How to Build Chemistry with Someone New

Chemistry feels like something that either happens or doesn't. Research on conversations between strangers suggests you have more influence over it than that. Three bodies of work point to two things you can do, and one mistake you can stop making.

Go one layer deeper than feels safe

Kardas, Kumar, and Epley (2022) ran 12 experiments with more than 1,800 people. Participants first predicted how a conversation with a stranger would go, then had the conversation. The real conversations beat the predictions. They were more pleasant, more connecting, and less uncomfortable than expected. The gap was widest for personal topics, which people assumed would be the most awkward. Participants who had both a light conversation and a meaningful one came away feeling closer to the partner from the meaningful one.

The researchers traced the hesitation to a single error. People assume a stranger won't care much about what they share, and most strangers care more than we expect.

To use this, move one step past the safe question. Instead of asking what someone does for work, ask what they wanted to be before that, or what they would change about it. Then offer something real about yourself in return. The evidence supports going a layer deeper than small talk, which is a long way from confessing everything on a first date.

Follow up on what they just said

Huang and colleagues (2017) ran three studies of live conversations, one of them using speed-dating data. People who asked more questions came across as more likable, and follow-up questions did most of the work. A follow-up shows you took in the last answer, and the other person experiences that as being heard and cared about. In the speed-dating data, people who asked more questions were more likely to be chosen for a second date. Participants didn't predict any of this. Most people underrate how much a good question helps.

To use this, listen for the detail in someone's answer and ask about that. If they say they grew up in Ohio, "What was it like leaving?" will take you further than "What do you do now?" Hold your own story for a minute. You'll get your turn, and you'll have earned more interest by then.

Don't trust your own review of the date

Boothby, Cooney, Sandstrom, and Clark (2018) ran five studies on how people judge a conversation after it ends. Again and again, people guessed that their partner liked them less than the partner actually did. The researchers named this the liking gap. It showed up among strangers in a lab, among college roommates getting to know each other over a school year, and among adults meeting at a workshop. Observers watching recordings could tell how much the two people liked each other, while the people in the conversation could not. The gap was larger for shy people and for anyone with a running stream of negative thoughts during the talk. Among the roommates, it lasted for months.

To use this, treat your post-date verdict as biased against you. If you want to see the person again, ask. They probably rated the evening higher than you did.

Where chemistry comes from

From the outside, chemistry looks spontaneous. In these studies it grew out of small choices: one more personal question, one well-placed follow-up, and one invitation sent despite doubt. That leaves a question worth considering: how many promising connections have ended because both people assumed the other wasn't that interested?

References

Boothby, E. J., Cooney, G., Sandstrom, G. M., & Clark, M. S. (2018). The liking gap in conversations: Do people like us more than we think? Psychological Science, 29(11), 1742–1756. https://doi.org/10.1177/0956797618783714

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), 430–452. https://doi.org/10.1037/pspi0000097

Kardas, M., Kumar, A., & Epley, N. (2022). Overly shallow?: Miscalibrated expectations create a barrier to deeper conversation. Journal of Personality and Social Psychology, 122(3), 367–398. https://doi.org/10.1037/pspa0000281