Cultivating an Experimental Mindset in Your Organization

[MUSIC PLAYING] Welcome to HBR on Leadership. Case studies and conversations with the world's top business and management experts hand selected to help you unlock the best in those around you. I'm HBR senior editor and producer Amanda Kersey. As a leader, you face uncertainty all the time. Experiments offer a weighted test assumptions, but it's not enough to simply run them. Their value comes from designing them carefully and being willing to act on what they reveal, even when the findings up end your expectations. Here's HBR idea cast host Curt Nickish with a conversation from 2020 about what leaders need to know to design rigorous experiments and then put the evidence to work. In science, the need for experimentation is cut and dry. You come up with a hypothesis, whether it's about how storm clouds move or how cells in the body die, and you set up an experiment to test it. There's a method, it's called the scientific method. And you test it over and over again until you're sure that it's replicable and your answers are right, or at least as right as they can be until new variables come to light or the landscape changes. In business, there isn't currently as much experimentation. Value has been placed on experience, on the intuition of managers and leaders. And that's a bad thing, says today's guest. Even in the most innovative industries we can think of, more can be done to set up experiments, test the results, and deliver better products and services to customers. And this goes far beyond AB testing at TechChance. Our guest today is Stefan Tomkitt. He's a professor at Harvard Business School. He's the author of the book Experimentation Works, the surprising power of business experiments. And he also wrote the HBR article, Building a Culture of Experimentation. Stefan, thanks for coming in. Thanks for having me. Just to start, pretend I'm a business leader. Make the case for me. Why do we need to experiment more in business? Well, first of all, it can generate a tremendous amount of value. Let me give you an example. Microsoft's Bing, which is its search engine. Sure. An employee working sort of at Bing came up with an idea on how to sort of display sort of ads. The manager didn't think much of it, and they kind of shelved it. But the employee insisted. At some point, the employee decided just to launch an experiment to run a test, a control test. And when you run the test, that little change, a few days of work generated more than $100 million of additional revenue, and that year alone. And, of course, more revenue going forward. It was, in fact, the most successful experiment that was run at Bing. So, you know, what made the difference? Well, the difference was that the employee had the power, essentially, or the authority to run the experiment, to launch it, and to test it. It's the test that actually told you what works and doesn't work. And not the manager. And not the manager. The problem is in a lot of innovation, especially when you're trying to predict customer behavior, we get it wrong most of the time. And so, rather than trying to follow our intuition or our opinions, why not just run the test and let the test tell us what works and doesn't work? And what's the answer to that? Like, why aren't people doing it? Well, there are lots of reasons why not people are doing it at scale, especially. Right. So, some people are sort of running simple experiments, because they refer to an experiment as something like a trial. You know, we're trying something. That's not really an experiment, sort of, in the scientific sense. And they don't do many of those, you know, because they either don't have the infrastructure to run many tests. They may not have the tools, sort of, to do so. It may be too expensive to run it. And then they may decide that, listen, you know, we run a test and we get some results. And then nobody listens to us anyway. Right. Do managers overestimate the downside to experiments and underestimate the upside? I think sometimes they are too overly concerned about the risk of running the experiment. For good reasons, you know, you have a lot of traffic, you know, you may not want to launch something that, you know, results in a loss of, you know, customers visiting your websites, for example. Right. If it goes down. If it goes down, and so if you don't have good stoppage rule, you know, kill switches and things like that, sort of, in place and, and then maybe a risk of version, it's also stepping into the unknown. And quite honestly, it's, it takes humility to admit that I just don't know. You know, walking into a meeting and we're launching this thing and everybody has some hypothesis about what the outcome is going to look like. And just go into the meeting and tell everyone, listen, honestly, I don't know what's going to happen. So let's just find out. Even though I get paid more. We get paid. I mean, charge. I don't know either. Exactly. And the higher you go, the more you get paid. Yeah. The more senior you get, you know, you get paid to make tough decisions. And you want to be a decision maker. And yeah, like create sort of an organization that takes a little differently sort of to do the sort of thing. By the way, I mean, it's not just the online world. It's also the physical world, you know, where companies are running experiments. And even there, you know, we have to make big decisions, sometimes very expensive decisions. And it's the experiments that can, in fact, adjudicate, you know, whether we want to do something or not, calls, you know, big retailer and so forth. So calls hire a consulting company and the consulting company basically does a cost analysis and they go to senior management and tell them, listen, we figured out that you can save a lot of money. If you open your stores an hour later, now here you are. You're running this company and you have to make a decision, should we do that calculating the cost savings is easy. But the big question is, what's actually going to happen to our revenue? You know, our customers going to buy less if we open an hour later. So how do you make these kinds of decisions, you know, we can analyze and analyze. But we won't know until we actually do it until we run the test. And in this case, they did. And so they ran control experiments in which they sort of set up these tests, opening an hour later and lo and behold at the end, you know, the result was that it made, didn't make much difference. So just so we're on the same page, like how do you go about setting up an experiment? Are there playbooks for this? Well, first of all, there are tools. A lot of companies that describe in the book built their own infrastructure, built their own tools because when they got started, many years ago, the tools ran around. So you look at Amazon, a Microsoft, a Netflix, a booking.com, I mean, you go through them. And it was about a dozen or so. They decided to do it themselves. So they just, they knew that they had questions they wanted to answer and they just figured out a way to do it. They figured this is going to give them a competitive advantage, you know, if they can kind of go out and just test a lot. And they knew that they often get it wrong. And so, so they started investing in infrastructure and so at a place like Microsoft, for example, you have a very, very large group that basically runs the infrastructure, you know, something like the last time I checked it was something like 85, 90 people or so that are just sort of doing infrastructure. But the good thing that happened a few years ago is there are now third party tools as well that can do this, both in the online spaces and in the brick and mortar spaces, which do sort of a lot of the heavy lifting for you, a lot of the statistical stuff and so forth. And so, so it's gotten a lot easier than say, if you wanted to start the five or 10 years ago, developing a culture for this is probably a little bit different. I think it may be potentially harder than getting the tools and building the tools, because now we're dealing with behaviors, with beliefs, with norms and all sorts of things. How does this show up in companies if the culture for experimentation is not working? What do you, what do you actually see and observe? Well, the classical example is they start running experiments. We have an experiments. We hand over the result to the group that asks us to run the experiment and then nothing happens, or they will start to challenge the experiment, something must have gone wrong. I remember a story where, you know, an angry person actually called sort of one of the, one of the tool vendors sort of in this space and complained about the tool being wrong. The person ran an experiment that actually showed, if you give customers less choice in his setting, you get better performance. And that was kind of just counterintuitive because everything that he believed in up to this point is that you should give people more choices. And so he was really disturbed by the finding. And so he called them and complained that there's a flaw in the tool. Having the tool must be wrong because the result doesn't match with the experience that he's had and he's been doing this for a long time. And so you run into that sort of thing, which kind of underlines your point that experiments bring new insights that you just can't develop on your own. Correct. There's a company called Booking.com, which most of us use. In fact, it's the biggest accommodations platform in the world, and more than 1.5 million room nights are booked on the platform each day. It's a two-sided platform. This is what we call it. It's got suppliers on one side, which are hotel operators, for example. And of course, it's got customers like us on the other side. And Booking.com runs a massive number of experiments. My estimates are, and I'm probably on the low side they told me it's my estimates. It's over 30,000 a year of experiments. And it's a really, really fascinating company. It's also a highly successful company. Their gross profits are in the high 90s percent, and they don't really have any assets. They don't really own any accommodation. So it's a super competitive industry, too. And so how do they get away with this? And the answer to this is there run a lot of experiments, and they created an experimentation culture where almost running experiments is like breathing. You kind of do it every single day. I mean, you have to encourage, you have to think about the numbers here. Even if I'm running a lower number of experiments, I mean, they're running more than 100 new experiments a day. You have to have an organization that can even come up with so many hypotheses. I mean, you mentioned the number of transactions that Booking.com does in a day. How key is that to being able to run experiments? Does that also work for places that just don't have data like that? Yes, it works for places that also have a lot less traffic. The underlying math changes, you know, sort of what you have to do algorithmically is very different. In fact, if you have very large sample sizes, you know, a lot of traffic, for example, you can really find one. You can sort of do very, very small changes, and you can kind of pick up whether that change actually causes something to happen. This your sample size shrinks, you know, you kind of have to go for bigger changes. We call it the power of an experiment. You have to power an experiment, statistical power. And so I recommend for companies that are sort of smaller that maybe they kind of run experiments that are a little bigger. Now, what happens also, and this is something that actually happened at IBM, when they started to do this, they realized that they have way too many websites. So yes, they had very little traffic on some of these websites, but they didn't need all the websites. So they actually led to a process of consolidation. And I said, listen, we don't really need all these things, so what we'll do is we'll consolidate and we get sort of more traffic on fewer websites, it's which that allows us to sort of run more experiments. I wondered if there are companies or industries outside of consumer-facing tech or outside of scientific or pharmaceutical companies where experimentation really feels for it. Well, I mean, the classical companies, I think, are sort of in the creative industries where the assumption is that everything is driven by creatives, right? And look at entertainment, for example, and look at what Netflix has done. So Netflix kind of flipped it around, and they operate in the creative industry, but they are completely experimentation driven. Right. It's a big wake-up call for the entertainment industry, because when you go in and you run Netflix, you are part of their ecosystem, their experimentation ecosystem, they run a massive number of tests, because they want to find out what works and does work. By the way, running the test and getting a result doesn't mean that you have to blindly follow what the result is, because sometimes there are good strategic reasons why you may not want to implement what the test tells you. Right. Or there are trade-offs to whatever benefit. For example, or maybe there may be a contractual violation or something like that. But what the test does, it actually adds transparency to the decision. So you cannot pretend that we're doing this because it's good for the customer, or something like that, or good for the viewer. It adds clarity that we understand from the test what's good for the viewer, but there may be other reasons why we may not want to do it. And adding that transparency to what you're doing, I think, is sort of the big value and it allows a company like Netflix to operate really in the creative industry with a testing approach. Yeah. I don't want to diminish the value of creative talent because creative talent is really important, but that doesn't create certainty in terms of decision-making. To me, the creative talent and the intuition is an important part of experimentation because it allows us to create hypotheses. You have to ask yourself, Kurt, where do these hypotheses come from? Yes. Still from people. Some people asking questions are having ideas. Absolutely. Absolutely. When I'm saying is they're running all these experiments, they're all hypotheses that came out of product groups, and it's the people who come up with these hypotheses, and so whether they get the idea as well, it's intuition sometimes. It's inside surprising, customer surprises, things that thought that were true, and then they observe something that doesn't quite fit sort of what they know. It's usability labs, so there's still, I mean, these companies all run qualitative research. But they do all the kinds of things that other companies do, but they do it for generating hypotheses, which are then rigorously tested, versus other organizations that generate the hypotheses and go directly from hypotheses to launch. Right. Based on whoever is the best kind of speaker or makes the best case in a meeting rather than-- Yeah. There's a word for that in the community called Hippos. Hippos. Yes. Highest paid person's opinion. Uh-huh. Yeah, yeah. Hippos. And we all know that Hippos are very dangerous animals. I think a lot of executives are probably also not used to knowing how much experimentation to do. How do you know what to experiment on, and how do you know what to let be? Yes. You have to empower people to make that decision. And the reality is, right now, I think most organizations test to a little. So you know, I don't think you should be too worried about testing too much. Okay. Yes. There's probably a point in which you test too much because you need an organization that can absorb all that knowledge or all sort of that, uh, all those findings that are generated by all these tests. That's true. And we need to think about that, but I don't think that's the problem in most organizations right now, right now they're doing, not doing enough. If you're bringing this into a company, do you try to do this company wide? Do you try to start with a team or a division and scale it up from there? So the different ways to organize your experimentation teams, the three models that are described in the book, you know, one model is really more centralized approach. I basically have like a center, a group that's responsible for experiments and they're like a service organization where you can come from a business unit, you can commission experiment and they'll run it for you and they give you the results. Oh, that's interesting. That's one model. And a lot of companies start out that way because they are kind of a little uncertain how if this is all going to work out, they don't, they don't, they may not believe that the company is ready to do this at large scale. It probably simplifies training and it lets people, let's people dip their toe in without look without really having to exactly any of a few experts and they kind of make sure that people don't do foolish things. Yeah. Then the another form is to have a decentralized, completely decentralized. So now we're shifting the autonomy basically to people and allow pretty much anybody to run experiments and and we don't centralize it anymore. And of course, there you have to trust people. You have to know that they're actually capable of doing this and and it's a way of course to rapidly scale things. But what happens there is when you start to put all these, you spread all these sort of your experts around and they're always sort of through the company, they get very busy. And you kind of lose the focus on building capabilities because you need to always kind of get better and better. And so there's no coordinated approach to this, you know, everybody kind of does their own thing. So what companies have found is they go from centralized to decentralized and they want to scale things, but then they realize that they need to have a more coordinated approach and then they create something which they call a center of excellence. And the center of excellence is kind of a hybrid model then where you have sort of a core group that actually is responsible for developing capabilities, experimentation capabilities, kind of know what tools to use and push the envelope. But at the same time, you take people out of that group and then place them sort of into the different organizational units that are doing this and they're basically there to help as well. And the company's found that that's actually sort of a very good compromise because on one hand, you kind of empower people to do things on their own at the same time. You actually have someone who centrally owns this capability as well. How do you know when it's really working? You know the way you're really working, I think it's a cultural test and I tell you here's the test. You sit in a meeting and you're discussing a decision and you know when it's working, either when someone asks, "Where's the experiment?" or when someone actually walks into the meeting and says, "Here is the experiment." When these kinds of discussions are happening every single day, without you having to ask for these things, then you know things are kind of working. I call it, it's like running the numbers, right? When you go into a meeting, you always expect people to do some financial analysis. It's almost a given, right? So it has to be like that. It has to be like running a financial analysis. It has to be a given that you kind of do a test, you run an experiment, unless you've done it, you know, you know, we're not going to make a decision. Say you're an individual contributor, you may be a manager, you may be a frontline worker, but you buy into this, like you see the value of experiments, you want your organization to do more. What do you do to try to bring a culture of experimentation to a place that is still relatively new to it? What you can do as an employee is first of all raise the awareness around sort of you. What does that mean? That means basically explaining sort of the people what sort of the value of the experiment or what experiments are, but then also I think at the same time is maybe try to do some of these things in the areas that you control. You know, yes, I see the difficulty sometimes, and I hear this from people saying, okay, I get you, but you know, there are two levels up, you know, I'm not sure that they do. So what can I do? So I always tell them, start small, get going, and then this is what often what happens that I've talked to organizations that actually started this way and then got bigger and bigger. They said, you know, we started out and we run an experiment and we went to the meeting and we told people what the experiment sort of showed us and so forth and they kind of listened to it and they gradually started to sort of understand the value of it. But you got to get started, don't wait. What kind of manager is then the successful manager in a company that has a culture of experimentation? Because in the past, maybe it used to be people who had experience, people who had intuition. Now when you run experiments, what is the type of manager who excels and advances in an organization that has a culture of experimentation? So you can ask the question, if everything is adjudicated by experiments or by tests, what's the role of the manager anyway? Right. I kind of break it down into sort of three different things that they should do. First role, I think of a manager is to set a grand challenge. What we don't want to do is we don't have an organization that just does experiments willingly with no direction. So there needs to be a grand challenge. A grand challenge, for example, could be, we want to have the best user experience in the industry. And that grand challenge then can be broken down into different pieces, which then can be addressed with hypotheses which are then tested. So you give them a directionality that needs to be a program, a systematic program that sort of aims for some big or goal. So that's the grand challenge. The second thing I think that managers need to do, especially in this kind of environment, they need to place the system's resources and organizational designs that allow for the large scale experimentation to happen. Things like that don't happen by themselves. You need to invest in tools. You need to make sure that you've got the right organizational design to start out with and maybe then change it when things don't work. So you have to think about that as well. And you need to make sure that sort of all the systems are in place. So someone like that employee at Microsoft can just kind of push a button essentially and just launch and run this thing if it takes employees weeks and weeks to set up an experiment. What are the odds of them doing it at large scale? It's not going to happen. So you're going to make it easy as well and you need to empower sort of people to do it. You need to democratize experiments. And the third one is they need to be a role model. They need to live by the same rules. So when we go into a meeting and we propose a course of action and someone says, that's really nice. We'll run a test and let you know what happens. We need to then have the humility to say, let's do it and let's do it quickly. So we need to live the same way. We need to kind of do the same thing that we ask our employees to do. So that's a different style of leading. Stefan, thank you so much. Maybe we'll try some experimentation on this show as well. Thank you very much. It's great to be here. Stefan Tumpke is a professor at Harvard Business School. He's the author of the book, Experimentation Works, the surprising power of business experiments, as well as the HBR article, Building a Culture of Experimentation. HBR on leadership will be back next Wednesday with another hand-picked conversation from Harvard Business Review. If this episode helped you, share it with your friends and colleagues, and follow the show on Apple podcasts, Spotify, or wherever you listen to podcasts. And while you're there, consider leaving us a review. And when you're ready for more podcasts, articles, case studies, books, and videos, with the world's top business and management experts, find it all at hbr.org. This episode was produced by Mary Doe and me, Amanda Cursey. And leadership's team includes Marine Hook, Rob Eckhart, Tina Toby Mack, Erica Trucksler, Ramsey Cabaz, Nicole Smith, and Ann Bartholomew. Music is by Koma Media. Thanks for listening.