, and score them with the PROOF Ladder or something like it. When a client or executive asks about press release open rates, show both lines and explain the MPP asterisk once, in plain words.
Then fix the inputs the surveys keep naming. Build lists by beat, not by outlet. Keep pitches under 200 words, as 69% of Muck Rack respondents prefer, and send them before noon. Offer sources and data in the first paragraph. Send one follow-up, since 50% of Muck Rack respondents call one follow-up ideal and 51% want it within three to seven days. If you want more context on how coverage turns into AI visibility, the Instant Press blog covers that side of the work.
Last, watch the methodology on any new benchmark you see. Ask who sent the pitches, how opens were counted, whether Apple’s proxy opens were filtered and how many messages were in the sample. A number without those answers is a guess dressed as a statistic.
As privacy features spread across more inboxes, the teams that win the next few years will be the ones who stopped counting pixels and started counting conversations.