A conversation with Kai Haley, Head of UX Methods and Processes & Founder of Sprint Master Academy at Google


“A lot of times, people will think that they’re experimenting because they’re being scrappy and moving fast, but they’re not learning as they go. The fundamental benefit comes when you stop and reflect and learn and improve.” – Kai Haley

This week on the Control the Room Podcast, I’m excited to speak with Kai Haley. In addition to being a founder and leader of their Sprint Master Academy, Kai is Google’s Head of UX Methods and Processes. Through the development of design sprint and design thinking curriculum and training programs, she has created a community of expert facilitators at Google.

Listen in to learn more about Google Sprint Masters, the delicate balance between data and design, and how knowledge sharing creates resilient product teams.

Show Highlights

[5:27] The delicate balance between data and design.
[11:08] The advent of the virtual facilitator.
[21:35] Google Sprint Masters.
[26:13] How knowledge sharing creates resilient product teams.
[31:31] Embracing asynchronous meetings.

Links | Resources

Kai on LinkedIn

About the Guest

Kai Haley, Head of UX Methods and Process at Google, works with teams to define product visions, drive successful user-centered products, and develop an innovation culture across all of Google. As Lead of the Sprint Master Academy, she also develops curriculum and training programs for Google’s expert team of facilitators.

About Voltage Control

Voltage Control is a facilitation agency that helps teams work better together with custom-designed meetings and workshops, both in-person and virtual. Our master facilitators offer trusted guidance and custom coaching to companies who want to transform ineffective meetings, reignite stalled projects, and cut through assumptions. Based in Austin, Voltage Control designs and leads public and private workshops that range from small meetings to large conference-style gatherings.

Subscribe to Podcast

Engage Control The Room

Voltage Control on the Web
Contact Voltage Control

Podcast Sponsored by MURAL

Full Transcript

Intro: Welcome to the Control the Room Podcast, a series devoted to the exploration of meeting culture and uncovering cures for the common meeting. Some meetings have tight control, and others are loose. To control the room means achieving outcomes while striking a balance between imposing and removing structure, asserting and distributing power, leaning in and leaning out, all in the service of having a truly magical meeting.

Douglas: Today I’m with Kai Haley, head of UX Methods and Process for the UX community at Google. She co-founded and leads to Google Sprint Master Academy, where they train and mentor facilitators within Google. Welcome to the show, Kai.

Kai: Thank you for having me.

Douglas: Of course. So, Kai, let’s share with the listeners a little bit of a backstory around how you became such a mover and shaker at Google.

Kai: I joined Google about nine years ago. And prior to that, I had been at Yahoo! for six years, working as a visual designer on the front page there. I started my design career right out of college by starting a design studio focused on complete communication and branding programs for startups in Silicon Valley. So I got my learning on the job initially, which was an amazing experience. Not always advised, though. And while I was at Yahoo!, I went back to school and got my masters in design at CCA, when I realized I really wanted to formulate my design process and practice. And at CCA, I was actually introduced to a designer and teacher from IDEO, and that’s where I learned about UX methods and process and got very inspired by the IDEO method and what they were doing there. So I brought some of those back to Yahoo! but really found when I got to Google that there was a much more nurturing environment for design thinking and that kind of collaborative work.

Douglas: You know, it’s so important to have that nurturing environment and the support to do the work. I was recently reading a lot of material around an FBI negotiator who was there in the early days, and there wasn’t a lot of support, you know? So it was like kind of two jobs. One is learning the material and getting really good at the craft, but then also the diplomacy of trying to convince everyone around you that this new approach is much better.

Kai: Absolutely. And I’ll be honest. When I first got to Google nine years ago, the UX community there was quite small. So they were just starting to develop the design-sprint way of working then. And a lot of the hard work that was done at that time was kind of building that culture or building the buy in of better ways to work together. So it wasn’t necessarily a smooth and easy sailing process, but the environment at Google has always been receptive to experimentation and working on how to work better together.

Douglas: I think that’s pretty clear from the Project Aristotle and all the research that goes into psychological safety and work rules and all those things definitely seems to be a culture of, how do we have better workplace culture and collaboration?

Kai: Absolutely. And that’s actually one of the areas that I am focused on right now. In addition to, obviously, making better products, we always want to make more user-centered products, and that’s one of our principles here. But my focus is on helping to make a great space for cross-functional collaboration, to really support the UX role and the UX discipline, both inside and outside of Google.

Douglas: So, I’m really curious. As an insider, you probably have some really interesting insight into this, but one thing I’ve noticed is this paradox between data-driven decisions and then kind of more qualitative type of decision making. And UX tends to kind of sit at the boundary of that, right? There’s things that UX does that’s quantitative and things that they do that’s qualitative. And Jake Knapp always talks about design sprint being so good at quick-and-dirty data before we’re waiting around for perfect data. So I’m kind of curious what you found, having to use and leverage qualitative methods and approaches inside a company that’s been known publicly to be so data focused and data driven.

Kai: You know, that’s a really interesting question. Actually, there’s a great video from my series Design Is […] called Design is [Data], if anyone is interested in deep diving into a case study on how design uses data to drive great design. When I was first—well, my second year at Google, I joined the search ads team, which is an incredibly data-driven team. And so that video outlines a project that we worked on, where we looked at how to leverage data to develop a design hypothesis and then use the data to push that design forward, meaning inform our decisions about design.

And working on Search at Google was incredibly transformative for me as a designer because it is so data driven and because it is so large, the user base is so broad, you really had to rely on the data to inform your intuition as a designer. So the qualitative aspect is really important, and we balance that at Google, where we do a lot of user interviews, more foundational research, where we try to understand how people think about things. And then we do experiment-focused either usability tests or we run an experiment on a specific portion of the traffic, like we would do a search, to take that hypothesis and see if it’s working. So I always see it as like a balance of both and sometimes quite a delicate balance, especially if you are a UXer, which involves a really strong collaboration with your engineering team.

Douglas: Yeah. I remember reading this blog post years ago that I think came out of Google—yeah, it was Google Maps—and they were talking about drawing together. And so rather than the designers just sketching something, some brilliant solution, it’s like, well—and rather than just sitting down and talking—they’re like, “Well, sketch out. Can you sketch with me while we’re talking?” And they noticed this pattern that came out of a lot of their conversations, and they just took a step back and looked at all the drawings that their interviewees had sketched out while they were talking. They all had this dog-bone shape to them. So there was like this epiphany that, oh, wow, everyone has this kind of dog-bone kind of reality, where it’s like you do a lot of stuff around your house and a lot of stuff around your work. And then there’s this tiny little connection between the two. And then that epiphany that comes from some of the qualitative stuff could then be married with some data, even ML type stuff, that you feed in this insight into something that’s more data driven then you could get some really powerful things out of it.

Kai: Yeah. And that’s where the design sprint as a structured collaboration process is really helpful because you can bring the data into the UX conversation early, and you can bring UX and data into the product conversation, right, the product-strategy conversation. So that convergence in an efficient and fast is really where we see a lot of benefit.

And to the point you were making earlier about fast-and-dirty data, sometimes it’s also making the time to engage with it, because we’re always working against a deadline or moving very quickly. But if we don’t stop and analyze what we see and what we know together with different perspectives, we miss a lot of opportunities.

Douglas: Absolutely. And I think that not only taking the time to smell the roses, but like you say, if it’s this confluence of threads, because the product team’s thinking about a certain set of concerns, the data team has concerns, the engineers have concerns, the designers have concerns, and that’s the power of the cross-functional teams. But we don’t always, even though we’re working as cross-functional teams, we don’t always take the time to stop and then kind of look around and ask some of these questions.

Kai: Yeah. And that’s where that—I mean, I had a real aha moment in the sprint with one of our partners, where we had a data analyst in the sprint. And, you know, I asked this question, like, “Well, where is the drop off happening? You have this funnel, and people are coming in here. Can you go see what is the level of drop off?” And she pulled up the computer right then and there and said, “Wow, actually, we’ve got a 40 percent drop off at this point,” and that helped us to figure out what was actually going wrong with the design, and we could use that to help problem solve. So I was so excited by the data-analyst role after that. I was like, can we get a data analyst in our sprint?

Douglas: Isn’t it amazing when you get the teams together with different capabilities and then folks within different parts of the organization are going to ask totally different questions, but they might ask questions that they can’t answer, because you’ve got the whole kind of, let’s say, a gamut of capabilities brought together, amazing things can happen just in those moments.

And, you know, I’ve seen things within a design spirit that were orthogonal to the work we were doing. In fact, it just happened earlier this week. John’s facilitating the design sprint, and I was dipping in here and there just to see how things were going. One of the times I was listening in, I heard them say, “I should totally use this in that community outreach project we’re working on.” So it’s like they’re making these discoveries that on the surface feel fairly trivial, but I’m convinced that without the design sprint, they would not have made those discoveries and that project would have suffered, or it would have been mediocre and not as good as it could have been.

Kai: Absolutely, yeah.

Douglas: I notice it because I love this stuff, but I think a lot of times folks have these tangential benefits that they don’t even realize or definitely don’t necessarily associate to the design sprint because it wasn’t even the core product they were working on. There’s just these ancillary benefits.

Kai: Yeah. It’s interesting how we get so many ancillary benefits from design sprints, particularly also with relationship building, shared vocabulary building, shared knowledge. But now that we are not able to all actually be in a room together, at least at current, at present time, we don’t automatically get those benefits. We actually have to think very intentionally about what do we need to get out of a session and design for it, whereas we used to get these added benefits without even necessarily looking for them. We would come in with, well, we need a product outcome, and then benefit from all these side conversations and the increased team building and all that stuff.

But now if we need to build relationships, so if we have a new team or we have new people that we’re working with that we need to actually form stronger connections with, we have to intentionally design that because the virtual experience doesn’t create it in the way that just being together does.

Douglas: 100 percent. And, you know, we’ve talked a lot about this notion that ultimately all of these virtual tools are in their infancy. It’s sort of like the way e-commerce was in the early 2000s. We were still trying to figure it out, and what are the models that are going to stick? And Daniel talks a lot about, we’ve had thousands and thousands of years of learning how to meet in person, and we can lean on that knowledge and those customs. But virtually, we don’t necessarily have the tooling and the skills where we can just walk into a room and it just naturally happens, like you’re talking about. And so I agree it has to be highly designed in and really thoughtful.

And we talked a little bit about that in the preshow chat. It’s had an impact on the training programs. I guess I’m just curious what you found you now have to focus on to make sure that Sprint Masters is successful in this virtual world.

Kai: Well, we have to teach them to pay closer attention to that, whereas initially we would sort of take that for granted. You’re going to get this by bringing everyone in a room together. The difficult conversations are going to happen. You just have to nudge a little bit to make space for it, to ask the right questions. Now you have to teach not just how to be a facilitator but how to be a very mindful facilitator that’s aware of the science and psychology of the virtual space, which, as you said, is very nascent. So we are teaching something as we’re learning it. We’re modeling it and learning as we go, which is amazing. Also, I can’t say that we’ve gotten it right right now. We’re trying it out. But at the same time, we also need to teach people, Googlers, the methodology, so we have the foundational knowledge that they need to gain, which is, what is a design sprint? When do I use it? What do I use it for? How do I scope it properly? All the basics. And then, how do I be a good facilitator? And then, how do I be a good facilitator in a virtual space that I’m maybe not comfortable in and maybe I have to find new tools and techniques for how to build presence or even to consider, what do my facilitator values like and how did they play out here?

Douglas: Yeah. I love this notion of values-based facilitation. It can be really powerful because you can show up in a real, authentic way.

Kai: Yeah, absolutely. And everyone has to find their own way, and that takes time to build your practice and what you feel comfortable with, your authentic self.

Douglas: Absolutely. I think there’s a reason that the notion of having a practice is to label it with the word practice because it takes practice. You know, we can’t just read a book and expect to be a great facilitator. We have to try things on for size and see what doesn’t fit and actually go do the work. We’re going to have to be willing to stumble a few times, pick ourselves up, and learn, okay, that didn’t feel very authentic. Let me try something different.

Kai: Yeah, it’s amazing. And the different ways that people learn is a thing that I have to always remind myself of, because how I learn is not the same way as how other people learn. So we’re trying to make space for sort of a diversity of approaches.

Douglas: Yeah. I think as much as we can be multi-sensory as possible to where no matter where someone’s at at any particular time, they relate to what we’re trying to convey. I think that’s pretty huge. And the training, you talked about connection, being able to observe people and draw them in. Have you found any good tactics or are there things you’re recommending to your Sprint Masters around how to think about that in the virtual space or what the tools can allow us?

Kai: Yeah, I think there’s a lot of great resources out there, but one of the things that we, just at a basic level, are trying to teach our facilitators is this idea of creating. I mean, we talk about creating and holding space in the physical space, but creating and holding the virtual space and making not just room for voices, but giving each person the opportunity to speak and be present. And we have such limited time, so it can be really hard to balance that. And I think that’s really forefront on my mind right now because the getting the product outcomes and asking the strategic questions and pushing the thinking, we all feel relatively comfortable with that. But to do it in a way where you create and hold this space, where people feel safe to be creative, they feel connected, like you’re saying, it really involves a delicate balance of speaking, being present, being called on, and a combination of, I hate to use the word icebreaker, but it’s like activities that can be used to help us get to know each other better.

Douglas: Yeah, you know, it’s funny. Icebreaker. Some folks dislike the term. And my perspective is that those are typically things that people throw around without much intention. Being true to what you were saying earlier, the virtual space requires even more intention than the physical space. And when we put those things in, we should be careful about what we’re trying to accomplish and just be honest about it and communicate it so that it’s well known.

Kai: Yeah. It’s a great point about, I think especially in my early days of being a facilitator, I didn’t have as much clarity when to use something like an icebreaker or an energizer or a “pair people up and talk about this thing” kind of activity and building that sort of the intuition of knowing, what do you need to do during, at what point throughout the sprint? When do you need to give people quiet moments to reflect? When do you need to build energy and get them feeling inspired and feeling seen or heard? Those are really nuanced things, and that’s why I say I hate to use the word icebreaker because an icebreaker is something that you can just be like, “Oh, yeah, everyone stand up and organize yourself like a fruit.” Sort by size and color or flavor. And that’s fun, and it builds energy in a room, but it’s not the same in virtual. So you have to pick very carefully.

Douglas: Absolutely. Even turn taking can be challenging in the virtual space. It’s like in the physical space, you can say, “Okay, let’s just go in a circle.” In the virtual space, we have to think about new paradigms and what’s our cadence and what are our signals.

Kai: Yeah. And it can be awkward and take extra time. And I’m just thinking, “Normally, this would take five minutes.”

Douglas: It’s amazing because some things move a lot faster. You know, if we’re voting inside of MURAL or some other tool, it can go quite fast. And so we might have to intentionally slow it down so that it becomes more thoughtful.

Kai: Yeah, absolutely.

Douglas: You know, I think you guys share something in common with us in the sense that you tend to sometimes, or even often, have larger workshops, whereas a classic design sprint’s seven people. We’ve discussed in the past how we’ve both had workshops that are quite more numerous than that. And I’m curious what sorts of things that you’re doing in the virtual space to accommodate for larger audiences.

Kai: Yeah. It’s hard. I’ll just acknowledge that because we used to do 20-, 30-person sprints. And also, when it comes to training, I used to—I’d be like, “Sure, I could train 60 people. No problem,” by myself in a room. But in the virtual space, because of the breakouts and it, just, it’s different, so I have not run as many large sprints, though I did do a 45-person one two weeks ago. But what we do is we really rely heavily on breaking out into groups of five with facilitators within it. So it’s just the scaled model. I just lean on my facilitator’s more, and we’ll make sure that each team has a POC or a lead of some kind.

I think it’s possible—a lot of our Sprint Masters right now are just saying don’t do it, which I understand that. And I think we used to say don’t do virtual either, because it’s just too complicated. So I think eventually we will find ways. And I really appreciate some of the tips and tricks that I learned from your large-virtual-meetings workshop. Seeing how you manage a group of that size I think is really helpful because you do want to approach it really differently. You’re not going to do the around-the-room “show me that special item off your desk and tell me the history” when you’ve got 45 people. So you are approaching it from a different perspective, but it can be scaled.

Douglas: Absolutely. So coming back to the Sprint Master training, I’m curious. Ultimately, Google is supporting this initiative, so there’s clearly an identified business value. For the business folks that are listening, what’s the real purpose behind—like, why spend the time to train all these folks? What does that really open up for Google?

Kai: Well, it enables us to work more effectively cross functionally, and I think I’ve—I don’t know how many times I’ve said the word cross-functional already. You can keep count—but there is so much value we all know that comes from that. And as a larger company, it’s not a standard way of working. I think smaller startups will in general work more easily more cross functionally. But when you’re in a larger organization with multiple product areas, working across products and working across end product and UX, more marketing, it’s really critical to the success of the company. It’s critical to understanding the business so that everybody who is working on product teams has that viewpoint across the company to working across those silos.

As the company grows, we need more and more people to do that work, to help with that work, and that’s where the Google Spirit Masters come in. We train people across all the PAs, primarily in the UX discipline, but we do folks in some other roles as well, provide the training to them. And ultimately, they’re not just running design sprints for their product area. So if you’re a Sprint Master on Photos, you might raise your hand to go run a design sprint for Maps or for Cloud. And that also creates cross pollination of ideas and allows you to leverage solutions that have worked in one part of the company in other parts. So there’s a lot of value there.

It allows us to work, we say, to accelerate our collaboration, accelerate our innovation, because it helps us to work faster as well. And it’s not just faster, but it’s more effectively as a group because teamwork is always harder than individual work.

Douglas: I guess one thing that I always loved to visualize the positive impacts is this notion of kind of flipping around your discovery process, because so often integrations start to happen later in the cycle. Even if we’re doing continuous integration and all these good things, it’s like if we’re not doing a lot of this upfront discovery work together, then we discover those kind of break points later. And anyone in this, doing this work at all, knows that the later the discoveries happen, the more expensive they are to address. And sometimes it’s too late because it’s not going to make or break the project, but it would have created a little extra delight for the user, and it would have been cheap to do if we would’ve known about it early. But now it’s too late.

Kai: Yeah, absolutely. And users don’t care that the Photos team is different than the Maps team, you know? Users don’t care that Search is a whole other team. Those product lines are not valuable to them. They pop back and forth across multiple suite of products, and we want that to be seamless for them. We want it to be the best experience it can. And when they’re owned by different teams, that requires a lot of coordination.

Douglas: Yeah. It reminds me of Conway’s Law. I don’t know if you’ve run across this, but most people that have actually written software have seen it numerous times. It’s basically the statement that says that any piece of software is going to resemble the structure of the organization that wrote that software. And so what you’re describing is exactly the manifestation of Conway’s Law, right? You’ve got these teams that are working on different portions, and at one level you could say they’re working on different applications. But when, to your point, one of the user’s using it, they’re using the system, not an application. And so they’re kind of moving fluidly around that. And how cohesive it is is going to depend on how much you can bridge those boundaries of the organization structure.

Kai: Yeah. And that’s, in some ways, what the Sprint Masters are an antidote to at times, right?

Douglas: That’s amazing.

Kai: Yeah.

Douglas: So you talked a little bit in the preshow chat about this notion of resilient product teams and how the sprint can help with that process. So I’m curious. I think our listeners would enjoy hearing a little bit about that because I think it’s fascinating.

Kai: Yeah. Actually, when I think about the resilience on my side, it’s maybe less about the sprint process itself and more about how you build a self-sustaining community. And that resilience comes from operating as, like, a learning organization, which means taking the time to reflect on what’s working and what’s not working, and then focusing on improving the process as you go. You know, that’s very standard when you think about a learning organization.

But the example that we recently had was as we had to pivot to moving all of our sprints to virtual and helping 400 Sprint Masters build confidence in the virtual world, also not being experts ourselves, we really turned to this group to build a platform where we could learn from each other. And that’s where I see a lot of resilience happening, when you have a system in place for people to solve together the challenges or the things that they’re facing, which means it’s not me coming in and saying, “I know the answer here, and I’m going to teach all of you how to do this,” but people joining forces and saying, “I’m trying this thing over here. I’ve got an experiment running. I’m going to feed it back in.” And my team being that central hub that creates a space for that amplification and knowledge sharing so that we can rapidly adapt and build new ways of thinking. And that sort of comes from how you approach that evolution, how you approach a learning organization.

Douglas: Wow, yeah. I love this concept of the learning organization and this group that can facilitate and be a conduit for these learnings, to kind of come and be dispersed throughout the other teams and whatnot, I think that’s a fantastic concept. And we talked a little bit also about embracing the experimentation mindset. And I’ve been referring to this as the sprint mindset in the context of design sprints and how it’s so critical for folks to not only see the benefit. You know, we talked about the direct benefits of how it’s moving your project forward and getting people aligned. And then also the ancillary benefits of, like, these weird connections or these epiphanies you have that influence other projects. Then, if you can leave that workshop and have these mindsets permeate all the work you do, then it has this kind of like compounding effect throughout the organization.

Kai: Yeah, absolutely. And it’s something I’ve been thinking about a lot recently, and I think you know because I reached out to you for more resources, because I think this is something that it’s not as easy as it sounds. You’re like, “Yeah, just design and experiment and learn from it.” There’s a way in which oftentimes we will do things and we haven’t fully articulated what it is that we want to learn from it. And I think that is one of the benefits or it’s sort of the strength of the sprint process, where you go in and say, “What do we know, what do we don’t know, and what do we want to know?” And a lot of times, people will think that they’re experimenting because they’re being scrappy and moving fast, but they’re not learning as they go. And that’s, like, the fundamental benefit comes when you stop and reflect and learn and improve. So I think it’s incredibly valuable for any organization to look at how do we embrace experimentation and really figure out how to bake it into the way that you work.

Douglas: Yeah. I love that. We picked up this concept from the military, which is, you know, people always talk about lessons learned, but basically their stance on this is, like, people use the term lessons learned inappropriately. And most of the time, what they’re talking about are lessons identified. And we haven’t really learned the lesson until we’ve actually addressed it and done something about it. So we can identify it all the time, but if we don’t actually make the change, we haven’t actually done anything about it. We haven’t addressed it. We’re just located a problem.

Kai: Yeah. And that happens a lot. We’re like, “There’s a problem. There’s a problem.” You move on to the next thing, and then you’re like, okay. It’s human nature, too, you know?

Douglas: Absolutely. And, you know, I think that’s one of the beautiful things about facilitation and a lot of these methods is they’re designed with human nature in mind and how we can take advantage of the good parts and use them against some of the things that are not so great about human nature.

Kai: Yeah, absolutely. I spend a lot of time saying, you know, “I’m trying to combat design fixation here. Don’t spend too much time on that sketch. I don’t want you to get too attached to that idea.”

Douglas: Yeah. I ran into a term recently called street psychologist, which I thought was pretty incredible. And I think essentially to be a great facilitator is you have to kind of have to be a street psychologist.

Kai: Kind of like an armchair—

Douglas: Yeah, exactly.

Kai: —an armchair psychologist, but we’re doing it from the street. I like it.

Douglas: I don’t know just enough to understand what we’re dealing with.

Kai: Absolutely.

Douglas: So I’m curious. If you could change anything about most meetings, what would it be?

Kai: Well, that’s a very broad question because I have all kinds of different meetings. A lot of my meetings are not bad currently. But I would say I find really, really large, kind of all handsy-type meetings to be ones that I often wonder why we do them. So I think there is a way in which sometimes we have meetings that don’t really need to be in person or synchronous, as we say. And I think we could be more creative about how we deliver content asynchronously to people and that when we do bring people together, we design for people being together as opposed to a one-way delivery of content, because I often feel like now, even more than ever, a one-way delivery of content is consumed one way. So I would say really embracing when things are bi-directional or multidirectional and designing for that to make the most use of people’s time when they’re all actually being asked to be together at the same time.

Douglas: Absolutely. We often talk about, is a meeting purely informative, and if so, there’s probably better vehicles than bringing everyone together unless we really want to get people’s reactions. But if we’re truly wanting to get everyone’s reactions, like you say, it needs to be designed in a bi-directional way.

Kai: Absolutely.

Douglas: Excellent.

Well, Kai. It’s been a pleasure chatting with you today. And I’m sure people are curious how they might be able to follow your work or get in touch or just kind of see more about what you are working on. How can they find you?

Kai: You can always find me on Twitter, @kaihaley, and LinkedIn. And then, of course, the designsprintkit.withgoogle.com is a great site where I post case studies and content for folks, if anyone’s interested more in our design-sprint information.

Douglas: Excellent. Thank you so much, Kai.

Kai: Thank you so much for having me. It was such a pleasure.

Outro: Thanks for joining me for another episode of Control the Room. Don’t forget to subscribe to receive updates when new episodes are released. If you want more, head over to our blog, where I post weekly articles and resources about working b