Zoom In... or Out? Why Face-to-Face Meetings Matter

Kevin Cool: If we want to generate better ideas, then we need to get people back to the office.

Will Tracy: First of all, it’s a rhythm thing, right?

Kevin Cool: Will Tracy was an executive producer on the HBO series Succession, a cowriter of the film The Menu, and was a staff writer for Last Week Tonight with John Oliver. During his career, he’s been in a lot of writers’ rooms.

Will Tracy: When you’re with a group of writers physically around a table in a room, there’s a sense of ideas building on each other following directly on the heels of somebody’s idea, maybe even in a way that feels not rude but collaborative, almost jumping on top of someone’s sentence. And so, it’s almost like you are crafting that idea together at the same time.

Kevin Cool: When writing a season of a television show, creators can spend up to three or four months in a writers’ room, following ideas, talking about characters in different situations. It is a highly creative and collaborative environment. In his many years as a writer, Will has never tried to do that kind of collaboration virtually.

Will Tracy: With Zoom, there’s always this kind of forced good manners where you have to wait for someone to come to a complete end of their statement. And then there’s a lag because people are also waiting for, did that go through, is there a time lag between anyone here?

So, someone finishes their sentences, and then there’s like a four-second gap, and then someone starts to speak again. And that just doesn’t have the rhythm and dynamism of real conversation. It doesn’t feel like a real conversation; it feels like people are making a prepared statement, then there’s a wait, and then there’s another prepared statement. And I don’t know how it works for other businesses. I think for writing that is not great.

Kevin Cool: Will sees the value of virtual meetings for some gatherings — in fact, this conversation with Will was done on Zoom — but he doesn’t think this technology is well suited for writers in a collaborative setting.

Will Tracy: I also have a feeling just knowing how writers are and how distracted they are and how they like to procrastinate and distract themselves, I always am suspecting that somebody — at least two or three people on the Zoom screen — are looking at their email or looking at another window on their laptop and finding a way to avoid writing and confronting something head on. Because they think, well, these other people are talking and no one knows what I’m really doing right now. And you can’t do that in a writers’ room. You’re right there on stage, basically. Everyone has to be focused on the task. You avoid that problem completely.

You need the grist of everyday life for a writers’ room. You need to talk about what you had for dinner last night. You need to talk about what’s going on in your relationship, what movie you watched. You need that sense of something to draw on, or else you are going to — when you’re writing a story with characters, your mind will resort to the stereotypical unless you can use a bit of experiential something from the room. And it usually comes from the room, it comes from people in the room.

Kevin Cool: We all accept that there are tradeoffs to virtual gatherings, but are they bad for creativity? How does virtual video affect collaboration? How do we decide what we do on Zoom and when it’s better to come to the office? I’m Kevin Cool, senior editor at the Stanford Graduate School of Business. Today we speak with Jonathan Levav, the King Philanthropies Professor of Marketing at Stanford Graduate School of Business. And we’ll talk about a paper Professor Levav cowrote with Melanie Brooks, now a professor at Columbia, asking about the effects of communicating via video on creative idea generation.

Jonathan Levav: What we wanted to contrast is working face-to-face versus working over video. And the reason that came about was in part because of a conversation that I’d had with a friend of mine who’s a noted Silicon Valley CEO who always said to me that in order for product development to happen, your engineers have to be collocated.

Kevin Cool: And I believe one of the exercises you had the study participants do was think of uses for bubble wrap, right? Now that seems like a fun exercise.

Jonathan Levav: Yes.

Kevin Cool: I mean, I would like to be a part of that sort of meeting.

Jonathan Levav: Yeah.

Kevin Cool: So, what did you find?

Jonathan Levav: So, what you’re referring to is what’s called in the creativity literature the Alternative Uses Task. What you do, essentially, is you give people an object and you say, “What are all the different ways in which you could use that object, okay?” In this case it was bubble wrap. And then your dependent variable is the number of uses people come up with, right? So, if I came up with 10 different uses and you came up with 5 different uses and I generated more ideas than you generated.

So, basically what we did in the study is we contrasted the number of ideas and the type of ideas. We can talk about that a little bit later. Both the number and the type of ideas that the pairs of participants generated, whether they were in person or if they worked over Zoom.

Kevin Cool: And what was the result?

Jonathan Levav: Drum roll, can we have it? And so, the result was that pairs that worked face-to-face generated more ideas, about 15 to 20 percent more ideas, than pairs of people that worked on Zoom. And everybody got the same amount of time to work and the same incentive to work, but people that worked face-to-face generated more ideas. But even more interesting than that was that if we analyze — we’re able to analyze the ideas that people generated, that pairs of people generated. And we could analyze those ideas not just for the number of ideas but for the type of ideas and how connected those ideas are to each other.

And so, imagine I tell you, “Kevin, I want you to give me as many alternative uses for a frisbee you can.” So, you can say frisbee, paint palate, plate, sled. Okay, those are very unrelated concepts that all could be alternative uses for a frisbee. Contrast that with frisbee, plate, coaster, tray. Those are all things that are related to serving food, so that’s a much narrower set of ideas.

And so, the interesting thing was that not only did people who worked on Zoom generate fewer ideas, but their ideas tended to be more clustered together than the people that worked face-to-face. So, it’s almost like you get a double penalty. One penalty is in the number of ideas. And what you want in, let’s say, a new product development process is to generate as many possibilities to select from in the idea generation phase of the new product development process. And you want those ideas to be as different from each other as possible because then that also allows you to connect things that are otherwise unconnected and create something that’s more novel. And so, working on Zoom was a double penalty: fewer ideas and a narrower set of ideas.

Kevin Cool: So, those of us who’ve been in creative meetings on Zoom — and that’s a lot of people — this does seem intuitive, maybe, to what our experience is, but why do you think that difference exists? What objectively is different about it?

Jonathan Levav: So, first of all, I don’t know that it’s intuitive. I’ve had that response from people. So, I remember when I presented this in a class of executives and I said, “What do you think would happen if I ran this experiment?” And they said, “Oh, people working face-to-face would generate more ideas.” And I asked them why, and they said, “Well, because you would. You would just generate more because you’re working face-to-face.” So, you can have the explanation be the experimental treatment.

And so, on the face of it, if the information exchange is the same, in theory there shouldn’t be a difference. And after we published the paper and the result went out and was picked up by the press and so on and so forth, we had a lot of people who really fought us tooth and nail to tell us how we were wrong. They didn’t necessarily present any evidence that we were wrong, but they just kept telling us how our experiments were terrible. They didn’t point to a particular factor in the experiment but just that they were terrible. And that’s when you start getting a sense that people here have more religious belief around this question than data.

And so, let’s think about having a conversation. So, right now we’re sitting in this studio. In the studio, there’s all kinds of stuff, right? There’s a poster on the wall, there’s the door, there’s the Stanford Business backdrop, just all kinds of wires, microphones and so on and so forth. If I look a way for a moment at something else, I’m still in this conversation, right? You know that I’m still here because our shared environment is this entire room, okay?

But think about what it means to work over video or on Zoom or whatever other platform you’re using. Now our shared environment is this narrow monitor, right? Typical laptop monitor size is 13 inches. Everything else is outside of that shared environment. So, the way to think about it a little bit is like the iPhone portrait mode, right? So, in the iPhone portrait mode, you focus on one object and everything else gets pixelated. That’s a little bit like what Zoom requires of you.

So, in order to remain in the conversation, I have to remain in the shared environment. In order to remain in the shared environment, I basically have to put blinders on. It’s almost like narrowing your gaze. Now there’s research that shows that people’s physical experience influences their cognitive style. So, if you think about it, I have to be super narrow and super focused. That’s going to translate into a cognitive style that’s also narrow and more focused.

And so, in a creative idea generation task, what you want to be is to think expansively, right? I want to generate ideas, lots of ideas, unrelated to each other. If you will, like a big, gigantic oak tree of ideas. Now I have a physical experience that’s narrowing me down, and instead of making an oak, I make a cypress.

And so, what we found is pretty strong relationship between people’s physical experience — and we measured that by actually filming them and seeing where their eyes went — and their creative output. And so, that’s the thing about Zoom that makes it particularly unique is that it’s not just an experience of talking to each other, but it’s actually where the physical experience of consuming information is different. And for some tasks like creative idea generation, that can potentially have an important influence.

Kevin Cool: So, if I’m a manager in an organization and I’m trying to evaluate when Zoom is okay and when Zoom is not okay or whether in person is much better than Zoom, we know the creativity based on your research. Those should be in person. How do we differentiate other things that we might do on Zoom, and how would that relate to maybe making an overall evaluation on when people should be at work and when they can be remote?

Jonathan Levav: Yeah, so there was another result that we found in our experiments that was weaker than the creativity results, but still there, which is that in addition to generating ideas, the pairs of participants had to select which idea they thought was the best. All of their ideas were then submitted to an external jury that judged it on novelty and usability, right? So, if we think about an innovative product, it’s both novel and usable, right? And so, we could measure how accurate each pair’s decisions were with respect to this external standard.

So, let’s suppose we generated 10 ideas and our best idea was rated a 6. And the idea that we submitted was, in fact, rated a 6, then we made a “perfect decision.” Had we submitted the idea that was rated a 5, then we would’ve scored a -1, right? That’s the size of our error. So, you can have a small error, you can have a big error.

And so, what we found is that actually the groups working on Zoom did a little bit better at the decision-making task than at the creative idea generation task. And if you think about the psychological requirements of the decision-making task, it makes some sense. Because to make a decision, I have to narrow down on something. I have to be analytical; I have to be focused. It’s the exact opposite of the expansive thinking that’s required in creative idea generation. And so, that gave us the hint that the story here isn’t Zoom is bad or Zoom is good, is that different modalities are matched to different task requirements.

And so, that’s a roundabout way to get an answer to your question, which is, look, what matters isn’t are you working at home, are you working hybrid, are you working in the office. It’s what are the psychological requirements of the task. And so, if it relates to expansive thinking, you want to be in an environment that doesn’t force this kind of cognitive narrowness. And if you have to be analytical and focused, then that cognitive narrowness actually might help you.

Our colleague Nick Bloom, who studied this for a very long time in the econ department, has a pretty famous paper that shows that call center workers — I believe it was in a travel agency in China — were 12 percent more productive when they worked from home than when they worked at the call center. And so, the conclusion from that for many years was, yeah, you’re more productive when you work from home.

But it turns out if you think about what that task is, it’s not that you’re at home that made you more productive. It’s that the task requirements matched the kind of environment that you have at home. What do I mean by that? If I’m a customer service agent sitting in a call center, I have to do two tasks, actually. The first one is to listen to the customer, and the second one is I have to suppress all the noise that’s around me. Maybe there’s not a ton of noise, but there’s some noise. I got to suppress that. So, now that’s going to naturally hurt how productive I can be because my attention is split.

Now imagine I go work at home. At home I can actually control the volume of the environment around me. Usually I’m working in quiet, right? And because I’m working in quiet, I don’t have to suppress these other noises, and that allows me to give more attention to the customer. So, it’s really about what does the task require and how the requirements of the task match the modality of the work environment.

And so, that’s what managers have to do. They have to ask themselves, okay, what really requires us to be in the office because it does require expansive thinking, because psychological distance does matter for that kind of stuff. What could be done better at home like something like that where actually being in the office could be a distraction rather than an aid. And what type of things does it not matter. And then I can let the employee really decide whatever it is that they prefer, right? So, I don’t know. If I’m doing my taxes, it might not matter where I am as long as I can handle some distraction while doing my taxes.

Kevin Cool: Right, right.

You’re listening to If/Then, a podcast from Stanford Graduate School of Business. We’ll continue our conversation after the break.

So, we’re starting to see companies, and tech companies in particular, starting to try to pull people back in, in some cases, for a third day, for example.

Jonathan Levav: Yeah.

Kevin Cool: Is there any evidence or are you aware of conversations, policies, that are happening in these organizations about what the criteria is for that, or is it more about they think for some reason philosophically that it’s better to have people in the office than not?

Jonathan Levav: I don’t know of any systematic study that looks at various categories of task requirements in the workplace and matches them to workplace modality. So, right now all we’re seeing is basically a tension between employers that want to have control over the production function and employees who would prefer to avoid commutes and are more happy to work in pajamas than they are to get dressed in work clothes. And they would rather be able to go to the gym at 11:00 in the morning as opposed to at 6:00 pm after they leave work.

Kevin Cool: Right. And you’ve put your finger on what that tension is. In one case it’s kind of lifestyle, and in the other case it’s production and religious belief, if that’s the case. So, assuming that organizations want happy employees, they want to retain them — and maybe in all cases they don’t, but let’s assume for a moment that they do — what are some ways they can convince people to come into the office rather than just stating three times a week for reasons that may be nebulous?

Jonathan Levav: Yeah, so there’s a little bit of an underlying assumption to what you’re saying is that what I want is I want to keep all the employees that I have because I’m motivated to retain my employees. And I think that’s only a partial yes. I think that what we’re going to find is that there’s some employees that maybe they just don’t fit the workplace modality. Like if I’m a fully remote company, it requires a certain type of employee. And if I’m a kind of company that wants to have people mostly in the office, then you have to have people who are willing to do that and who actually prefer it or who embrace it.

And so, I think that you can imagine there over the next couple of years being like reshuffling depending on people’s preferred work styles. Now there’s no doubt that it’s more convenient for me to stay at home than it is for me to be in the office. So, I think that what employers need to start thinking about is to ask themselves what’s the job that being at home is doing for people that they’re so rabid about it. And I think that job is convenience, it’s avoiding commute, it’s flexibility, it’s a sense of familiarity and comfort and safety. And so, you have to start thinking as an employer, how do I provide that in alternative ways.

So, right now a typical office, a day starts at 8:00 or 9:00. But guess what else starts at 8:00 or 9:00? Schools, which means that there’s always going to be more traffic around 8:00 or 9:00, which means something that should take me 15 minutes can take me 45 minutes or an hour, in some cases an hour-and-a-half. So, maybe we just have to start the workday at 10:30 or at 11:00 or some kind of staggered form, okay?

Maybe if we know the people who like the flexibility of being able to work out and go to their Pilates class at 1:00, actually make that part of the workplace norm and say explicitly, “We want, almost expect you to go exercise once a day whenever you want as long as it fits your workday. Here’s a gym membership that allows you to go at any number of gyms in the area. Right? But the kind of tasks that you engage in, really it requires you to be in the office. We want to have a nod towards the things that made home such a convenient place for you.”

Kevin Cool: What would you advise a manager or a senior leader who’s trying to find that sweet spot between the mix of office and work, and what should they be thinking about when their people are remote?

Jonathan Levav: I’m going to give you an unsatisfying answer, which is an answer that’s actually going to require you to answer some questions, not you specifically but the listener. And so, the thing that people have to understand is that all those things that you learned about how to lead in a face-to-face environment pre-pandemic, because everything we’re talking about now is pre- and post-pandemic that skillset and those norms that were developed in the workplace, that culture, those forms of communication, those habits, that transfer of knowledge, you have to relearn it and redo it in the remote work experience.

And what you have to do, actually, is you have to engage in a process that says, okay, what were the elements of the culture that I had that were critical to me, what were the things that we did that were carriers of that culture, and then say, okay, if I’m in a hybrid environment or a remote environment, what do I have to do in order to be able to create that same kind of culture or that same kind of style or whatever, that same kind of workflow.

So, before, one of the ways we used to connect was everybody would end up going to coffee at 9:00. We’d all converge for coffee at 9:00. Now that we’re working far away, we actually have to deliberately set up an alternative. In March 2020, we moved from working face to working from home, we just kept on doing the same stuff, except now we’re talking to each other through a screen.

No, but then there’s all these other things that we did around that were essentially the scaffold around which work was done and then fell by the wayside. And we actually need to actively replace those because it’s all those scaffoldings that actually make the work environment meaningful for people. And so, I can tell you, teachers here at Stanford, I remember essentially one day I taught face-to-face, and then the next class I was showing the same slides and doing the same things online, and it doesn’t work. It doesn’t work because when you’re online, people tend to drift.

When I’m face-to-face, I can see when they’re on their phone. I can see when they’re on their email because I know where their eyes are going, right? But when I’m working on Zoom now, nobody’s really looking you in the eye on Zoom because the camera’s in a different place. It’s a different place than the screen. So, it always looks like you’re not looking at me even though you might be looking at me, right? So, we’re always miscommunication and there’s no way to fix that unless some kind of AI fixed the gaze.

But at some point I learned that, wait, if my class is all about engagement and engagement is important to me, when I was face-to-face I could create engagement by walking towards somebody, by changing inflection in my voice, by moving my arms in a certain way that would make people understand just like a conductor that uses his arms or her arms in order to be able to really move the orchestra. I can do that face-to-face. On Zoom I can’t do it; sometimes you don’t even see my hands.

So, I had to make a transition. I said, “You know what? If leaning in is such an important element of my class environment, I’m just going to cold call.” And I would have entire 80-minute periods where I would cold call exclusively. There was no hand raising. And what that meant was that people knew that any point in time I can call on them, so then they had to lean in.

But that was an adjustment I had to make, and I had to make a conscious change. You can’t just default into it; you have to make a conscious change. And so, to really manage in this new environment, you have to stop and say, okay, what are the things I care about? Okay, now how do I enact those in this remote situation? Because what you used to do, it doesn’t work, right? It’s not the same stuff. There are not the same features of the environment.

Kevin Cool: Right. So, let’s pivot just a little bit. You’ve done a lot of research on consumer behavior, how to influence people, decision making.

Jonathan Levav: Sure.

Kevin Cool: Are there any parallels from that research to what we’re talking about now with respect to managing a team and incentivizing or getting them to make decisions or think about something differently? How does creating a situation where they are evaluating this thing as compared to that thing as an option? What’s the process that they are engaged in their decision making and how can that be manipulated, if that’s the right word, or changed, or how can you lead them to a decision that you want them to make?

Jonathan Levav: So, what you’re describing is something that in the literature is called choice architecture. And the idea in choice architecture is that you — Think about it broadly this way. You have two choices. Suppose you want to get someone to engage in a certain behavior. You can either try to convince them to engage in that behavior, and then hope that when the decision point comes, they choose the option that’s congruent with the thing you try to convince them to do. So, that’s called the persuasion approach.

The other approach, the approach of choice architecture, says, wait, if I designed a decision environment taking into account people’s psychological tendencies and the way the mind works, then I can influence people’s decisions by virtue of the situation, right? Sometimes I tell my students, I don’t have to admonish them to do something. I put them in a situation where the decision I want them to make that’s better along some dimension that I’m trying to maximize is a natural outcome of the psychological process. And I do that, and it’s not that I’m incentivizing to do thing versus do another, but I’m creating a situation where they’re more likely to do one thing than another.

One of the philosophers of business that I love quoting in my class is the rapper Snoop Dogg. And Snoop Dogg says that it’s too easy for kids to join gangs and do drugs. We should make it easy to do football and academics. He’s absolutely right. He has the intuition for choice architecture: make easy the decision that you want people to make, and then they’re more likely to make it.

If I tell people that it’s really important to recycle, but to recycle they got to go to their garage to throw something out, they’re not going to do it. But then when garbage can manufacturers started doubling up the cans, right? So, now you have a can for trash and a can for recycling. Now all of a sudden people recycle. Why? Because you’ve made it easy. That’s the idea in that body of research.

Kevin Cool: So, you talked about and you used Snoop Dogg as an example of make the decision easy.

Jonathan Levav: Yeah.

Kevin Cool: How do you make the decision easy for people to come back to work?

Jonathan Levav: I think that if you make work be a place of greater meaning for people above and beyond the actual work itself. People search for meaning in their lives. They search for connection. They search for relationships. They search for opportunities to express their identity. They search for opportunities to express their individuality. They search opportunities to show off. They search for opportunities to relate. You want work to be the place that they naturally think about as the outlet for all those desires.

And so, you have to build the work environment with those desires in mind. And so, then you become easier. Then you also become easier on some issues like conveniences. If my commute to work is short as opposed to long, now all of a sudden it’s easier. I’ve lowered the cost of going to work.

If work provides me things that I can’t get at home. So, we talked earlier about gym memberships. If work gives me that or has a gym in the office, now all of a sudden I don’t have to leave the place where I’m at in order to be able to get a workout in, assuming that I value the ability and the time to exercise. Then that makes it easier.

I think tech companies did that to some extent with food. For a long time, the food experience at a lot of these companies was extraordinarily central. So, it makes it easy to sort of — that solves my lunch problem or my dinner problem or my breakfast problem. So, those are ways to make the office decision an easier decision.

Kevin Cool: Yeah, yeah. You talked about working from home in your pajamas. I showed up once in workout pants because I went to exercise and forgot to bring something else, and my boss said, “Nobody cares. Wear what you want.”

Jonathan Levav: That’s great, right? So, now he allowed you — or she, I don’t know — to express yourself in the way you want. And so, now he’s made being you easy. So, you’re you vis-à-vis other people. When you’re at home by yourself, there’s no other people. So, at least the office now is an easy way to be able to express yourself.

Kevin Cool: What interested you in this kind of research in the first place. Why are you studying the things you’re studying and why did you choose the career you chose?

Jonathan Levav: It’s a funny story and it goes back to my mother, but not my mother as in my mother inspired me to do it. I called my mother when I was a junior in college considering what I was going to take in the spring semester. And I was going to take five classes and I had chosen four.

And I called her and I talked to her and she says, “Well, what classes are you taking next term?” And I said, “Well, I’m taking [unintelligible] 1, 2, 3, 4. And the fifth one I’m not sure if to take this one or this other class.” And I honestly don’t remember which two classes I was considering, but one of them was a class called the Psychology of Decision Making and Judgment. It was PSY 321 WWS 312, that was the code. And she said, “Who’s teaching that?” And I said, “A guy named D. Kahneman.” She goes, “Danny L. Kahneman. Oh, you should take that class. He’s really famous.”

Kevin Cool: Okay, you got to tell me why she knew that.

Jonathan Levav: So, my mother’s a neuropsychologist, and she said, “Yeah, he’s really famous.” And I was like, so what do you do? I was like, sure, I’ll do whatever my mother tells me. And so, I took that class literally because my mom said, “You should take that class; that guy is really famous.” And so, I took that class and I just loved it. Yeah, it was the second year that Kahneman taught at Princeton.

Kevin Cool: What’s the next question you want to have answered? What’s the next thing you want to study?

Jonathan Levav: I don’t know. Listen, the next great revolution of our time is artificial intelligence. And I think that once we get past this like, “Oh my God, wow, look what ChatGPT just wrote for me” phase, we’re going to have to start understanding the psychological interface between the human being and the machine.

And so, we talk a lot about copiloting, and copiloting can mean a lot of different things that can affect us psychologically in a lot of different ways. And by the way, when I’m in a cockpit of a 737, there’s two pilots. There’s a captain and a first officer. Is the machine going to be the first officer or is it going to be the captain, and how does that change — and when, and how does that change things? So, that’s clearly one place.

Another place that we really want to keep pushing is hybrid work is here to stay, and remote communications are here to stay, and want to dig in further into that. So, we’re studying a bit now about people’s emotional responses to each other face-to-face versus on video. And eventually what I want to really do, I want to study networking. Because a lot of what we do at work is network. And if the type of information that we exchange differs when we’re face-to-face versus over video, then it can change the type of relationships that we build.

And so, the question is how does that change those relationships, how does that change the information that’s exchanged, who does it affect the most and in what way. So, I think those are the directions that I’m going to go in, but we’ll see. I’ll ask ChatGPT.

Kevin Cool: Or your mother.

Jonathan Levav: Or my mother, who by the way at 79, soon to be 80, still works fulltime as a neuropsychologist. Yeah.

Kevin Cool: Thanks, Jonathan.

Jonathan Levav: Pleasure.

Kevin Cool: If/Then is produced by Jesse Baker and Eric Nuzum of Magnificent Noise for Stanford Graduate School of Business. Our show is produced by Jim Colgan and Julia Natt. Mixing and sound design by Kristin Mueller. From Stanford GSB, Jenny Luna, Sorel Husbands Denholtz, and Elizabeth Wyleczuk-Stern.

If you enjoyed this conversation, we’d appreciate you sharing this with others who might be interested and hope you’ll try some of the other episodes in this series. For more on our professors and their research, or to discover more podcasts coming out of Stanford GSB, visit our website at gsb.stanford.edu find more our YouTube channel. You can follow us on social media at StanfordGSB. I’m Kevin Cool.