Conversations Don't End When We Want Them To - Ideas for Leaders
Idea #805

Conversations Don’t End When We Want Them To

This is one of our free-to-access content pieces. To gain access to all Ideas for Leaders content please Log In Here or if you are not already a Subscriber then Subscribe Here.
Main Image
Main Image


Successful and productive social interactions, in the form of our conversations with others, are at the heart of our lives and our work. A new study reveals, however, that most conversations don’t end when people want them to end. Conversations, it seems, are much harder to manage than we realize.


As human beings, we are spectacular at pulling off all the complicated tasks it takes to have a conversation. The one conversation challenge we have not mastered is coordinating the end of the conversation. In most cases, according to recent research, a conversation ends either later or earlier than either partner wants it to end, and for two good reasons. The first is that partners in a conversation do not want to leave the conversation at the same time. As a result, one partner may cut the conversation short or keep the conversation going longer than the other partner would like. The second reason is that we hide from the other person when we would like the conversation to end for fear of hurting feelings or appearing rude. When we are ready for a conversation to end, and the other person is still speaking, rather than declaring, “I’d like this conversation to end now,” we say nothing… and the conversation continues until, perhaps, we run out of patience and use an escape route, such as, “I’d love to continue, but I have an appointment.” By that time the conversation has lasted longer, sometimes much longer, than we wanted. Or, perhaps, it is the reverse: the other person ends the conversation before we are satisfied with its outcome. In sum, it is difficult for partners in a conversation to get what they want, when they both want different things and don’t know what the other partner wants. 

A new study reveals just how badly the duration of most conversations fails to satisfy either partner. The first part of the study involved one-on-one conversations among people who knew each other well (romantic partners, friends or family members). In an online survey, 806 participants were asked a number of questions about their most recent conversation, beginning with how long the conversation lasted whether at one point they wanted the conversation to end, and if so, how much longer the conversation continued after they had wanted it to end; or if they believed the conversation ended too soon, how much longer they had wished the conversation had continued.

The second part of the study involved one-on-one conversations among strangers who agreed to participate in a laboratory study. The participants were asked to converse with a single partner for at least one minute and no more than 45 minutes (to reduce the temptation to finish this conversation with a stranger very quickly, participants were told that whenever the conversation finished, they would have tasks to complete for the remainder of the hour). As with the first study, participants indicated whether they believed the conversation went on for too long or ended too soon, and by how much. Participants also guessed how their partners in the conversation would answer the same questions.

The results of the study showed that 66% of the people talking to intimates and 68% of the people talking to strangers had been ready to end the conversation before it finished. In the first part of the study, the participants wished on average that the duration of the conversation with their intimates had been different by a remarkable 56% of the actual duration—in other words, the actual duration of the conversations were no where near the desired duration. The laboratory experiment involving strangers yielded similar results: the participants wished on average that the duration of the conversation had been different by 46% of the actual duration. 

Unlike the online survey responses, the laboratory part of the study yielded data from both partners in the conversation—data that confirmed that both parties were unsatisfied. Specifically, less than 2% of conversations had ended when both partners wanted them to end, and less than 30% of conversations had ended when at least one partner wanted them to end. 


Conversations are the engine of life… and business. Whether you are a leader, manager or customer-facing employee, a salesperson, human resources manager, or an independent consultant, your ability to manage conversations in a way that leaves both sides equally satisfied is key to your success. Standard attributes such as people skills, communication skills and emotional intelligence rest on successful conversations. 

As indicated earlier, conversations don’t end when people want them to end because, first, conversation partners want the conversation to end at different times, and second, not wanting to be rude or hurt the feelings of others, we don’t let the other person know of our desires. We also try to guess what the other person is thinking without asking them, and, this study also shows, we often guess wrong. 

In sum, the problem with conversations, as revealed by this study, is a coordination problem: we don’t collaborate with our conversation partners to coordinate the ideal duration for all parties. A potential solution, therefore, is intentional coordination. For example, before the conversation begins, determine together how long a conversation on a specific subject should last. Determine together, also, the outcome of the conversation: what would make both partners happy when the conversation ends. Finally, ask the final question: This is when our conversation is supposed to end; are you happy with the result?

This prescription is more appropriate for a business setting. When speaking with romantic partners, friends and families, we will probably continue the guessing game—in many cases, a wise choice.



  Adam M. Mastroianni’s profile at Harvard University
  Daniel T. Gilbert’s profile at Harvard University
  Gus Cooney’s profile at The Wharton School of the University of Pennsylvania
  Timothy D. Wilson’s profile at University of Virginia


Do conversations end when people want them to? Adam M. Mastroianni, Daniel T. Gilbert, Gus Cooney & Timothy D. Wilson. Proceedings of the National Academy of Sciences (March 2021). 

Ideas for Leaders is a free-to-access site. If you enjoy our content and find it valuable, please consider subscribing to our Developing Leaders Quarterly publication, this presents academic, business and consultant perspectives on leadership issues in a beautifully produced, small volume delivered to your desk four times a year.


Idea conceived

March 2, 2021

Idea posted

Nov 2021
challenge block
Can't find the Idea you are after?
Then 'Challenge Us' to source it.


For the less than the price of a coffee a week you can read over 650 summaries of research that cost universities over $1 billion to produce.

Use our Ideas to:

  • Catalyse conversations with mentors, mentees, peers and colleagues.
  • Keep program participants engaged with leadership thinking when they return to their workplace.
  • Create a common language amongst your colleagues on leadership and management practice
  • Keep up-to-date with the latest thought-leadership from the world’s leading business schools.
  • Drill-down on the original research or even contact the researchers directly

Speak to us on how else you can leverage this content to benefit your organization.