End every message with a question. The strongest lever against ghosting. (Part 2)
Thirteen points for one line
Messages that end with a question get a reply in 85.1% of cases. Without a question it is 72.2%. That is almost 13 points of difference, from the last line alone.
For first replies, meaning the first message you send in response to an inquiry, it becomes a chasm: 79.2% with a question, 61.8% without. Almost one in five additional replies hangs on that single line.
This is part 2 of the series. You can find the overview of all five patterns in the hub article: Why clients ghost after your reply. The methodology is there too: real booking chats on tatme.com, test accounts removed, a reply counts as a client message within 14 days. And as always: these are correlations, not proven causation. But the effect holds across every cut of the data.
Why a question triggers so much
A question at the end works for two reasons: it makes clear whose turn it is, and it gives the client something concrete to reply to.
Without a question, your message ends as a statement. "The forearm works well for this, I'd do it as fine line." That is a good, competent message. But there is no ball for the client to hit back. They read it, think "cool", and put the phone down. Not out of disinterest. There was simply nothing to do.
The research on this is clear. Huang et al. (2017, Journal of Personality and Social Psychology) showed across multiple studies that people who ask questions are perceived as more responsive, and their counterparts react far more often. The effect was stable enough to measure in speed dating: those who asked more questions got more second dates.
Outside the psychology labs, the same thing shows up. Boomerang analyzed 40 million emails: messages with one to three questions are 50% more likely to get a response than messages with none. An analysis of 770,000 sales emails (Regie.ai) found the same lever: messages that start or end with a question had a multiple of the average reply rate.
People are conditioned to react to questions. An unanswered question feels unfinished. An unanswered statement does not.
The combined effect with speed
Question and speed reinforce each other. First replies with a question, sent within 72 hours, sit at an 86.5% reply rate. First replies without a question, sent after more than a week, at 55.6%.
That is the biggest gap in the entire dataset: 31 points. And both together cost you not a single extra minute per message. Speed has its own part in this series (part 6, the 3-day rule). What matters here: even a fast reply loses a large part of its effect without a question.
Which question fits when
The question has to match where the conversation stands. A good question is concrete, easy to answer and moves you one step forward. Here is what that looks like across a typical inquiry:
| Conversation stage | Weak ending | Ending with a question |
|---|---|---|
| Inquiry comes in | "Thanks for your inquiry, sounds good." | "Where on your body should it go, and roughly how big?" |
| Idea is clear | "I can picture that well." | "I'd do this as fine line. Does that match what you have in mind?" |
| Price is named | "The range would be 300 to 400 euros." | "The range would be 300 to 400 euros. Does that work for you before we look for a date?" |
| Appointment is on the table | "I'd have the 12th or the 19th free." | "I'd have the 12th or the 19th free. Which works better for you?" |
The difference is never the content. It is always just the last line. The weak versions are all friendly and correct. They just end the conversation every time, without meaning to.
Important for the price stage: the question does not replace the information. First the price range, then the question about it. Why the order of idea, price, appointment matters so much is covered in Why clients ghost after the inquiry.
What doesn't work
The question bombardment. Three questions in one message feel like a form to the client. The Boomerang data shows the effect flips past a certain amount, and the reply rate drops again. One question per message, at most two if they belong together (placement and size).
The pseudo question. "Okay?", "Questions?", "Let me know if you need anything?" technically count as questions, but they give the client nothing concrete to do. They are the written equivalent of a nod. A good question can be answered in one sentence, and the answer moves you both forward.
The buried question. A question in the middle, followed by three more paragraphs of info, gets lost. The client reads to the end, and at the end there is no action. The question belongs at the close, as the last line.
The oversized question. "How do you envision the whole project?" is honest, but it demands an essay. The bigger the question, the more likely the answer gets postponed. And postponed, in booking chats, often means: never.
One thing you can do this week
For every reply you send this week, take three extra seconds and check the last line: is there a question the client can answer in one sentence?
If not, add one. You don't have to change anything else in the message. After ten messages you will have a feel for it, after twenty it becomes automatic. It is the cheapest behavior change in this entire series, and the data says: the most effective one.
Sources
- Huang K., Yeomans M., Brooks A.W., Minson J., Gino F. (2017): It Doesn't Hurt to Ask: Question-Asking Increases Liking, Journal of Personality and Social Psychology 113(3): Harvard Business School
- Boomerang (2016): 7 Tips for Getting More Responses to Your Emails (With Data!), analysis of 40 million emails: Boomerang Blog
- Regie.ai: The One Trick That Will Boost Your Reply Rates, analysis of 770,000 outbound emails: Regie.ai
- Internal data basis: 3,539 artist messages and 743 first replies from anonymized booking chats on tatme.com, as of August 2026.