Outreach & relationships
How to write a first message to someone who dislikes pitches

Put the reason, the topic, and the size of the request in the first message, and say plainly what you are not asking for. The reader should be able to say yes, no, or nothing without writing back to find out what you want. Leave out flattery and “quick question” openers, because people who dislike pitches have learned to read those as the start of one.
The line on her website says “no pitches”
Dev is a student who wants to understand how bike repair shops decide which jobs to take and which to turn away. Marta runs a small repair co-op and wrote a long post on the subject. At the bottom of her contact page is a line: “If this is a sales pitch, please skip it. If it isn’t, tell me what it is.”
Dev has no mutual contact, no budget, and nothing to sell. He still has to write a message that looks, from Marta’s side, a lot like the ones she has learned to ignore. The same is true for anyone writing to a researcher, an owner, or a busy specialist for the first time.
Why a cold request is easy to misjudge
People who ask for help tend to expect more refusals than they actually get. Zhao and Epley (2022, Psychological Science) ran scenarios, recalled experiences and live interactions with 2,118 adults in the United States. People who needed help consistently underestimated others’ willingness to help, underestimated how positively helpers would feel, and overestimated how inconvenienced helpers would feel. A Stanford business school summary of Flynn and Lake’s earlier studies reports a similar pattern: askers thought they were more likely to be turned down than they were, in part because people feel social pressure to help. These are findings about requests in general, not about written messages to strangers.
That finding sounds like encouragement, and partly it is. But it has a second edge. In one experiment on email requests, students asked strangers to fill out a short survey. According to the Association for Psychological Science’s summary of the study, email requests were accepted far less often than the requesters had predicted, and in person they were accepted far more often. We leave out the exact percentages because two summaries of the same study report different figures and we could not read the paper itself. The authors suggest requesters focus on their own discomfort and do not account well for what the recipient is looking at: an unfamiliar message from an unknown sender.
So the work of a first message is not to be charming. It is to answer the questions the recipient is silently asking: who is this, what do they want, and how much of my day does it cost?
What the first message has to carry
A short message can hold all of this if each part is a sentence, not a paragraph. The checklist is practical guidance, not a research finding.
- Why you chose her. Name one specific thing she wrote or did, not “your amazing work.”
- What you want to learn, in one line a stranger could repeat to someone else.
- The size of the request: a time, a format, and what you will do with the answers.
- What you are not doing, stated once. If there is nothing to sell, say so.
- A way to decline that costs her nothing, such as “no reply needed.”
A draft that makes her do the work
Here is a first attempt many people would send. It is friendly and honest, and it leaves out almost everything Marta asked for. She has to reply just to find out whether the “chat” is about hiring, buying, or press, and “whenever works” hands her a scheduling job before she has agreed to anything.
It also uses the exact phrase her contact page is written to filter out. “Pick your brain” and “quick chat” are common in real pitches, so in our judgment a reader who has been burned is likely to treat them as a signal, whatever the sender meant.
“Hi Marta, I’m a huge fan of what you’re doing at the co-op! I’d love to pick your brain for a quick chat sometime this week. Let me know what works for you and I’ll make it happen.”
A draft she can answer in ten seconds
The stronger version puts her question first. It cites the post she wrote, says what Dev is trying to learn, gives a time and a format, and states what happens to her answers. The “not selling anything” line appears once, plainly, instead of being repeated nervously.
The offer to answer by email is deliberate. A twenty-minute call is a larger ask than three written questions, and a recipient may prefer either. Giving the smaller option lets her accept without rearranging her week. It also admits something true, which is that a stranger’s request is allowed to be smaller than the stranger hoped.
“Hi Marta, your post on turning away jobs that cost more to fix than the bike is worth is the clearest account I’ve found. I’m a student writing a short report on how repair shops make that call. Could I ask you three questions by email, or take 15 minutes by phone if you prefer? I’m not selling anything, and I’ll send you the report if you want it. If it’s not a good time, no reply is needed. Thanks for reading. Dev.”
If she says no, or says nothing
Silence is the most likely outcome, and it carries almost no information. A busy owner may not have seen the message, may have meant to answer, or may simply not want to. One short follow-up after about a week is reasonable if it adds something, such as making the request smaller. After that, stop.
If she declines, answer in one line and mean it. The aim is to leave her with no reason to dread the next message from a stranger, and possibly no reason to remember yours at all.
If she replies with a question or a condition, answer exactly that question. A reply is a small yes to continuing, not an opening to restate everything.
“Thanks for letting me know, Marta, and no problem at all. If you ever do want the report, my address is below. Best of luck with the co-op.”
What the evidence does not show
The research here is about how requesters misjudge other people, and about how requests in different channels fare. None of it tests this message, and none of it shows that a particular wording raises the chance of a reply. The email experiment used students asking for a short survey, which is a far smaller and more familiar request than an interview.
The phrase “no reply needed” is worth including because it is respectful and true, not because it is a trick. A 2023 pre-registered meta-analysis of “but you are free” wording (Fillon and colleagues) found a medium average effect across 52 experiments, but the seven studies rated low risk of bias showed no clear effect, and the authors rated replicability indicators as very low. Treat it as courtesy, not leverage.
This advice also assumes the person has not told you to stop. If someone has declined, asked you not to contact them, or you are writing in a setting where contact is restricted, do not send another message. The scenario here is fictional; Dev and Marta are invented, and a real person has reasons no script anticipates. Nothing here shows that practicing with Leetsoul causes better outcomes.
The linked exercise lets you write a first message to a different person with a similar stated dislike, and check it against the same elements: reason, topic, size, and an easy way out.
Practice a first message to a busy stranger
Try your response to a fictional conversation, then read the guidance and decide what you would change.
Open the practiceThree starter practices are free. Soulful opens the full library.Sources and limits
Dev, Marta, and the co-op are fictional, and the two messages are original examples written for this article. The sources show how requesters misjudge strangers and how weakly some wording tricks are supported; they do not show that any message in this article will get a reply.
- Zhao and Epley (2022), Surprisingly Happy to Have Helped: Underestimating Prosociality Creates a Misplaced Barrier to Asking for Help (Psychological Science)
Abstract read via OpenAlex (journal page not opened): total N = 2,118 adults in the United States; people needing help underestimated willingness to help and overestimated how inconvenienced helpers would feel. Not a test of written messages.
- You’re Less Persuasive Than You Think Over Email (Association for Psychological Science, summary of Roghanizad and Bohns)
Read as a summary of two experiments on email versus in-person requests to strangers, not the paper (the journal page was blocked). Student samples. The percentages in the summaries conflict, so the article uses none.
- Stanford GSB summary of Flynn and Lake (2008), If you need help, just ask
Read as a summary of the study, not the paper: askers thought they were more likely to be turned down, in part because people feel social pressure to help. Not a test of written messages or introductions.
- Fillon et al. (2023), The effectiveness of the “But-you-are-free” technique: meta-analysis and re-examination (Meta-Psychology)
Pre-registered meta-analysis of 52 experiments (read on the journal page, abstract level). Medium average effect (g = 0.44), but the seven low-risk-of-bias studies showed no clear effect (g = 0.11, interval from -0.18 to 0.40) and replicability indicators were very low.
Published by Leetsoul. Sources checked . Examples are fictional. Read our evidence approach.