Vivian Sarratt, Head of UX Operations at Google
At Google, you might have a team of 300 engineers, two dozen designers, 10 researchers and a handful of product and project managers. All unique humans with different opinions and ideas. All working together on enormously broad, complex problems. Clipping away at a speed of maximum velocity.
Given the scope, speed, team size and sheer messiness of human nature, this might sound like a recipe for disaster. But it’s actually intentional by design, a fine-tuned system based on a simple truth: cross-functional collaboration creates better products.
If Google can do it, why do so many of us struggle to work together across our teams? Vivian Sarratt, Head of UX Operations at Google, joins us to spill some of the secrets to success. Tune in to explore the complicated relationship of collaboration and strategies to make it work.
Carl Smith: Hey everybody and welcome back to the Bureau studios. You know, I got a lot of friends out there that are digital project managers. And one of the conversations I hear a lot is that career path is a challenge and I've always felt that if you've got those skills to be a project manager, you can really go anywhere. I think today's guest kind of shows that that's true. She is an Emmy Award winning producer in advertising across multiple mediums and today is the head of UX operations at Google, but she got her start as a project manager. It's Vivian Sarratt. How are you Vivian?
Vivian Sarratt: Hi, I'm doing great. Thanks for having me on the show.
Carl Smith: I'm so glad you're here. Do me a favor and share with everybody a little bit about how you got your start and then transitioned on?
Vivian Sarratt: How I got my start in project management in general was a little while ago. I was actually talked into the role by a friend who was a project manager. I didn't really know what project management of digital advertising meant, but turns out I kind of have a knack for it. I didn't go to school for it and it was a really good fit. Someone talked me into it and I found a role in New York City. I talked them into hiring me. It was a really pretty big stretch, I think, for all of us. But once I got the hang of how to make things move forward, I was kind of hooked from there.
Carl Smith: There you go. And so you move on. And I know at one point you jumped into managing seven different product design teams.
Vivian Sarratt: That's right.
Carl Smith: That's got to be crazy going from managing projects to managing those teams. What was that like?
Vivian Sarratt: Well, I think that at Google we don't really have project managers in the same way you see project managers at other companies. And the reason why is the scope and the velocity is just too broad. There's so much going on. In a project management capacity, you might have a team of 10 people, but in a program management capacity you might have a team of 300 engineers and maybe 20 designers and 10 researchers. So you kind of have to step back from a lot of the handholding that you do as a project manager. And as a program manager, you're really managing the program. So you're thinking about things really differently and you have to switch your mindset. You have to switch your team's mindset. You have to zoom out way more than I've ever expected to zoom out because it's a lot of work. It's a lot of stuff.
Carl Smith: I can only imagine when you get into that program manager perspective for the teams that are that large soft skills have to be a huge part of keeping things moving forward.
Vivian Sarratt: I think that you're right, soft skills are the hardest thing to articulate and grow and identify. What kind of soft skills were you thinking about in terms of big program management? What would you say?
Carl Smith: Part of it, I think, especially when you've got that many people. First of all, you have to have that safe environment where people feel they can speak their minds about what's going on. And so, especially if you look at the designer engineer, designer developer kind of dynamic, it's about being able to talk to people in a way that they can hear you, and I would imagine when you're moving really fast and trying to accomplish a lot those communication skills are just critical.
Vivian Sarratt: Yeah, I think that soft skills kind of encompass communication. I think you're right and that part's critical, but really feeling the vibe of the team is something that you can't really articulate, but you can feel it as a program manager, you can really sense it. And you really have to take stock and step back. I really like the fact that I'm able to be kind of that neutral bridge between design, research, engineering, and product because all of them kind of need advocates and all of them kind of need guard rails, if you will. And that helps them work better together, I think.
Carl Smith: You know, that happens to be the title of your talk for Design Leadership Days this September in Seattle, better together. And it's all about how cross functional teams... Hold on. And it's all about how cross functional collaboration creates better products. So is this part of what you do now? You're kind of helping the vibe of the team. Talk about that.
Vivian Sarratt: Yeah, it is all that I do now. I think that there's two sides to the brain when you deal with this stuff. There's the soft skills and then it's the hard skills. It's assessing how much work we have and when it could possibly be turned in and what a reasonable milestone could be. Because we're living in the software world, we're not living in advertising world where you launch a thing, get it out the door, never see it again. Software world is so different. You're launching a thing and then you're checking on how it's doing, you're monitoring that thing, you're talking to your users, you're looking at the results and you're improving on it, hopefully, based on data and based on research analysis. So you have to kind of have both sides. You're always juggling that warm and cozy vibe for the teams, is everyone happy, is everyone getting what they need? But also really trying to crack the whip, to be totally honest, and align everyone to big picture goals that hopefully everyone has agreed to in the beginning. So I talk about that a lot in my talk. I think it's kind of funny how you have to bring it back to basics. What are we all doing here and how are we getting it done?
Carl Smith: And with a team that size, I'm sure you've got some amount of turnover as well. So you've always got people in different phases of their relationship with the team. So you have to keep all of that in mind as well.
Vivian Sarratt: It's so funny that you mention the turnover, because 300 engineers, that could have... Half the team could turn over in two years, and that's a lot. And also, being at Google you're really encouraged to find other roles after you've mastered your domain. You're encouraged to stay at Google, they really want to keep and retain all this talent that they're growing. So how do you get all these people when it's not a revolving door but it's just a natural part of life. Right?
Carl Smith: Right.
Vivian Sarratt: I think what it comes down to is being hyper-organized with artifacts that lay out everything from the product requirements to what the design doc looks like for the engineering side to what the user is trying to do and understanding the user first and foremost and how that permeates everything that you're doing.
Vivian Sarratt: PGM's really manage all... PGM, sorry, program managers, we really manage all of that and I think that it's crucial to have that trail head because you're going to have people coming in and out and they need to know, a one stop shop, how do I get my information, what is the whole point of what we're doing? And the goal has to really be articulated and everyone should be able to say what the North Star vision is and it should be super clear. But how we do stuff is always different. Every team kind of works differently and has different priorities. So that North Star vision is crucial, I think, to articulate.
Carl Smith: This reminds me a lot. I was having a conversation with the owner of a product company and he told me, he goes, "I've finally realized my job is chief repeating officer. I just have to say the same stuff so many times that I think everybody's going to get mad at me. But then I realize 4 of the 10 people in the room weren't here last time when I said it."
Vivian Sarratt: Yeah. It's kind of funny how that works. Right? We do a lot of surveys at Google in terms of how's management doing and how's leadership doing? What we constantly hear is we want to have managers that are consistent, that are fair, that aren't playing favorites and that are clear. That's really all people want. So I think we all have to strive to it in how we manage programs in addition to how we manage teams of people.
Carl Smith: And that gets back to some of Daniel Pink's drive stuff with intrinsic and extrinsic in terms of how you're being treated. And to have that voice, which Google is well known for giving everyone that opportunity to share. And I'm curious when you have people coming on the team, I think about an onboarding process and I've heard everything from thrown into the fire to three months of having people come on. And if you've got a two year window as sort of a norm and you have people who may have already been in the organization but are coming onto a new team, what is that onboarding process like?
Vivian Sarratt: It really depends on the product. I think I've really been drawn to enterprise products. I feel like they're really complicated and fascinating, and it gives me a deeper challenge than just a traditional consumer product. But interestingly enough, Google to me feels like lots of startups within a large organization. So it really depends on what team you're on and how organized that team is and how the vibe of that team is and what leadership is bringing to the team. So, in terms of... I think I just wandered a little bit. What was the question?
Carl Smith: No, you're totally fine. I'm going to ask you a new question. It's kind of related. So, so tell me about some of the favorite products that you do work on when you talk about the enterprise products.
Vivian Sarratt: Well, I think that I have to give some shout outs to the the flagship Google product, which is ads. I was really fascinated. But coming from the advertising world and understanding some of ads, some of what drives advertising, I was fascinated to come to Google and see the inner workings and the inner machinations of AdWords. I think that AdWords generates... It's pretty well known. AdWords generates almost all of the internet's revenue and generates almost all of Google's revenue. So when I started at Google, this is a public knowledge thing, ads were making $3,000 every second.
Carl Smith: Whoa.
Vivian Sarratt: How unfathomable. And that was six years ago. So I don't know [crosstalk 00:10:21]-
Carl Smith: Oh, my goodness.
Vivian Sarratt: But if you think about ads generating 96% of the internet's revenue, that's a really interesting and slightly scary proposition. But I really love ads because there's so many... So I was working on seven different product areas within ads and my favorite one was, believe it or not, my favorite idea of this whole thing was when we took a really hard look at our platform because we were starting to ship our organizational dividing lines. We were starting to really think about things in terms... We weren't thinking about things in terms of the user. We were really thinking about things in terms of each organizational group at Google, and we had to tear apart that kind of approach. And we had to think user first and framework first. So we really took a hard look at the ad words kind of front end and we really revised how users go through it because it was a really fragmented process. People were designing this platform, this ads platform, without talking to other parts of the ads groups.
Vivian Sarratt: So, as an example, you have video ads, you have search ads, which are of course SEO, and then you have display ads which are banners or html banners of course. So those three groups weren't working together enough. They were rolling up to different organizations. And really, when you come down to the advertiser, let's just take what's in front of me, Purell, let's take Johnson & Johnson, they don't look at their advertising budget separately really, they're looking at a big chunk of money and then how do they spend it. So, the challenge is how do we treat our platform the way that our advertisers, our users, need to use it. So it was a really good experience to think about a major rehaul of ads, basically.
Carl Smith: And I remember when I was running, I had a digital shop, and I remember trying to dive into ads and using it and being blown away that we spent like $100 and end up getting an $8,000 project. And I was like, "More of this, more of this." But I was struggling, too, because I didn't really understand how to craft it or what to do.
Vivian Sarratt: It's complicated.
Carl Smith: And then, when you have this opportunity now and you've got the team working on video and banners and organic, or I'm sorry, on SEO, so how do you get them together working together? Did they have, in the inner workings, did they have different KPIs? Did they have different metrics they were measured against or was it easy to get them to start working together?
Vivian Sarratt: No, it is so hard to get engineering [inaudible 00:13:11] together. I think that's really the key of everything. It's like how do you get these disparate groups within each of these kind of verticals, right? Banner ads or video ads. So they have to work together and then they have to pop their heads up and say, "Hey, what does our entire experience look like across the whole platform? And how do users really traverse through the UI?" And I think what we had to do is we had to really think about the core and we had to dissect it to think about what are the most impactful places and use cases for our advertisers? And then we had to kind of prioritize because you can't take forever to build things. So we had to cut, we had to have a really clear cut line of what we can and can't do.
Vivian Sarratt: So what we ended up doing was just ruthlessly prioritizing at Google. We really just have a strict engineering P0, P1, P2 layout of what P0 is mission critical, P1 is we should do it, P2 is that's a nice to have. So getting everyone on the same page with these things and then really, bringing it back to project management, really scoping the work. How long is this thing going to take? And if the requirements changed, this thing is going to take a lot longer than we expect it. So how are we going to deal with that?
Vivian Sarratt: So there's a lot of wrestling and a lot of, I hate this term, sausage making, but the upfront planning, this is the core of what we do. So sometimes you can kind of use time crunches in your favor and you can say, "Well that means since we don't have..." A financial quarter is only 12 weeks, so we have to think about it in terms of what can we do in 12 weeks, we can't think about it in terms of what can we do in 52 weeks, you really have to break down the work into digestible chunks, I would say. And I always encourage my team to articulate what does that chunk look like, how long is it going to take, how many people are on it, and just be the project manager and the lead. And then the program management comes in when you think about the whole system and how the whole design of the system fits together.
Carl Smith: And the idea of the ruthless prioritization, especially when I've been on teams where one person thought something was mission critical and the other person was like, "Eh, nice to have," and they never communicated that, so it was no wonder they were always in disagreement because they had totally different understandings of what the priorities were.
Vivian Sarratt: That's exactly right.
Carl Smith: Yeah. So bringing them together and getting them on that page. And then when the program manager comes in, once you've got that kind of organized in your world, they're responsible for the holistic UX and the feel of it?
Vivian Sarratt: Well, I think that the program manager is really tasked with developing an operational plan to get to that vision. So, that North Star vision is really defined by those, that product manager lead, that engineering lead and that design lead. And I partner program managers with design managers. And together I feel like design manager brings the vision, program manager brings the operations to support that vision. And a lot of the time that means everything from how are we kicking off the work, how are we reviewing the work, how are we presenting the work, how are we making sure that this matches what databases are doing over here and needs to match what storage is doing over here. But that also needs to match what artificial intelligence is doing, because all of those things live on the same platform. So what are the common patterns and components?
Vivian Sarratt: Because we're designing a system, we're not special snowflakes. We're in the inner machinations of Google so we really need to make sure all these things play together. And that's really where that North Star vision comes in. So it's a lot of shared goals between different vertical teams. The shared goals are great because then you have the key results that are different. So at Google we have a system called OKRs, and a lot of people use this. It's really standing for objective and key result. So this is how we plan all of our quarters, getting all the leads on the same page in terms of what is the objective. And then the key results, they can be different. Each person has a different tactic that they're working on. They have a different heads down thing. But making sure you're popping your head up and saying, "This aligns to the objective that we all agreed to."
Carl Smith: OKRs, it's great to know that that's something that you're using at Google because I hear so many people talk about them. So many people using them. I'm sure everybody has their own flavor, but that also has to be just an amazing way to keep the velocity up for a team that size in these 12 week increments.
Vivian Sarratt: Yeah, I think it's a really interesting point because the velocity is sometimes different for different teams. So you UX might work on an individual who's a UX or it might work on four different projects in a quarter. But an engineer might work on one. So the scope of each person is very different and there's more context switching for design compared to engineering. Additionally, our resources aren't fungible. We can't just grab a designer and put her on something else because she's got all this context and knowledge and background that she's bringing, especially with enterprise users, they're so complex. It's really interesting.
Vivian Sarratt: But I think it's really important to call out that, I'm a little biased because I'm program manager for designers, but I think that designers have it the toughest, to be totally honest. We have context, which all the time we have to be really helping the PMs lead them toward understanding the user better so that they can write better requirements for the products. And we have to be able to justify our design decisions with data. So, we're getting it from all sides. I think the designers having the roughest. To be honest, I have the coolest job.
Carl Smith: The coolest job is definitely part of it, but also a job that has just now come in over the past few years come in to a position of respect as opposed to a position of almost nice to have.
Vivian Sarratt: Exactly, exactly. Google really emphasizes user first thinking, and who better knows the users than researchers? We try not to let PMs do the research because researchers are really trained in methods that aren't going to bias the user. So it's better to partner PMs with researchers and train them and teach them how to best get info from our users.
Vivian Sarratt: There's really good ways to ask questions and I'm not good at them because I always ask leading questions. But researchers are so solid at getting the data that's not biased and they have all these interesting and fascinating methods and we also have to find the right users for our studies. That's really complicated too. We have products on cloud that don't have more than 3,000 users and that's not like search for Google. That's quite a different user base, or like android for Google. We can have tons of studies with android users, but try finding a networking user in cloud, it's very difficult with a certain amount of knowledge, so lots of challenges from the design side.
Carl Smith: I can only imagine. And one of the things that feels like an underlying theme and everything that you're talking about is personalities and relationships, so I'm confident this is going to be a big part of what your focus is, and also what you're going to share in Seattle.
Vivian Sarratt: I think that I have to, I mean you can't... My philosophy is if people don't like each other, they're not going to make great products. You have to at least get along with your team in order to hear all the divergent opinions and ideas because that's really what makes the product better. If we all thought exactly the same, our products would be terrible. So more data, more opinions. I think it's really great for collaboration.
Carl Smith: It's so funny. That reminds me, there used to be a show, it was like inside the actor's studio, but it was for CEOs. And it was at Stanford and they were interviewing Jack Welsh and he got a question from the audience and the students said, "What's one piece of advice you'd give for somebody who's starting their own company?" And he said, "Don't hire people you don't like. You're going to have to work with them." And it's so funny.
Carl Smith: Now, obviously, you can get into a real situation of avoiding diversity and doing all kinds of things if you only hired people that you like. But getting people you can get along with and work with is totally different. And that's kind of what I picked up on from what you were just saying.
Vivian Sarratt: Yeah, I mean, I think that that's actually one of the rubrics that we use to interview at Google. The rubrics are how well does this person know their role, that sort of thing, what are their skills? But also are they a Googley person? So, not necessarily does this person look like me or act like me or have the same background? But what kind of creative problem solving do they have? Are they proactive, are they positive, are they kind of coming in and poo-pooing everything? Or are they going to be okay with the status quo or are they going to be like, "How can we change this and make it better?" So that's a big part of what we look for. And that's also what I look for on my team. I've hired some people that have completely different experiences as me and I really liked that. But also there are people that if we got stuck in an airport for a delayed flight, they'd be fine to talk to.
Carl Smith: Well, there you go. I think that's a great balance. Well, Vivian, I'm so excited to have you joining us in September and I can't wait to meet you. And thank you so much for spending time with us today.
Vivian Sarratt: Absolutely, Carl, it was super fun. Thank you again, and I can't wait to see you in September too.
Carl Smith: Sounds good. Everybody listening, thank you so much, and we'll talk to you next week. All the best.
Show Notes
Thank you to our amazing partners for making The Bureau Briefing possible!
We have a new digital services salary report that’s coming out, thanks in part to our friends at VOGSY. Be sure to give them a look, and connect with them to see how their professional services automation software can help make your business smarter.
Mailchimp does SO much for the community, and they have so many great tools and resources to help your team and organization. If you haven’t already, sign up for their Partner Program to access exclusive benefits.
