🎧 Listen Now:

Our guest today is Rebecca Hinds, a Stanford PhD who heads The Work Innovation Lab by Asana, a think tank for a better world of work.

Rebecca studied remote work and AI and received the Stanford Interdisciplinary Graduate Fellowship, considered one of the highest honors given to doctoral students at Stanford pursuing interdisciplinary research. 

Her research and insights have appeared in publications including Harvard Business Review, New York Times, the Wall Street Journal, Forbes, Wired, TechCrunch, and Inc.

In this interview, Rebecca shares:

1. Neuroscience and AI: People have always been afraid of technology and innovation, and neuroscience shows us that people avoid what they fear, and AI is triggering our fear of the unknown, of being replaced.

Leaders must understand this and transition from people being scared of AI to creating optimism and excitement by highlighting AI’s potential benefits.

2. Three Gaps for Organizational Adoption of AI: According to Rebecca’s research at Asana, there are three big gaps between how executives and individual Contributors view the technology:

• Optimism Gap: Executives see the promise and potential of AI more than Individual Contributors do. 
• Transparency Gap: Executives think they are more transparent in using AI than they are according to individual contributors.
• Resource Gap: 25% of executives say they provide AI training, but only 11% of ICs agree.
  Leaders must close these three gaps to gain the benefits of AI in organizations.
  

3. AI: From Individual Productivity to Team Results: People currently use AI for admin tasks, content production, and data analysis. While beneficial, this creates a risk of focusing too heavily on individual productivity versus team productivity. Doing so can lead to “collaboration overload,” where people get buried in notifications and task assignments from AI.

In the future, we will witness an increase in the benefits of AI for teams and organizations. It will become better at assigning work to the appropriate person, at the right time and in a way that makes them more interested and engaged in their work. This will ultimately result in higher job satisfaction and retention rates due to the increased interest in the work assigned.

4. Solving the Meeting Overload Problems: Rebecca co-wrote the famous HBR article Meeting Overload Is a Fixable Problem.

This article highlights the Meeting Doomsday case study, which let people delete all repeating meetings before adding them back, saving participants up to 2.5 work weeks per year.

Cutting meeting time is important, as Rebecca shared, because it makes time for what matters, like creative thinking, relationships, and brainstorming.

5. Tech Stacks: Less is More: Rebecca’s team recently released a new study about tech stacks in partnership with Amazon Web Services, acknowledging that we probably use too many platforms.

Using too many platforms is counterproductive especially because of switching costs, the time we lose when going between platforms.

To solve this issue, the subtraction mindset is key: do less, not more. As humans, we are inclined to add, not remove. An intervention is needed here to jolt people out of the status quo.

6. The Innovation Score: Rebecca's Work Innovation Lab has developed a new AI-powered model for predicting a company's ability to innovate, which is critical for its success. The model is based on data from Asana collaborations.

Called The Innovation Score, it captures four key predictors of innovation: cohesion (how well people work together), velocity (how quickly ideas flow through the organization), resilience (how stable or robust your organization is as people switch teams), and capacity (how much bandwidth do your people have to their best work.)

7. And finally, a call to action from Rebecca: work needs to change, so be bold! Grab a couple of people and test better ways of working.

🔔 Available on:

You can find the full episode and transcript here:

Transcript:

Daan van Rossum: I'm so excited! I know you just came back from London, where you did a conference on neuroscience and AI. Can you share a little bit about what you presented and what that conversation was about?

Rebecca Hinds: This was a talk I had the pleasure of doing with Shivvy Jervis, who's an amazing thought leader over in London and globally now.

We had a conversation around neuroscience and technology, in particular neuroscience and AI. I've long been fascinated by the psychology of technology, and I think neuroscience in general can help us explain a lot about how technology is adopted, whether it's embraced by workers or employees. I think the most important thing is that neuroscience can help us understand humans' natural resistance to change and change in technology.

I've given this example a few times, but when we look at the history of technology, we see it time and time again with every major technological leap. We see fear, uncertainty, and resistance. When the telephone was first introduced, people thought it was capable of transmitting evil spirits.

When electricity was first introduced, people thought it could cause internal human combustion. We see this fear and uncertainty in our research with AI as well. I think the resistance stems from fear of the unknown, replacement, or obsolescence. What neuroscience shows us is that fear is a natural brain response to a perceived threat.

The challenge that leaders in particular and organizations face right now is to manage this fear and transition from their employees seeing AI as a threat to viewing it as a tool that can really amplify their skill sets.

Daan van Rossum: We all see the potential benefits of AI and the way that AI could hopefully make work better, but then there is that resistance and fear. When you talk about neuroscience, that means people putting people into MRI machines while they're on their laptop doing AI work. What does that look like?

Rebecca Hinds: I think it can, and there's been some fascinating work by many different companies in terms of using brain imagery to understand how we respond to different interventions and changes.

But I think, most fundamentally, it's about using our knowledge of what happens when people are put into new environments to understand their natural resistance to change. I think I see it all the time right now, especially with AI, that there is a halo effect associated with AI, and for good reason, we're seeing a lot of promise and a lot of potential around the technology.

But what we see in organizations is significant gaps in terms of how executives are perceiving the technology versus how individual contributors are interpreting the technology and viewing its impact on their workflows and day-to-day work. I think using neuroscience as a lens to understand that there is fear, uncertainty, and anxiety, especially at those lower levels of the organization.

We know this is natural. We know that we can't shove it under the rug. We need to address it and help our employees move from that place of fear to a place of excitement and optimism. That requires hard work and change management, and I think it's really helpful to look back at the history of technology and know that this is natural, and we need to do the work to be able to fully get that organizational-level adoption of AI, which we know will be critical to harnessing its full potential.

Daan van Rossum: So you were saying that senior leaders and other people in the organization look at it differently. What are some of the differences that you see?

Rebecca Hinds: We see three key gaps in particular. 

The first of which we call an optimism gap. When we ask senior executives, how optimistic are you about AI? How likely do you think AI is going to help you reach your company goals? Executives are much more likely to see the promise and potential in AI compared to individual contributors.

The second gap we see we call a transparency gap, and this one I think is really interesting and important. When we ask executives whether they've been transparent in terms of how they're using AI and how they're bringing AI into their organization, executives are much more likely to say that they've been transparent compared to what individual contributors say.

Then the third gap, which I also think is critical right now, is what we call a resource gap. When we ask executives how much training or how much learning and development they've allocated towards AI and whether they're truly offering that training to employees, they're much more likely to say yes compared to individual contributors.

So, I think these disconnects are pretty startling to see, especially when we look at the resource gap. We see about a quarter of executives say that their organization offers training and learning and developments around AI, but only 11% of individual contributors agree with that sentiment.

There's a significant gap across these three different dimensions. It's going to be critical to the change management approach and to effectively harnessing the full potential that AI can offer.

Daan van Rossum: What are some ways that you see AI being integrated into companies right now? I'm really interested in it. What is it now? And you're obviously way closer to this than most people. What do you foresee for maybe one, three, and five years into the future when it comes to AI in the workplace?

Rebecca Hinds: It's a big question, but I think it's interesting because when we look at right now and what we see AI being used for and how we see it being adopted, we see that it's rather predictable in terms of the use cases that employees are gravitating towards. So, we see three at the top of the list in our research: admin tasks, content production, and data analysis.

What's common about those three use cases is that they are all very much oriented around boosting individual productivity in the short term. I think that makes sense, especially in a world today where productivity is under a microscope. Humans are more likely to gravitate towards those short-term use cases where there's a quick win in terms of boosting their individual productivity.

I think that when we look to the future, we're going to see much more of a focus on team and organizational-level productivity gains associated with AI. I think what we see in the research and what we're starting to see in some of the companies we work with is that an overwhelming focus on individual-level productivity can come at the cost of team-level productivity and organizational-level productivity.

In research, it's sometimes called The Tragedy of the Commons, where the collective good is sacrificed for these individual productivity gains. The example I often give is that an employee can use AI to become highly productive as an individual. So they can auto-assign tasks to other people. They can delegate their work.

They can send notifications and pings, but that can lead to a storm of what we call collaboration overload—unmanageable demands for other people on the team. If they're not taking into account other people's levels of burnout, bandwidth, and priorities.

I think what we're going to see in the years to come is that vendors and organizations focus much more on team and organizational-level productivity gains, which AI has enormous potential to help us really crack in ways that we haven't been able to in the past, just because of the sheer volume of data.

Daan van Rossum: That's a super interesting shift. What are some of the ways that AI will then contribute to being more productive as a team? Because I know you study this kind of stuff, do we just talk about productivity in terms of output, or are we also talking about the quality of that productivity? What are we actually doing? What are we actually delivering? Are we solving a problem? What are some ways to think about that? 

Rebecca Hinds: At the Work Innovation Lab in particular, we are pretty laser-focused on collaboration as well, and AI has enormous potential. We're starting to use AI ourselves to understand what those helpful and harmful collaboration behaviors are that lead to different outcomes. Sometimes it's productivity, employee engagement, or employee growth, learning, and development.

I think there are several exciting potentials for AI. One is matching individuals to tasks in a way that we haven't been able to in the past. So you can imagine a world where we look at all employees within a team or an organization. We take into account their skill sets, career ambitions, and history of working on different tasks and with different people.

We can start to work much more intelligently as new projects are spun up, and we can put the right people on that team to execute the work most effectively. I think we were talking before we hit record about meetings and how broken meetings are. You can imagine a world where we intelligently take into account people's level of workload, when they're most energized throughout the day, when they're in the office versus not, and start to put together much more effective and intelligent meeting calendars.

So there are a whole host of different outcomes. I think when we think about collaboration too, we should be able to proactively tell individuals when their collaboration with other people or other teams is declining, when they might need to schedule that meeting, schedule that touch base, or schedule that status update, or if there's been a lag in communication, and again, start to more intelligently work with other people in a way that we haven't been able to in the past because of the sheer volume of data.

Daan van Rossum: You would get some much more helpful diagnostics, let's say as a team leader or a manager. Is everyone really utilizing their full potential? Are they doing meaningful work? Are they doing the stuff that they want to be doing? That sounds almost like we're moving towards an internal talent marketplace where, let's say, AI knows what I really love doing and spots an opportunity for me to contribute something. That otherwise would have been completely unknown to me and unknown to the organization and make that match.

Rebecca Hinds: There are so many factors at play. It's employee engagement; it's retention. You can imagine a world where, yes, assigning you to this one task might boost your productivity or might be in short-term service for the good of the company.

But it might not be interesting or exciting to you. And if it's not, then probably you're going have lower engagement, you might stay at the company for a shorter period of time and you can start to use AI to do this complex calculus that we haven't done in the past.

Daan van Rossum: You mentioned one of our keywords, which is meetings. I would love to go back on that because you wrote the very famous HBR article, “Meeting Overload Is A Fixable Problem.” Whether it's AI or thinking differently about meetings, what are some of the ways that we can fix meetings and all have more time to do actual, meaningful work?

Rebecca Hinds: A bit of background on that article. I wrote that in collaboration with one of my mentors, Bob Sutton, at Stanford. It was based on a couple different interventions that we led at the Work Innovation Lab.

One of which we called Meeting Doomsday. It started as a pilot experiment with a small team at Asana. This was happening throughout the pandemic when we were seeing people in more meetings; they were in longer meetings, and their calendars were filled up by meetings that they thought maybe could be performed and executed more effectively.

We asked this small team to delete all their small recurring meetings from their calendar for 48 hours. Then, after 48 hours had elapsed, we invited them to re-add the meetings back to their calendar. But do so in the way that they thought was going to be most valuable. So they could change the length, they could change the cadence, they could change the number of attendees, and they could delete them entirely.

This was incredibly successful, beyond our expectations. Each participant in that pilot experiment saved 11 hours per month. About two and a half work weeks per year just by eliminating those small, unproductive meetings and going through this 48-hour activity.

It was shocking in terms of the time savings gained back. Bob and I are both adamant about the fact that meeting efficiency is important, but it's important because it enables us to have more space to do the things at work that should be more inefficient.

So creative thinking, brainstorming, and developing relationships with each other and with our customers. These are the things that we want to make space for, and these are the things that shouldn't be highly efficient, but meetings should in general be efficient, and they're broken in organizations today.

They're ubiquitously considered a time sink. They are ubiquitously considered to come at the cost of deep thinking, creative thinking, and collaboration. It's an opportunity, especially with AI now, to challenge some of our core assumptions about when we need to meet, why we need to meet, and rethink the status quo.

Daan van Rossum: What are some recommendations that you would give either to individuals, to managers, or maybe to company leaders? How should we look at meetings? Should we all do this Meeting Doomsday and get a better schedule?

Rebecca Hinds: I've gone through so many meeting interventions and studies, and the most common one is the meeting audit. To take an audit of your calendar, look at which meetings are effective and which are not, and make some sort of judgment call incrementally in terms of which meetings you want to change.

We've done this study where it doesn't lead to the same impacts as this Meeting Doomsday because it doesn't fundamentally challenge our core assumptions about how meetings happen.

I do think we've done it; we've just finished a study where we took a similar approach to tech stacks and did a doomsday on tech stacks. This type of complete reset—a fresh start per se—does encourage this healthy behavioral change. Bob Sutton has studied what he calls the subtraction mindset quite extensively.

This idea is that, as humans, we're naturally inclined toward addition; we're inclined, especially in a workplace setting, to add meetings and processes.

In particular, meetings and technology are panaceas for so many different things. This healthy behavioral change of “let's do a reset" in 48 hours is going to do so much more help within the organization than harm. It can be scary. It can be uncertain, but there's no better way, in my experience, to jolt people out of that status quo and encourage that mindset shift.

Then, I think there's real value in no meeting days as well. We see from the research that no meeting days consistently have a positive impact across a diverse set of factors for organizations. I think there are common skeletons for meetings. You need an agenda, and you need to have follow-ups.

But I think fundamentally doing a Meeting Doomsday every six months or every year is an incredibly helpful practice. If nothing else, then to encourage that mindset shift, we need to be consistently challenging the status quo and rethinking some of these work practices that we know are broken and equated.

I think another good example of this is that meetings often default to weekly cadence. They so often default to 30 minutes or 60 minutes, and it's so arbitrary that the time of length and the cadence, and when we look at our research, we saw that when people participated in the Meeting Doomsday, they often changed the length of meetings to be unconventional length.

They changed 30-minute meetings to 25-minute meetings. They changed weekly meetings to monthly meetings, or every six weeks. It encourages us to also not just default to what our calendars project in terms of what should be a meeting, but think consciously about the content of the meeting, who's involved, what's the outcome, and what's the most appropriate length and cadence for those different factors.

Daan van Rossum: Definitely. That's why I love when Microsoft makes these small changes, or we use Google, make these small changes about setting the default value of a meeting to 25 minutes. They probably impact more people than anything. If they can just get people into the right behavior. I also love that you said to “jolt” people out of behavior, and I love that you use the word “intervention” because we can theoretically always just think about auditing our meetings. Do we still need them? But it's very theoretical. I like the idea that we're going to do an intervention. You're going to start with no meetings and then maybe build them back up, but you're going to look at them very differently.

Daan van Rossum: You just mentioned something about a similar research study around the tech stack and how we also often try to get new technology in to solve problems. Can you share more about that research? I don't know if it's public, but it sounds really interesting.

Rebecca Hinds: We did just publish a HBR on this as well. It was done in collaboration with Bob Sutton as well as a fantastic professor at UC Santa Barbara, Paul Leonardi, as well as in collaboration with AWS, Amazon Web Services.

It was largely inspired by Meeting Doomsday. We thought, can we apply this subtraction mindset to tech stacks? And we see, especially throughout the pandemic, a ballooning of tech stacks within organizations.

And I think that was, again, this example of what we lost in many cases: the physical connection to people; we lost in many cases the opportunity to have one-on-one interactions with people in person. And so the natural response for many organizations was to invest more technology and, in particular, more collaboration tools, and we see so much context switching and so much collaboration overload in organizations today.

We were trying to better understand this problem and what organizations can do to help their employees reduce collaboration overload. So it was a group of Asana employees and Amazon employees where we asked them. We had them in 2 different groups, but essentially, we asked everyone within the experimental intervention to stop using a certain number of their core collaboration technologies for 2 weeks, I believe. So we asked them to stop using their technology. They had to log any time they deviated from their approved tech stack, and we saw significant changes in how people worked.

We framed the article in terms of the fact that there was good news but also bad news.

The good news: was that people in general became more mindful of this technology overload. They reflected, and we had them complete surveys throughout the experiment, consistently reflecting on the value of this type of intervention and truly rethinking all the different technologies they use at work.

We also saw that more than half of our participants ended up saying that they could eliminate one or more collaboration technologies from their core tech stack by the end. So in just two weeks, it's made pretty significant changes.

But the bad news: This was inspired by Professor Leonardo's work. We measured something that we call Digital Exhaustion. So this is how exhausted the technologies you use at work make you feel, and what was surprising to us is that we saw that for participants, digital exhaustion actually increased throughout the study.

As we were pairing this finding with the survey data, we realized what was happening, and that was that as people were asked to reflect more deeply on the technologies they use at work each day, each week, they became more aware of just how exhausting that context switching is and that use of so many different technologies and the guesswork that goes into understanding what work to do in which technologies.

So, we saw their digital exhaustion increase as well, because there was this recognition, unlike the meeting work, that interventions related to technology need to be much more top-down because there's so much systematic nature when it comes to technology and technologies are so interdependent.

My technology use is interdependent on what my customers use, what my direct team uses, and what my cross-functional team uses. In contrast to meetings, which are still pretty interdependent but not as much as technology, where there are deep-seated histories that need to be taken into account, And so one of the reasons why digital exhaustion increased as well was because there was this recognition that the employees themselves could not enact all the change they wanted to in terms of rethinking their tech stack and reconfiguring it.

It really did require that team leader or executive-level intervention, and so when we think about the implications, it's definitely true that leaders need to play a more proactive, more active role in helping their organization understand what that ideal tech stack is and when it's becoming too big. Doing a tech audit and a tech reset every six months, just as you might do for meetings, can be a really healthy practice.

Daan van Rossum: Rebecca, there are so many amazing research findings that are so applicable, like they're immediately applicable to the way that we work.

Now I noticed you guys also recently launched a work innovation score. Maybe you can share a little bit about that. Again, what can we learn from that in terms of how we work? 

Rebecca Hinds: Thank you. The work innovation score was inspired by the recognition that executives right now have an extreme hunger for more visibility into how work is happening.

We ran a survey about a month ago looking at the North Star metrics that matter most to executive leaders today and the common ones that top the list. It's productivity. It's innovation. It's employee engagement.

But when you ask the same executives whether they actively measure and track those north stars, there's a significant gap. So a lot of the work innovation score was inspired by the opportunity to give leaders and organizations more visibility into how work is happening and whether that work is contributing to positive outcomes.

We launched the Work Innovation Lab about a month ago. The development of the score was led by an individual on my team, Dr. Mark Hoffman, who used to be a Stanford professor and is skilled in understanding the sociology of work, organizational network behavior, and how people within organizations collaborate.

So this work innovation score we developed is a score out of a possible 100. It's powered by AI by neural networks. So we've done extensive research in terms of understanding what those core predictors of innovation within organizations are. This specific score is designed to assess a company's propensity for innovation based on how they're collaborating and how they're collaborating on Asana.

What we find is that there are four key drivers of innovation within companies. The first is cohesion, so how well are your employees working together? Velocity is a measure of how quickly ideas and work flow through the organization. Resilience, which I'm hearing top of mind for executives today, is essentially a measure of how stable or robust your organization is as different people switch teams, as teams are disbanded, or as people leave the organization.

How quickly can work be picked up? How quickly does work break down in those instances? And then the fourth one, which we've spoken about, is capacity. So how much capacity or bandwidth do your employees have to do their best and most important work?

We see that those four drivers matter most for innovation, and we're able to use those drivers to understand and assess a company's potential for innovation and package it as this work innovation score.

Daan van Rossum: Just to close out, like any final thought on the future of work, so you're studying this very up close, maybe more than anyone else, what is something that you wish for when it comes to work in the future?

Rebecca Hinds: I think there are a couple of things. One, we've talked quite a bit about AI, and I think one of my hopes and wishes for AI is related to this concept that is being called human-centric AI.

This idea is that as we bring AI into our organizations, we can't just think of it as a technology that we're adopting and rolling out to employees. We really need to position it in a way that is in service of humans and helps amplify our potential as humans.

One of the most interesting findings from our AI-specific research is that we see right now that more than 40% of employees say that they're choosing their next employer in part based on whether that employer has adopted this human-centric AI approach.

It's incredible and really promising because we know that the companies that don't adopt that human-centric approach are going to be at a disadvantage in terms of talent recruiting and retention. 

My hope is that we see AI as first and foremost an opportunity to amplify and augment human potential and as we advance AI technologically, I think the true measure of progress should be how well those innovations and that advancement support and augment human potential. And not only building smarter machines but also creating work environments that augments our abilities as humans and our catalyst for human growth, learning, and well-being.

Then, I think more generally, a lot of our research at the Work Innovation Lab focuses on challenging the status quo and rethinking some of these core aspects of work that we know are broken, antiquated, or outdated and so my hope is that as we continue this journey, organizations continue to challenge the status quo in terms of how work happens and be bold even if it's on a small scale, even if it's taking 20 employees and testing what the world would look like if we fundamentally rethought some of these practices that we rely on without second-guessing them.

Daan van Rossum: I love that. Be bold and challenge the status quo. Rebecca, thanks so much for being on. This was amazing.

Rebecca Hinds: Thanks so much, Daan. I love the conversation.