Do Professors Actually Reply to Cold Emails? (2026 Data)

Before writing a single line to a professor, almost everyone asks some version of the same question: does this even work, or am I about to spend an evening on an email that lands in a void? Most advice on the topic answers with a vibe, some version of "it depends, just try it." There is actually a real number behind this, from a real study, and it is a lot higher than most applicants assume.

The real study, and the real number

Researchers Katherine Milkman, Modupe Akinola, and Dolly Chugh ran a field experiment sending fictional meeting-request emails from prospective doctoral students to roughly 6,500 faculty members across 258 U.S. universities and 89 academic disciplines. Each email asked for ten minutes of the professor's time to discuss a PhD, phrased to sound like a genuine, individually written request rather than a mass mailing.

The headline number: 67% of those emails got a response. Two out of three professors, cold-emailed by a stranger with no prior relationship, replied. That is the base rate for this specific type of email, a short, sincere request for a meeting from a prospective doctoral student, not for cold email in general.

Why that number is genuinely surprising

Context matters here, because "cold email" usually carries a much worse reputation than 67% would suggest. Large-scale studies of cold email in sales and business outreach, drawing on millions of emails sent through outreach platforms, consistently report overall reply rates in the low single digits, with even strong, well-personalized campaigns typically landing somewhere in the high single digits to high teens at best. A generic, unpersonalized cold email in that world routinely gets under 1% back.

Set the academic number next to that and the gap is stark. Professors are not behaving like overworked sales prospects deleting anything that smells like a pitch. Most of them are, in some real sense, doing their job when they read and answer a genuine question about their research and their lab. That is a structurally different situation from a stranger trying to sell software to a VP of Marketing, and it is worth remembering on the days this process feels hopeless.

The number is not evenly distributed

The 67% is not the whole story, and treating it as one uniform number would be misleading. The same study varied the applicant's name to signal gender and race or ethnicity, and found real, measurable gaps in who got a response.

• The widest gap showed up in business, where names signaling a white male applicant received a response 87% of the time, versus 62% for names signaling a woman or a minority applicant, a 25-point spread within a single discipline.
• The humanities showed the smallest gap, small enough that it did not reach statistical significance on its own, but women and minority applicants were still collectively ignored at roughly 1.4 times the rate of white male applicants even there.

This is a documented, real finding, not a hypothetical, and it deserves to be stated plainly rather than smoothed over: some of the gap in who gets a reply has nothing to do with the quality of the email and everything to do with bias in who reads it. No template, subject line, or amount of personalization closes that gap on its own. What it does mean practically is covered below.

What the remaining third of silence usually is

A third of emails in the study went unanswered. Some of that is the bias described above. A lot of it, based on how academic inboxes actually work, is simpler: a professor genuinely not taking students that cycle, a message buried under a hundred others during a grant deadline, or an email that arrived at a moment when a reply would have required more time than the professor had. None of that is something you can diagnose from outside the inbox, which is exactly why treating silence as a considered no, rather than an unresolved maybe, tends to waste real opportunities. A single, well-timed follow-up after 7-10 days converts a meaningful share of that silence into an actual answer, because the most common reason for no reply is a buried inbox, not a decision.

What actually moves a specific email from the 33% into the 67%

The study measured a generic, sincere meeting request. It was not testing what happens when you stack every controllable advantage on top of that baseline. Based on what separates a forgettable email from one that gets read, three things are worth taking seriously:

1. Specificity beats general interest, every time. The gap between a 1% cold email and one that lands closer to double digits, in the general cold-email data above, is almost entirely explained by personalization. In academic outreach the equivalent move is naming a specific, recent paper and one real reason it connects to your interests, not a sentence that could be copy-pasted into fifty other inboxes. See the exact structure that gets replies for what that looks like in practice.
2. Reading the paper first, not skimming the abstract. The line that actually proves you read something is usually buried past the abstract, in a specific method, dataset, or result. A 10-minute targeted read is enough to find it without turning every email into an afternoon project.
3. Timing, at the margin. Late summer and early fall, when professors are actively planning next year's intake, tend to produce more attention than December, when departments are underwater in applications. Timing will not turn a generic email into a good one, but it can turn a good email into one that gets read the same week instead of two months later.

What 67% actually looks like across a real list

It helps to see the number applied to an actual-sized search rather than treated as an abstraction. If the 67% baseline held evenly across a list of 30 well-targeted professors, roughly 20 would respond in some form. That is not 20 offers, most replies are a polite "not taking students this cycle" or a request for more information, but it is a very different picture from the fear most applicants start with, which is closer to "almost nobody will answer at all." The real bottleneck in a funded outreach search is rarely getting a reply. It is turning a reasonable number of replies into an actual funded yes, which depends on fit, timing, and budget far more than on how the email was written.

If your own response rate looks nothing like this

If you have sent a real number of emails, 15 or more, genuinely personalized, and gotten close to zero back, that gap between your result and the study's baseline is worth investigating rather than absorbing as bad luck. A few things are worth checking honestly before assuming the whole approach is broken:

• Is the "specific" detail actually specific, or does it just sound specific? "Your work on machine learning" is not specific even if it names a real field. A professor can tell the difference between a genuine reference to their paper and a sentence that was reworded by a template to look like one.
• Are you emailing professors who are still active? A lab page that has not been updated in years, or a professor with no recent publications, is a much weaker target regardless of how good the email is, since there may be no lab left to reply from.
• Is the timing working against you? A batch sent in mid-December, when departments are buried in admissions files, will underperform the same emails sent in September, independent of email quality.
• Is it actually reaching the inbox? A CV attachment from an unfamiliar file host, or an email sent from an address that looks automated, can get filtered before a professor ever sees it land.

What volume has to do with any of this

Even at a genuinely favorable 67% baseline, a response is not the same as a yes to supervising you with funding, and most individual professors, regardless of how warmly they reply, will not have space or budget that specific cycle. That is precisely why the applicants who end up funded are consistently the ones who reached a real number of well-matched professors rather than the ones who wrote three perfect emails and stopped. A high base response rate makes volume more worthwhile, not less necessary.

What this means if you are worried the bias applies to you

If your name, background, or where you are writing from puts you in a group the data above shows gets replied to less often, that is a real, structural disadvantage in the process, and it is not something a better email fully offsets. What it does argue for is not writing fewer emails to compensate for lower odds per email, but more: reaching a wider, well-targeted list so that the professors who do reply, and plenty will, are not left to chance on a handful of names. It also argues for not reading a string of silence as a personal verdict on you specifically, since the study shows plenty of that silence is structural, not individual.

Where a tool like this fits, and where it does not

GradScoutFunding finds professors filtered by field and country, confirms they are actively publishing (the clearest available signal a lab is still running and likely still taking students, though no tool, including this one, can verify in advance that funding is actually available), and drafts a personalized email grounded in a real recent paper for you to review and send yourself. It does not send anything automatically, and it does not, and cannot, promise a reply. What it removes is the hours of manual searching and paper-reading that specificity requires, so reaching a real number of well-researched professors is a realistic week, not a burned-out month. Your first 2 credits are free, no card required.

Common questions