“What do you do?” is a common opening question because it is easy to answer and often gives a conversation somewhere to go. It can also feel reductive when work is precarious, unwanted, difficult to explain, or simply not how someone wants to introduce themselves. Context decides a great deal.
Research does not show that people who are good at small talk avoid asking about work. No study cited in the original version of this article compared that question with other opening lines. What the evidence does suggest is that curiosity, follow-up questions, and room for reciprocal disclosure can help a conversation feel more attentive.
The job question has not been put to this test
A professional reception in Copenhagen is different from a neighbour’s dinner, a parent group, or a conversation with someone who has recently retired. At the first event, asking about work may be expected and useful. In the other settings, it may be one option among many.
The original headline turned a contextual judgement into a rule: if you are good at small talk, do not ask what someone does for a living. That conclusion sounds research-backed, but neither of the article’s sources studied the question itself, assessed conversational skill, or identified which topics skilled speakers avoid.
There are good social reasons to keep the question available. Work can be meaningful, surprising, or an easy point of connection. The problem is not the subject by itself. It is the assumption that a job title supplies a complete picture of the person answering.
Follow-up questions have stronger evidence
A 2017 paper in the Journal of Personality and Social Psychology offers more direct evidence about conversational quality. Across three studies of live conversations, Karen Huang and colleagues found how question-asking affected liking. People who asked more questions, particularly follow-up questions, were generally better liked by their partners.
The researchers linked that result to perceived responsiveness: asking about something the other person has just said can signal listening, understanding, and care. The studies included online get-to-know-you conversations and an analysis of face-to-face speed dates. They did not establish a script for all encounters, and they did not test whether a work question helps or harms.
The distinction between an opening question and a follow-up matters. “What do you do?” can become a useful conversation if the next question responds to the answer rather than placing the speaker into a category and moving on. Conversely, an apparently imaginative opener can still fall flat if it is delivered as a trick and the answer is not heard.
Self-disclosure can be rewarding, within limits
The original article also cited work by Diana Tamir and Jason Mitchell. Their 2012 Proceedings of the National Academy of Sciences paper found that people valued opportunities to disclose information about themselves. In several experiments, participants sometimes gave up small amounts of money to answer questions about their own opinions rather than answer factual questions or judge somebody else.
That is evidence about the reward value of self-disclosure in controlled tasks. It does not show that strangers want to discuss their hopes, childhoods, or inner lives on demand. It also does not show that a question about employment blocks meaningful disclosure. Personal questions can invite connection, but they can also feel intrusive when trust and context are missing.
A considerate conversational partner therefore watches the response. A brief answer may signal that the topic has run its course. An animated answer offers material for a follow-up. This is ordinary interpretation rather than a laboratory finding, and it remains sensitive to culture, language, personality, and setting.
Awkwardness is often judged too harshly afterwards
Another study used in the original article concerned the “liking gap”. Erica Boothby and colleagues found across several settings that people tended to underestimate how much a new conversation partner liked them and enjoyed their company. The work covered laboratory conversations, university flatmates, and people meeting at a workshop.
This does not mean that every conversational risk is imaginary or that every partner is secretly delighted. It is an average difference between perceptions and reported liking. Still, it offers a useful correction to the tendency to treat a pause, a clumsy transition, or an imperfect question as proof that the exchange failed.
Small talk does not need to become immediate intimacy to be worthwhile. A modest exchange can be enough.
A better principle than banning one question
There is no scientifically approved list of openers. Questions about the place, the occasion, food, travel, hobbies, or something already mentioned may work because they give the other person choices about how much to reveal. Questions about work can do the same when the setting makes them natural.
The best-supported lesson from the cited research is not to avoid occupations. It is to ask with genuine interest, listen to the answer, and follow where the other person gives you something to follow. That is a smaller conclusion than the original headline promised, but it is the one the evidence can carry.