Second date tips: should you go if there was no spark?
Is one flat first date a verdict? Speed-dating research suggests chemistry is hard to predict before you meet. Three rules for choosing a second date.
Go to the second date if the first one was pleasant and nothing about it felt wrong, even if you felt no spark. Skip it if you felt dismissed, pressured or unsafe, or if something came up that you could not live with. The evidence for that rule is a result that sounds like a failure: researchers using machine learning on speed-dating data could not predict who would want whom before the dates, which makes a single flat evening a weaker verdict than it feels.
What the failed prediction says
Samantha Joel, Paul Eastwick and Eli Finkel ran two speed-dating studies with romantically unattached participants. Beforehand, each person completed more than 100 self-report measures of traits and preferences that earlier research had linked to mate selection. At the event, each person met every participant of the opposite sex for a four-minute speed date. The researchers then tried to predict who wanted whom with random forest models, a machine learning method.
The models could predict something about each person's general tendencies. They explained 4 to 18 percent of how much a person desired others overall, and 7 to 27 percent of how much a person was desired. What they could not do was predict desire for a specific partner, over and above those general tendencies, from anything reported before the dates. The authors conclude that the compatibility elements of human mating are challenging to predict before two people meet.
The limits are real. The dates lasted four minutes, the participants were people at a speed-dating event, and the models used what people reported about themselves beforehand, not photos, chats or anything that happens in a longer relationship. The study does not say attraction is random or that nothing predicts a good match after meeting. It says that a long checklist of who someone says they are and what they say they want does not reveal who they will click with. That fits a second finding by Eastwick and Finkel, also from speed dating: the preferences people stated before an event failed to predict what inspired their actual desire there, and the authors suggest people may lack introspective awareness of what drives their romantic judgments.
For a second date, the point is narrow. If you ruled out a person on paper, or ruled them in, the first meeting is the first real data. It is also only one data point.
Why a flat first date misleads
Two other findings pull in the same direction.
Erica Boothby and colleagues found that after conversations, people systematically underestimate how much their partner liked them, a gap that lasted for months among college dorm mates. If you left a date thinking it went badly, your own replay is the least reliable witness, and our page on first date nerves explains why.
Harry Reis and colleagues ran two experiments in which pairs who had never met interacted for different lengths of time. The more they interacted, the more attracted they were to each other, and the researchers traced the effect to three things: feeling that the other person was responsive, feeling more comfortable and satisfied during the interaction, and feeling that they knew the other person. Those pairs were same-sex and not on a date, so this is evidence about liking, not about romance. It does mean a second meeting can feel different from the first for ordinary reasons: the nerves have gone and there is more to talk about.
There is one counterweight, in a study of online daters. Liesel Sharabi surveyed 186 people before a first date with someone from a dating site or app, and 94 of them again after it. On average, social attraction, how much people liked the other person, fell after the first face-to-face meeting. If you built up an impression in the chat, a first date that is a little flatter than the chat is the usual experience in that sample, not a sign the match was wrong.
Three rules for the decision
These are our judgment, built on the findings above. No study tests them.
1. Ask different questions of the evening. "Was there a spark?" is the question the prediction research says you cannot answer well from one meeting. Ask instead: did I feel respected, did I get curious about this person at any point, and did we laugh at least once? A yes to the first and one of the others is enough for one more date.
2. Spend one more date, not three. If the answer was no spark but no discomfort, say yes once, in a different setting (ideas below), and make the call after that one. Giving it a "few more tries" is how a polite no turns into months.
3. Treat deal breakers as deal breakers. The rule of giving it time is for chemistry, not for values. Bruch and colleagues found that people apply sharp cutoffs on things like age, smoking and children. If you learned on the first date that you disagree about children, or you felt dismissed or pressured, more time is not the fix. Stay with the first impression and say no kindly. Our page on saying no without ghosting has wording.
Second date ideas
If the first date was a drink across a table, make the second one side by side. A new setting gives you new things to react to, and you no longer have to carry the whole evening on questions. Second date ideas of ours that do that and still leave room to talk:
- A Saturday morning flea market or farmers' market, with coffee after.
- A walk with a destination: a lookout, a bakery across town, a bridge you have never crossed on foot.
- Pool, darts or bowling, where being bad at it is half the fun.
- A free museum where each of you picks one room to show the other.
- A food hall where you order for each other.
Our date ideas page has fifty more, sorted by stage.
Questions to ask on a second date
You now have material, which the first date lacked. Before you go, pick one thing they told you and ask about it early. Karen Huang and colleagues found that people who ask more follow-up questions are better liked, and in their speed-dating data those people got more yeses to a second date.
Then go one step deeper than last time. Michael Kardas, Amit Kumar and Nicholas Epley found that people overestimate how awkward a deep conversation with someone new will be and expect the other person to care less about what they share than that person really does; their participants felt more connected after deep conversations than after shallow ones. They studied people meeting for the first time, not daters, but the advice carries. Some second date questions of ours:
- "Last time you said your sister was moving to Denver. Did she go?"
- "What's something you've changed your mind about in the last few years?"
- "What does a really good Sunday look like for you?"
- "What were you like at sixteen?"
- "What's something you keep meaning to learn?"
Answer your own questions too. More are in first date questions.
Second date tips for the day itself
Tell them in the message that you want to see them again and name the plan. "I had a good time. Want to check out the flea market on Saturday morning?" is easier to answer than "we should do this again." How long to leave between a first and second date has no research answer; our rule is to suggest a plan within a few days and meet within a week or two, while the first conversation is still fresh. Then two suggestions of ours:
- Give it an end point. Plan something with a natural finish, such as a movie or a market that closes at two. If it is going well you can extend it; if not, nobody has to invent an exit.
- Judge it the next day. The liking gap suggests that on the way home you will underrate how much they liked you.
The first-date basics, from how long to stay to who pays, are in first date tips.
Third date ideas
By a third date you can plan for longer and let the date show more of each other's lives. Third date ideas of ours:
- Swap planners. Whoever did not choose the second date chooses the third.
- A tour of one person's neighborhood, ending at their regular coffee place.
- An afternoon in two halves, such as a matinee and then an early dinner, with a natural point to check in.
- A picnic where each of you brings one dish from your family or hometown, which starts a conversation on its own.
For what comes after, first date ideas has a fourth and fifth date table.
Frequently asked questions
Should you go on a second date if there was no spark?
Can attraction grow after a first meeting?
What is worth deciding on after one date?
How do you say no to a second date?
How can you tell if a first date went well?
Sources
- Joel, Eastwick and Finkel (2017), Is romantic desire predictable? Machine learning applied to initial romantic attraction, Psychological Science Abstract read on October 11, 2026 (the full paper was not read). Supports two speed-dating studies with romantically unattached participants who completed more than 100 self-report measures, 4-minute speed dates with every opposite-sex participant, models that predicted 4 to 18 percent of actor variance and 7 to 27 percent of partner variance, and no ability to predict relationship variance, the desire for specific partners, from anything reported before the dates.
- Eastwick and Finkel (2008), Sex differences in mate preferences revisited: do people know what they initially desire in a romantic partner? Journal of Personality and Social Psychology Abstract read on October 11, 2026. Supports the finding, from speed dating with longitudinal follow-up, that ideal preferences stated before the event failed to predict what inspired actual desire at the event, and the authors' suggestion that people may lack introspective awareness of what influences their romantic judgments.
- Reis, Maniaci, Caprariello, Eastwick and Finkel (2011), Familiarity does indeed promote attraction in live interaction, Journal of Personality and Social Psychology Abstract read on October 11, 2026. Supports two experiments in which previously unacquainted same-sex pairs interacted for varying amounts of time, the finding that the more they interacted the more attracted they were to each other, and the three mediating processes: perceived responsiveness, comfort and satisfaction during the interaction, and perceived knowledge. These were same-sex pairs, so the study is evidence about liking and attraction in a general sense, not about dating.
- Boothby, Cooney, Sandstrom and Clark (2018), The liking gap in conversations, Psychological Science (University of Essex repository copy) Abstract read on the repository page October 11, 2026. Supports the finding that after conversations people systematically underestimate how much their partner liked them and enjoyed their company, in the laboratory, among dorm mates and among workshop participants, in conversations of varying length, and for several months among dorm mates. Rechecked October 11, 2026.
- Sharabi (2015), Modality switching in online dating: identifying the communicative factors that make the transition from an online to an offline relationship more or less successful, doctoral dissertation, University of Illinois at Urbana-Champaign Full text read October 11, 2026 (the dissertation behind Sharabi and Caughlin 2017, Personal Relationships). Supports the sample of 186 people surveyed before a first date with someone from a dating site or app, with 94 followed up after it, and the report that, on average, the sample declined in social attraction after the first face-to-face meeting.
- Huang, Yeomans, Brooks, Minson and Gino (2017), It doesn't hurt to ask: question-asking increases liking, Journal of Personality and Social Psychology Full text read October 11, 2026. Supports the finding that people who ask more follow-up questions are better liked, and the speed-dating study of 110 people on 4-minute dates in which a higher follow-up question rate went with more agreements to a second date. Rechecked October 11, 2026.
- Bruch, Feinberg and Lee (2016), Extracting multistage screening rules from online dating activity data, PNAS Full text read October 11, 2026. Supports the finding that people apply sharp cutoffs on attributes such as age, smoking and children when deciding whom to look at and write to, which is why those items are treated as deal breakers on this page.
- Kardas, Kumar and Epley (2022), Overly shallow?: Miscalibrated expectations create a barrier to deeper conversation, Journal of Personality and Social Psychology (Europe PMC record) Abstract read on October 11, 2026; the full text was not read. Supports the findings that conversations between people who had not met before felt less awkward and created more connectedness than participants expected, that participants especially overestimated how awkward deep conversations would be compared with shallow ones, that they felt more connected to deep conversation partners after having both kinds, and that they expected others to care less about their disclosures than others did.