UX STRAT Interview: Dr. Susan Petrick, Google

Paul: Hi Susan, thanks for taking time out to talk with me today. Can you start by telling us a little about yourself, your career, and what led you to your current role at Google?

Susan: Sure. I studied math and psychology initially as an undergraduate at Yale, and was more interested in the cognitive psychology part than the math part, thinking about a career ahead. So, I went right into a graduate program in cognitive psychology at MIT and studied processes that went on during fluent adult reading.

Bell Labs, at that point, was hiring lots of cognitive psychologists to do what was then usability testing, usability work. Of course, much more physical device-related, ergonomics, in addition to some amount that was more like programming and digital stuff. So, I worked on small business telephone systems at that part, getting everything from the handset to be comfortable, the dials to work, to getting people to be able to program functionality onto the buttons. It was a great experience in that it was a success. The first product that I got to work on is still out there--this was in the early ‘80s--and it’s still out there, I still see them when I travel around. I get to talk to people about their experience continuing to use it. Even to this day, some of them are still working. So, it sold me on the whole thing about being able to have lots of impact on product design, and lots of impact on lots of people that used it, because of such a good initial experience at Bell Labs.

I went on and did consulting work for some number of years, and again the work broadened out into web-based work, to voice recognition-based work, lots of different areas. And then when I went back to in-company work rather than consulting work, I worked at Compaq and then what became HP. I worked for Intuit immediately before Google, where I am now. And while at Intuit, I worked on the usability and the design decisions for online payroll systems, for empowering small businesses to be able to not have to hire expensive accountants to handle their payroll work, but to be able to do it themselves.

I then came to Google, worked first on the advertising systems used by online advertising systems, again used to grow small and medium-sized businesses, as well as used by agencies for high-spend clients. In January, I moved over to start managing the user experience research team that works on Google Travel-related products and tools, and that’s a team where a little bit more than half of them are based in the Zürich office and the rest in Mountain View, California, which is where I sit.

Paul: Ah, that’s where you are right now, is Mountain View?

Susan: That’s where I am right now. The team uses a lot of different methods, so a lot of what I do is coaching researchers and working with designers to make sure that we’re doing the right sorts of research to be able to impact the directions of product strategy. So, understanding underlying user needs that might or might not be being met well, understanding the current experiences that we’re creating. There’s a design lead, Lucia, who’s in the Zürich office, who I work with on a daily basis, making decisions for our travel-related products; in addition to design leads here in Mountain View.

My team sits with designers, we work with designers; we sit with product managers; we work with both the early phases before there’s any concrete design work being done, and then we continue the work throughout a product’s life-cycle, with the actual design being instantiated and incorporated, all the way through until things are being evaluated once they’re out with real users, being tested and then being used for real.

My group includes both qualitative researchers and quantitative researchers, so we do data science and logs analysis sorts of work, and in addition we do what’s more typically qualitative work, using ethnography, using in-lab or remote unmoderated testing. Lots of different methods, which makes it a fun job to get to advise on all of those and coordinate work for all of those, ensuring that whatever people are doing in my team, that it maps up to the most important goals for Google, the most important goals within the product area, and that work we do is actually having impact as opposed to not having impact.

That’s the fun part of my job, is to do that coordination with all the different cross-functional roles at Google now within Travel. I guess just to make sure that the user is at the forefront of everybody’s mind as decisions are being made. There are a lot of complex business decisions that need to get balanced. I enjoy working at Google especially because the focus is on the user, respecting the user, organizing the world’s information to make it universally accessible and useful for everyone. It really is something that Google not only says but acts on. So, it’s a great place to be doing the kind of work that I’m doing.

Paul: You mentioned ethnographic studies. Do you get to travel as you do the ethnographic studies?

Susan: Especially now that I’m within Travel, I do get to. I arranged for some research that was with travelers in India last month. So, I didn’t end up going myself, but I got to do all the planning and arrange for an in-market researcher to be with the team as we start learning about lots of the interesting challenges that travelers in India face, having to do with rail travel, having to do with flights, having to do with all the different travel methods that are prominently competing in people’s minds when they need to do traveling, either for leisure or business travel.

Paul: I was thinking more like cruise ships, Club Med, things like that.

Susan: It does include all of that. We have some work going on right now on packages. So, packages are a very interesting way to buy travel, and as you’re talking about some of the things you just raised, those would count as packages, where your lodging and your travel to get there, and some amount of the entertainment during the days while you’re there all could be included as one thing.

It’s something that in the US I never thought much about personally, as a traveler. It’s not your first thought. When you ask people who are traveling, often it’s like, “Okay, how much will it cost me to get this flight? And how do I get all of the information to ensure that I’m making, cost-wise, good choices, and get myself where I need to be when I need to be there?” Often times the lodging is a soon thereafter but secondary thought. And then finally what you’re going to do while there can be a distant thing.

There are lots of different kinds of trip planning, different funnels for making those decisions based on all the circumstances and the context of the travel. But in Europe, very different in terms of the attitude toward packages. In the US, cruises would be thought of as that kind of combined travel purchase. Whereas in other countries, like Germany, or other countries in Europe, it’s a first thought, a first way to start doing research for a possible trip. So in my group, what’s really cool, having people located both in the Zürich office and in Mountain View, we have easy access to several different language-speaking populations. And when I hire new researchers for my team, having two, three, four fluent languages is a real plus, being able to work with users in their actual language. At first we tried things like, “Well, we want to know what people in Germany think about what they’re currently doing in terms of purchasing these packages.” So, we thought, “Well, it’s a person in the California office that’s most interested in doing this work. We’ll just have her talk to people that are bilingual, that speak both German and English but that have lived in Germany their whole life.”

Well, that was a good way to do it, except that those may or may not be typical of travelers in general in Germany. And it turned out, as you’d expect, that they’re people who know a little bit more about Google, possibly have a little bit more trust in the results that Google would be providing them, and are just more likely to be familiar with Google as a US-based company, as a place to turn for information. So, having people actually located in Zürich, who are native German speakers to do the work with German-speaking country travelers is a much better way to do it.

Paul: That makes sense. I think of Google typically as an engineering company turning out these fantastic products, a lot of them free, search-related, etc. But what you’re describing pulls user experience and user research to the formative and even generative side of things. So, that’s a little bit of a surprise. Has that been recent, or has that been going on a long time?

Susan: It’s something that, well certainly the industry as a whole, in the time that I’ve been doing user experience research or what was usability testing, has migrated a lot in that direction. So, it started out that we were seen as testing things that engineering built, to kind of bless them before they went out the door. Then it got to be more of an iterative cycle, where they actually expected we would have something worthwhile to say and they might want to make changes. But this moving it to more upfront part of the process, hiring people with ethnography skills and knowledge, is certainly somewhat more of a recent trend in the industry. Back in the ‘80s when I was first doing applied work, it was cognitive psychologists who studied the cognitive processes that went on in a tangible, physical process, and now the research we do is much more broad, and earlier in the development process.

One nice thing about Google and the structure here is that my team is part of the engineering organization. We’re not a separate, far away branch. We’re not a marketing-related branch, though we do do some marketing-related work using conjoint analysis or MaxDiff. We might use marketing methods, but because we, organizationally, are close to and are physically sitting close to our product managers who are doing the early stage work, and engineering leaders who are making decisions early on about funding or not funding things, I get to be there at the table, I get to be contributing earlier in foundational sorts of work that the group has built up over the last few years.

Things change; there are a couple of really interesting things that are happening, which is why my group stays gainfully employed and useful to the process. So, as different technology trends emerge, people change what devices they’re using and when they’re using them, to do all sorts of tasks, including travel planning, or ads if we’re talking about the business population. Just the demographics of who’s traveling throughout the world have been changing in really interesting ways and are predicted to continue changing in the next few years. So, I’m going to be talking about some of that in my talk, and that keeps us current and needed for evaluating new efforts and for helping prioritize what Google actually builds when, and how it gets built. Similarly, changing traveler expectations based on both travel planning and other experiences that are happening makes a big difference in what we build and how we have to build it. So, a feeling of “I like to work with systems that know me well and that are tailored to what I need” is something that’s happening more and more throughout all the services that people touch during their lives, and it’s a big demand, especially if it’s something like travel, where it’s a relatively high consideration purchase. For a lot of people, it’s the most expensive thing they’ll actually purchase in the course of a year. And often the planning process is still tricky enough for people that there are a lot of touchpoints throughout that that they could use help so as not to end up frustrated or discouraged about being able to get their trip together at all.

Paul: I would think that the millennial generation and those following them would be quite open to an assortment of digital assistance tools to plan travel, or even to use as they are traveling. Without giving away any secrets, are you thinking in terms of a product that is going to be Google’s travel solution, or are you thinking more broadly of an ecosystem of products and interactions?

Susan: Right now it’s certainly been more of an ecosystem model. If you look out there to what Google does, starting from search results, Google Flights is something that has kind of a brand identity and a lot of people go to directly. But just being able to coordinate and plan to make sure that people get easy access to whichever kind of information they’re needing at the point in the planning journey or process that they currently find themselves is very much what my group is working on. Getting that knowledge of travel planning state to inform what kinds of information we’re serving up to people.

Paul: My 16-year-old son seems to have really high expectations of technology, and I can imagine him needing or being interested in, as he’s traveling, lots of fluid resources. I can imagine Google search would be a huge one of those. But I think he would be really open to giving up a lot of information just for the convenience of being able to say, “Hey this really fits you. You probably want to rather do this than do that.” Whereas I would not be as open to that. I would be more likely to discover as I go, doing it my old-fashioned way.

Susan: Yeah, it’s a difficult balance when we’re trying to make things frictionless and easy for people based on all the things we know about where they are in planning. But to make it transparent to people what sources of information we’re using, why we’re showing certain information or not showing other information. Just little tactical questions like if people want to do things on a map, to see certain kinds of information on a map, how information dense do you make that map? Do you show them every single point that qualifies for what they’re looking for? Do you show them only a small subset? What will work best for people? How much do you require people to tell you information about where they are and what they want to see, vs. how much do you try to automatically figure out from other signals that you have about the actions they’ve been doing as part of their research. That makes it super interesting and super challenging, too. But the important thing, again, is, from a user’s point of view, that things feel safe and responsible and understandable, in addition to easy and facilitated for their research and decision-making.

Paul: It seems like the travel industry in particular has a lot of complex layers. Because it’s not only the cultural differences in Asia and Europe and the USA--we have that across a lot of different verticals. But to then have just all those legal entities involved across all those, as well. So, it’s not only cultural, but also different countries are treating digital products and services differently. It seems like you have a rather complex ecosystem that you’re dealing with.

Susan: It’s really important to help people understand, again, what information you’re about to use, or you’re using, and in what way have they provided that information. Did they intend to? Do they want to? And give them easy ways to say, “No, I don’t want that factored in,” or, “I do want that factored in.” So, that’s a lot of what we work with or that my group specifically is interested in, is issues having to do with what engenders trust or what could cause lack of trust, and that we wouldn’t want to go there.

My team is in an interesting position because we work a lot with researchers who focus on other kinds of search and other kinds of search results. Also, people focus on other kinds of maps. So, there are certainly plenty of things that people would look for that wouldn’t really be travel planning, associated with both of those. We want to make sure that all the different experiences are good--travel and others. Same thing with assistants. More and more of the travel planning people are doing with natural language queries, either via search or via speech, and so we’re trying to make sure that, specifically for travel research, or even for booking, that people can do that if they choose to with an assistant-like experience. So, finding out how people do those kinds of conversations back and forth as they’re learning about different travel destinations has been another interesting part of the work my group has gotten to do recently.

Paul: It seems like a number of new technologies are starting to emerge: artificial intelligence, virtual reality, augmented reality, blockchain, machine-learning. It seems like some of these are going to impact your work possibly quite a bit as they come in. In your crystal ball, which of these do you think might be more significant for the travel industry, or at least your part of it?

Susan: Yeah, you certainly hear exciting things about augmented reality, virtual reality as a good way to learn about destinations. I’ve seen relatively less of that happening already, but it’s certainly something that we’re interested in when it does provide a better experience. But immediately the machine-learning and AI-related work is crucial to give people. If you’re trying to personalize an experience, you want to make sure that you understand, as best you can, traveler intent. That’s the whole thing.

If you think about something like giving people weather information about a particular location to which they’re planning a trip, it’s just hugely important to know everything you can know about things like time course prior to the trip. At some point, when people look for weather, if it’s far enough in advance, they’re trying to figure out, “Is this place going to be rainy when I go there? Is this place going to be hot and steamy when I go there?” So, the right kinds of information to give at that point, if you know when the trip is, if they’ve done any early date-based research at all, the right thing to give is usually generic information. What’s it usually like then, and what kinds of things am I likely to need to know. As it starts getting closer, though, to the time of travel, it’s like, “Gosh, am I going to need to have my raincoat accessible when I get off that plane, or can it be packed away in my checked bag?”

So, if there are things that people want to do while they’re at a location that are really weather- sensitive, they are often trying to get their day’s itinerary together based on “Is it going to rain this afternoon or is it going to rain this morning?” And so if somebody asks, “Should I go to the gardens in the morning and not the afternoon?” If we were to give them generic information about, “Well, the climate this time of year in this place is usually like this…” That’s a horrible user experience, and we want to make sure to never do that. And that’s a very trivial example, but it plays out through all the different kinds of information and the different timeframes that people would be needing it.

Paul: Well, I need to know all that stuff for June in Amsterdam.

Susan: [laughs]

Paul: Is there anything else you want to tell our audience before we sign off?

Susan: No. I think that your questions were great. It was fun talking to you, I enjoyed it. I’m looking forward to attending myself, certainly. I haven’t been to Amsterdam before. I’ll have to start my own travel planning.