[&] Design Research Tokyo Season 1 Episode 5 - Dara Gruber / Google
Design Research Tokyo Season 1 Episode 5
Dara Gruber / Google
Building Empathy Through Qualitative UX Research
歴代正社員 Google Japan のUXリサーチャー３名
This is Dora and
Alicia two gentlemen.
Thank you. Like Johnny said, my name is Sarah. I am a UX are here at Google. And it is my pleasure to welcome you guys here today.
So this this talk about giving is on building ethically through qualitative UX research. And I really like to do this talk, because there are many other types of research. And I do believe that you need to advocate for both qualitative and quantitative research in order to form a new picture. So this is our
computers at once.
But first, I'd like to give you guys a brief history of us are in Japan. And and the reason why we do this is because you know it, we realize that UX for Google Japan is not super
know that there are a bunch of us here. If you have a grandma, any Google, you actually do this. We're recruiting for understand or raise your hand.
I see over there. Here we go. Yeah. So we we exist.
But in terms of UX research, we've actually you would have is hosted three. So Judy, Judy shades actually still lives in this area. I was the founder of us at Google and, and then Adrienne came in and helped us a lot of products. And then I turned me, the full term researcher here as a beginning of 2018.
And, but as as most of you know, search is a fairly social thing. And so it's very hard to do it alone, especially when you don't speak Japanese in Japan. And so, we, over the years hosted a bunch of Google errs, who devoted their 20% of time to help him with
UX research projects you need being one of them.
And also some who have who have been in the lobby champion read today and like I said, we're better
are really impactful quality of research is what's going to make the difference in a lot of building for real user needs.
So I believe that there are three
main tenets to impactful qualitative research. So this this research should be radical, transparency, and fun, because you know, if you want to find that it's kind of boring, or research should be designed with always with the replicability in mind, right? So you want to strengthen your resolve elements across across your projects and products and cheese
But then you also want to be transparent, right? Where the research process which is unfamiliar to action, and then it should be understandable. So that you can build us with your stakeholders with your design partners, your product or service engineer partners, who may not actually understand that process and the ins and outs of how good research is done.
And then lastly, it should be fun, fun and social, right? research is a very, very social profession and you talk to users and we like to work in groups so that we can you know, constantly make our own
methods and ways of working and workshops better.
So generally speaking, there's this ideal research process right where for, for a lot of products out there, you you want to start with yeah
Bring up all the information and research right? And then we go through an expert review media how that product works. You'll hear heuristic evaluations about me, I live there's a sample in the news
and then you really get into Okay, how am I going to test this afternoon found a problem right and we we kind of have it needed some sort of
a product idea or solution
or hypothesis, you you create this replicable research design right that can be expanded
through our researchers and other in other products in other areas. Then you go to your testing and synthesis and, and then you tell everyone about what you found. And then you drive solutions. And within there, you can kind of see where these three main tenants that I believe the qualitative research at all
We're replicable is is more of the testing transparent is is with the analysis through kind of laying out your findings. And then fun is overly bunch of beautiful thing.
So I will go through each of these three tenants, and a little bit about them.
So wrapped ability applies across populations, devices, time and products.
And so a reputable study design
can really be thought about in a few different ways. So, so you can reuse and adapt other people's study designs and this is highly encouraged in this field. Because you know, what scales and methods are already out there. You know why they would over not work for your particular study, and they're already usually pretty validated. But the problem with this sometimes is that individual product research research needs a little bit
Custom customized ability from that original valid method. And so you really do need to understand what about that research design needs to be adapted without compromising the integrity of that method?
And then you kind of agents to systematized What have you do. So you figure out how to best scale
your method to be useful to others. You can gather feedback for researchers, and stakeholders. And then keep detailed immaculate records of everything you do so that other people can just read and go into a self service type type model.
And then you convince other people to use it right being convincing is a huge part of being a researcher
by explaining the benefits of having comparable data science.
And then last but not least, for replicable study design, you kind of use I use the term combined, and I use this in two different ways.
The first of which is to show the power of data by running cross analyses and comparing data sets. Right? So there's immense power in the combination of relatable data sets, but then also combine as as kind of teamwork, right? So you want to team up with other researchers who are using the method or similar one to discuss issues or benefit. And maybe how about how to adapt it to two different markets, right, that comes up pretty quickly.
So here's an example
of a reputable site.
sometimes you need more specificity that then already previously validated, right
as I mentioned earlier, so
user satisfaction. Interest is kind of one of those standard methodologies that people do use them to measure satisfaction and trusted products.
And so, really kind of the exercise here is defined similar products were or trust, which is a really big idea, right? Trust is a really big factor can be broken down into similar similar smaller factors. And components, such as examples for just would be technical competence, personal identity, information, privacy or help support. And if you can break down really, really big factors into small pieces and figure out what products have overlapping small pieces. That is a little bit easier to find the right methodology to reuse and then how to adapt it specifically to the product.
Okay, so the next time we have here
so that the main kind of takeaway here is that it's easy to hard truths. And of course, the beta. So obviously build a foundation of trust the stakeholders, and also kind of build your credibility as a researcher.
So here's some, here's kind of a breakdown of how I think about transparency.
So first is awareness. You want to make sure your stakeholders are aware of the research and design process, not after back back very beginning, you want to tell them exactly what this ideal process will entail. And for you start working with them.
You want to show them where research and it's into the overall product development cycle, at what points and why. And then include them at various points throughout that cycle, so that they're kept in the loop about what's happening.
And then you kind of want to manage expectations. So properly managing expectations for what research will and will not provide, or sometimes research people, people think that research will just provide you all the answers. But oftentimes, that's not the case. And I'd say almost every time when properties
so really being honest about what certain research will offer, in terms of how they will make decisions, strategic decisions to the border.
And then good research takes time. I think this is probably
the least understood among a lot of companies that good and research takes time. Right? And you see this in academia where, where people spend 510 years doing a really, really good quality research study. Right. And we're here often expected to do on this week.
Right. So you can kind of imagine what that quality trade off is, in terms of good research design. But you really just need to convey and be honest with the stakeholders on, on what you can accomplish and what quality that will be in the time that you have right work advocated for more time so that the quality is even better.
And then the last thing is communication. So, communicate early and often. and communicate your research and analysis plans before you start actually conducting research. be thorough and transparent by linking everything to your plans to your survey questions, sometimes raw data in your final reports, right? Because if you if you are transparent and if you really believe you're doing the research and you're transparent with with your findings with the data, they know you're not hiding anything and it actually build a really good rapport between researcher and the product team.
And then always connect the dots and show how you get from, from data to your recommendations, right? Because that's really what what researchers own as a skill set and trying to explain that as best as you can. To your stakeholders is a really great example of transparency. And so here I have kind of
Just involvement as it relates to transparency. And so in general, I'll collect research questions from many stakeholders at the very beginning, when someone comes to me and says, Hey, we're generally interested in this topic, right? So like, so user satisfaction, and I'll go to them and I'll say, Okay, everyone working on us on on these products, and have questions on user satisfaction, right, all of your questions down and give them to me, okay. I was like, seven and then I'm
Create a study plan to provide a rough timeline. Right and, and kind of work out how to tackle the most important questions. And then I'll tell them, right, I'll have, I'll have my stakeholders with you what this looks like and see if it matches the idea of, of when they would like to launch a product or feature.
And then as I'm planning them users, right, I like to always ask me folders if they want to watch sessions or take notes, or be involved in any way because really, any interaction that your your counterparts have with real users is kind of kind of game changing. Right? A lot of times they don't get to talk directly to people who use their products. They just talked to the researcher who talks to the people who uses the products and so that that first degree connection can really actually change the minds of some stakeholders. And then after you run research, right, involve them in your synthesis. Right. How's this
Code your user videos if you don't want to do them all.
And really kind of
captured their interpretation of what happened. And then again, I said link to your plans and your notes and your data because it really promotes and fosters the system of trust a team. And then review the report was key stakeholders before presenting it to other people. Right. So like, this is some really small thing, but it's also really important because you don't want any crazy surprises in a meeting room full of full of managers, right. One of the people you work most closely with, have it refunded refrigerators.
Favorite cars. Okay, fun.
Right? Oh, it's not loading. There we go. real picture of the most boring room ever. Right so you can turn the most boring thing into something funny.
Room amaze from full pipes right can you can be turned into something kind of crazy and fun. And I believe that is true for a lot of research that we do.
So fun should be shared. I mean,
I'm sure it's possible to have fun by yourself and I much rather have fun with other people.
So I like to think about, share fun in the workplace in these ways where
you can you can do icebreakers with your team, and ice breakers, I usually turn to like get to know other people's names and whatnot but but within a group setting.
improv games are actually a really great way to bond with team members to introduce play into the workplace, and also to like to lower inhibitions nice and when you do a creative exercise or or brainstorm a lot of times people are kind of it takes a while to get out of their their bubble or Amazon
little comfort zone. But if you played in Prague game, a lot of times it actually kind of makes them open up a little bit and, and studies actually show that in five exercises lead to 36% more ideas when it's then follow back to me. It's like, really simple games. And I'll show you guys some examples, but I'm actually going to call out some people and demonstrate an improv activity where you guys, right? I'm not gonna explain to you what it is and it's going to be really loud. It's really weird, but it's just an example of something that you can do with a small group of people. It's kind of crazy kind of childish, but also fun. So if you don't set up a call option back there, will have Courtney. Alice there.
me Get over here. YouTube
I get up you get YouTube. Perfect.
Alright, so please do get in a circle.
And you're just gonna have to learn as we go.
They know what's happening I use this month multiple times during workshop. Okay. So
so so so this game is is called Samurai and it is a kind of a communication game if anything else. And generally what happens is you you need to you need to be very loud, very intentional with your emotions. And then really you kind of just have fun and you may be competitive and so there's a rhythm and whoever breaks the rhythm is out.
All right, who wants to start? I know you guys know this Irie. You want sorry.
... Wow wow game
okay so I should just say Nick thank you guys yeah
So that's an example of something that works in all languages, right? We're making a certain noises and pointing at each other.
And it makes people laugh, it makes people open up, right? And then eventually, once you actually get to doing real collaborative work together, it's just a lot more fun in general.
So so the next section is on engagement.
And the way I like to explain engagement is like, think of things that people don't do on a normal basis, right? Like you, you want to break them out of the monotony of their everyday kind of cycle by peaking their interest, doing something unique, makes them laugh or smile, right? It just, it makes them remember the activity base and remember the research and it makes them remember the concepts and topics. And then the last thing is in the recent hasn't been passion. If you are not
excited about it doesn't matter what you're doing, it can be really boring. But if you the moderator hosts are not excited about it, no one else can either right?
And then empty. So a really good way to build them as the, as I mentioned earlier, is to really connect your stakeholders directly to your users. We here at Google run a lot of events that are aimed at Jimmy Mack will be able to train our engineers and our product people to conduct short one on one interviews with participants outside of Google to see how they use their products.
But within your team, and a way to build empathy within your team is actually to just enter for a really dramatic, emotional aha moments, right? Let them feel things that they can't really put into words. But then sometimes it's really hard so give them a little bit of time afterwards to realize that they
They had a new family.
Great. So here are some examples of activities that
kind of encouraged this idea of fun. So yeah warm up activities which we also semi successfully demonstrated for you where you can encourage movement and enable creativity 13 so you can play where, where your each person says a word and you go around and try to form a sentence right or collaborative creative icebreaker exercises. If you Google improv games, homeless and instructions on your company's you can do in a short amount of time.
idea generation sessions, which you know are people people run brainstorm sessions and ideation sessions and in various ways, but the general gist of those are that no idea is a bad idea. more is better. generally have fun
synthesis workshops are way to kind of engage your stakeholders again throughout the process. By having them do affinity diagramming with you. Right? have them go go through the motions of clustering and organizing raw data. It helps you, it also helps them to understand what's happening with
also, you know, use them to take notes, and watch sessions and code videos, because a lot of times and work structures right, as you may kind of mentioned earlier there, there are fewer researchers, and then there are everyone else. Right. And when you're expected to bring in a lot of data.
It takes a long time to sit through and go through that data, right. Please have your stakeholders help you. You just have to train them to do it in the way that you want it done.
And then last but not least, fun ways to relieve by
date. So you want to get your findings out there and get them notice and make stakeholders feel like the users.
So here's how we're not getting funding findings. Notice I'll go through a few examples
of each of those, those points. So sometimes just just
in the marketing kind of way, just get the information out there, get it in front of people, be a little bit obnoxious about it.
But because we get hundreds of reports, right, and it documents and Jackson to our emails, weekly basis, right, but do something different. So far, important findings. Don't Don't be afraid to do something like create a user core on a shoelace. Right and like, just the chick choices are cheap, people wear shoes dirty. I will just where the shoe laces around the office and everything
User quote, or Oh, yeah, I remember that research. Like, do you
use fortune cookies? Okay, bye bye bunch of fortune cookies in the office, tell everyone to have a snack. Break over the board with your local user finding what that's
like, and you'll be surprised that it generates a little bit of fun, right? There's a little bit of surprise like, unfortunately people did you get
Did you get right? But, but in general, like you're still communicating something to them, right.
Okay, this is a little hard to explain.
I did not kill anyone in this report. So for an empathy activity.
Use personas as an example your how many how many people are familiar with, with personas? Yeah, okay. So it's a it's a good way to kind of segment users
into kind of groups and buckets
and so a lot of times the new teams great are so many as you want them to start using them in their language everyday language and thinking like a certain type of user. But you know, if they read a document on all of their documents, it's it's fairly unlikely that they'll actually remember all of those personas by
if you invite them all to a working lunch and send them all invitations and tell them to act like a persona and dress like one of the personas and then you set up a scenario like a murder mystery Marie oh no someone this use this product they know someone dead right? I have to solve a mystery while acting like one of the personas that you have. Oh, so kindly suggested to them.
It makes for really interesting time. But the bigger benefit is that
Get to understand how these user types think and make decisions. And that's the piece of knowledge, right? That is really important to convey to your product teams. That's not just a funding, right? really understanding how someone makes decisions is is a hard thing to do through research findings, but directories where you make them do some kind of role playing, you get a little bit closer, and it's really funny to watch.
So that's generally all I have for you on those three kind of tenants of, of building empathy for qualitative research.
岩： powers of ten で、クリエイティブ教養みたいな。イームズの作品みたいなのを
The Making of Marvel Studios’ "Avengers: Infinity War"
Dan DeLeeuw VFX Supervisor Marvel Studios
Jen Underdahl VFX Producer Marvel Studios
Swen Gillberg Additional VFX Supervisor Marvel Studios
Kelly Port VFX Supervisor Digital Domain
Russell Earl VFX Supervisor Industrial Light & Magic
Matt Aitken VFX Supervisor Weta Digital