How To Get To Your Customers Underlying Needs?

Kate Martynova, Gianni Clifford, Mall Allpere, Louis Wilkinson at UXDX Europe. Video: https://www.youtube.com/watch?v=eKStMb5AH10

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

Introductions

[00:00:00] Louis: Yes. So this is the panel discussion, and this is where we're going to come into some topics following up from the sessions that we're running over a couple of days and from our speakers there. To kick things off, it makes sense that we just have a quick introduction round. Perhaps we have a little round robin, so start with myself. My name is Louis Wilkinson. I'm one of the Senior Project Managers and a UX researcher at Testbirds. I mainly focus my work on the telco industry, but also some betting and gaming is in there also.

[00:00:33] Mall: So I can continue. I'm Mall Allpere. I'm a Design Strategist at Swedbank and focusing on scaling our team, and also making sure that everything we do is infused in what everybody else is doing. So, climbing up that ladder of UX centricity.

[00:00:50] Kateryna: Hey everyone. My name is Kateryna. I work for Preply, which is a global marketplace for online tutoring. I'm Head of the UX research and market research team.

[00:01:03] Gianni: Awesome. How's it going everyone? My name is Gianni. I'm a Product Design Manager at Zendesk. Zendesk are a company that makes sales and support tools for better customer engagement. I've been here about two and a half years at Zendesk, and I'm mostly focused right now on our Zendesk Talk product, which is a telephony solution that builds right into our ticketing software, and on Zendesk Explore, which is an analytics and reporting tool at Zendesk.

[00:01:31] Louis: Right, thanks to everybody, and thanks, Gianni. As we go through this session, stay for this panel discussion. Please feel free to send in any questions that you have on Slack or just into the Zoom meeting chat there, and I'll try to put that forward, add to the relevant people or to the group as it comes in.

How much time goes into problems versus solutions

[00:01:56] Louis: So I guess to start off conversation, let's see, how much time do you tend to spend trying to uncover problems versus validating solutions? That's quite an interesting starting point for us.

[00:02:07] Gianni: Where's that? Yeah.

[00:02:11] Louis: So, whoever's brave, somebody jump in on that.

[00:02:18] Kateryna: Well, I can jump in, because we've been setting up our quarterly OKRs, and we've been reflecting on this, and surprisingly it was 50-50, although I was always thinking that it wasn't that equal. I think it depends on different product squads in our team, because some of them really prefer concentrating more on discovery, direction, and testing everything and validating everything via small A/B tests. However, some teams do prefer to test their own ideas. And I think it's equal because different product squads have their own cultural insight. But for now it's indeed 50-50.

[00:02:56] Mall: As I said, we're working on getting higher on the maturity ladder, and there we have to balance both. So a lot of teams are working in the latest stages and just making things better and better. And then we are also talking a lot about the double diamond and understanding who the users are, so we're doing that also. For me it's really impossible to say how much, but I think the most important thing is just to do both at the same time. Get going and keep an eye on doing both and having, as you say, OKRs that are tracking both.

[00:03:35] Gianni: Yeah. And I guess, I think Kate kind of mentioned that it's kind of different for the different squads in her business, but I think for us it really depends on the part of the life cycle that an individual designer is involved in. They could be during one quarter really focused on a kind of a problem discovery space, really targeting a lot of customers, doing a lot of research. And then following on from that, they could move into a later part of the double diamond, as Mall mentioned there. So it really depends at a level, either the teams, individuals or whole product areas could be at kind of different stages.

[00:04:12] Gianni: The short answer, though, for us really at Zendesk, we do invest a lot of time in that and feel that it's important that we truly understand what the customer problem is before we start getting through those later stages of solutionizing. We're trying to figure things out, because we inevitably end up just building things or designing things that are based around biases. And when you can't prove what you're trying to do and trying to solve as a designer, it gets very easy for a roadmap to kind of go astray, you know what I mean. Other things come in and are more of a priority, and you won't be able to prove firsthand, if you haven't done the research, that this is exactly what the customer needs, this is the exact customer problem that we should be fixing.

Techniques and toolboxes

[00:04:50] Louis: Sure. And so, Gianni, for trying to find and dig a bit deeper into some of these problems, do you find that you use a range of different techniques there, or perhaps you have some that you use a bit more regularly, kind of your go-tos that you could share with the group?

[00:05:09] Gianni: Yeah. So essentially the way we work is, and this is definitely my team, we try to have quite a dynamic approach, but across our whole, like our design team is distributed in 10 different offices around the world. And essentially we use, I know Mall just mentioned the double diamond, but we use a triple diamond. So you have the two conventional areas of problem discovery and then solution discovery. Essentially we've tacked on a third one that just follows the process all the way through to the point of getting it into customers' hands. And this is the track, the time the designers are working with engineers and they're actually doing the final designs, getting things ready, and then putting it out to the customers. Maybe there's EAP stages, there are beta stages, they're where we're doing final iterations.

[00:05:48] Gianni: But along this triple diamond that people are kind of very familiar with, we look to our individual designers to have different tools, workshops, processes and techniques that they can roll out depending on what they're trying to figure out or solve. And any time we hire a new designer, or look to any designer, we really hope that their experience or past experience is different. They've done different types of workshops, they've rolled out different types of processes. And that is, as we look at it, I almost think of it like (inaudible) belts. We add this one more workshop, one more process to your belt, and know that you can roll that out when you're trying to figure out any given thing. So the structure there, the triple diamond, is there, but it's kind of (inaudible) of the people, what they can apply or what they need to do depending on the exact thing they're trying to understand or solve.

[00:06:36] Louis: Yeah. So Mall, is that ringing some sort of close situation to your work also?

[00:06:42] Mall: No, absolutely. It depends on how you talk about it. Whether it's at the end of the double diamond or the third diamond, doesn't really matter, as long as you have one story within the company. I was thinking more that we are having this fantastic research team, very experienced researchers who are supporting all of the designers in our quite large community on getting the right tools for every situation. They are of course creating a toolbox, but then they're also, and this has helped a lot, we're having a larger kind of training for everyone who wants to join in, that Swedbank runs, and now we've trained about 700 so far on just customer centric design. So they understand the tools and processes as well. And that creates a kind of a pool to join into the usability testing and we get much higher traction there. And also joining in and starting to ask the right questions early on, for the people who are not product people, who are not designers. So I think that's been a fantastic effect of having these very senior designers who are able to just train anybody on this.

[00:08:02] Mall: And in addition, I also want to mention that, to be able to do the right things at each point, we have 10 to 15 lead designers for each of the value streams, and they are in turn supporting four, five designers, doing the right things, doing design critique, making sure that we're working in the right way. So it's kind of a mesh as well of experts and also leads in the products that kind of helps everybody to do better all the time.

[00:08:36] Louis: That sounds a good system. Kate, is this something that's quite similar to your setup? At Preply, how do things look on your side?

[00:08:39] Kateryna: It's pretty similar. However, at Preply we have fewer employees, it was like an early stage startup, right? So our team is really stretched out. So one researcher may work with three plus different squads, not only with product squads, but also with growth squads, right? So sometimes it's a matter of proper onboarding of designers or copywriters and how they can do studies on their own, what tools they can use on their own. So in our reality it's more like a mentorship type of job, and also like assistance in the research itself. But regarding tools, I think it all depends on the type of team and type of chapter, right? So if we talk about content makers, UX writers, we have separate tools that they can use to estimate the efficiency of their work. When we talk about designers, we have a separate toolbox for them. So this is how it works at Preply.

Recruiting the right participants

[00:09:40] Louis: Okay, thank you. So I think, moving the conversation in a bit of a different direction now, is actually doing some user research. And for the example of interviews, I'd really like to know a bit more about how you guys find and source these specific participants for your study. So maybe if I kick that back, start with Mall, perhaps.

[00:10:02] Mall: We're using sourcing agencies[?], because we are a bank for the many. So we're not buying for affluent people or so, we need to get everybody in the Baltics and Sweden to understand. So therefore we have agencies[?] to help us. And we are working closely with our researchers to screen properly, to get the right mix of people and make sure that we're not kind of too narrow in our understanding. So that's what we do. Sometimes we do guerrilla testing as well, and then we go out in the bank office and then we try to do it in kind of a spread of bank offices, not only downtown (inaudible). But in doing it, since we need to have a simple language, you need to go to a place where people don't know Swedish or the language well enough, and we need to reach out to seniors and et cetera. So that kind of thing we do as well. But the main part is just good screening.

[00:11:03] Louis: And so, yeah, Gianni, you mentioned previously about sort of trying to find out what the right questions are. So once you're trying to work out during that process, and also once you feel you have the correct questions, how do you go about putting those questions to the correct person?

[00:11:25] Gianni: Well, in terms of finding the actual candidates and recruiting them, it can be fairly different. I know that Mall and Kate will be dealing directly with consumers or individuals, whereas we're dealing with businesses and being in a company at scale. At Zendesk there's lots of challenges in actually finding those customers, and there's numerous different ways that we go. But there's even internal stakeholders that we need to be aware of before recruiting customers, like account executives for those accounts, and solution managers and stuff like that. They're working and communicating with the customers all the time, and we need to be aware of these relationships when we're recruiting.

[00:12:03] Gianni: What we do normally, though, is we start understanding what the questions are, as you mentioned earlier, understanding what type of businesses we want to ask these questions of. And from there, there's numerous different ways we can recruit. Sometimes we could be after a really maybe specific type of customer. So, to give a broad example, we could run a query on our own data of all our different customers. So we're looking for a customer with a specific number of Zendesk seats, based in maybe a certain location, and a bunch of different metrics like this. And we can run a query, we can find who those customers are, and then see about how we're going to reach out to them.

[00:12:43] Gianni: If the problem has initiated from the customer side, so let's say we're hearing it through our voice of the customer team, we're hearing it through our community pages, maybe we're hearing it because it's someone emailing it through an executive, then we're kind of already starting with a problem space. We're starting with a customer who has a problem, and we can connect directly with them. It makes it quite a lot easier. And sometimes what happens is we talk, or someone talks to someone in the voice of the customer team, talks to them. When it gets to the stage where design is getting involved, it's on the roadmap actively, then we'll have a bunch of customers, and with this request that we do, we start asking those questions.

[00:13:27] Gianni: And in terms of the questions we ask, that will be designed around what we're trying to figure out. Of course, we always have to make sure our questions aren't leading. We really want to get to the root of what the problem is. And I talked last year at UXDX all about research and setting the tone for customers so that they don't feel like you're tricking them, or they don't feel that you're asking questions that are too obvious when you're trying to get to the root of their problems, to make sure that they know that sometimes you may be a little bit silent on the call, but you don't want to be directing them. Sometimes you may ask them questions that seem overly simple, and those questions aren't there to trick them, it's just to get to the range of what their problem is. And once we get there, to think, is there another way we could solve this on top of the current way that's there, or the current way that they're suggesting?

How the interviews are run

[00:14:21] Louis: Yeah, that's an interesting point. So I think it's kind of, how do you actually operate these interviews? What's the kind of usual structure, or does it depend on your journey for how you actually operate these sessions? What have you found that works well for you?

[00:14:40] Gianni: Yeah, mostly our sessions are run by product designers and a PM in partnership. Now we do have a research team, and it's not unlike the way Kate described hers, where they're spread out across many different projects. Our research team is amazing and has done this, and we're so lucky to have a great team and a great resource. And they run their own rolling studies and interviews, and they have their own projects that they're focused on. So the broader team kind of uses them as consultants and they help kind of guide and shape their interviews. And we find that works really well. We have all their expertise, but we also have the PMs and the product designers having those conversations first hand, so there's no intermediary there.

[00:15:21] Gianni: The PMs and designers talk directly to the customers. I think customers find that this is really positive, that they're really being heard when they're talking directly to a product manager and a designer who is going to try to dig deep to understand their problem and then go forth and try to solve it at some point. And so mostly it will be a designer and a PM. Sometimes they will have, as I've said, research might shadow that call. And then also sometimes we could bring in some partners depending on what we're trying to solve, like content strategy or engineering people, or people from maybe even localization, we need, depending on what we're trying to solve for.

[00:15:56] Louis: And so, Kate, you mentioned that Preply, being at that kind of start of that stage of the enterprise, how do you kind of approach the situation with interviews and structuring it? Do you have any standardized practices already, or is that something that you're developing as a company at the moment?

[00:16:14] Kateryna: Well, we have templates and guides for those who just want to start doing interviews, right? Like with the proper reporting questions, with an introduction, how to probe for different pains, with all the different techniques that can be just adapted to the specific research question. But usually we try to still kind of control the design and to make sure everything is so we can properly add on to our expertise. But yeah, we try just to use common practices, common techniques to the standard, in-depth interview practices for example.

[00:16:53] Kateryna: And what works for us perfectly is just contributing a lot to the intro part, because we have lots of people who just start learning the languages, right? And sometimes they're shy to talk and to show their accent, et cetera, et cetera. So we need to show that they're on the same level, we're the same people as they are, so they would be open to us. And so I think that the intro part, the rapport part, is really, really important. So I would emphasize it.

[00:17:21] Louis: Yes, definitely, I'd agree with that. Making the kind of safe space and encouraging discussion. And really there is no sort of right or wrong answer, we want to remove those barriers as best as possible. And I'll just take this opportunity now to see if there's any questions from the audience, or to invite anyone, a reminder again, to send in anything they'd like to ask. Specifically as we go, we can do that, and we can also check up at the end.

Laddering evidence into insight

[00:17:47] Louis: But maybe on a kind of a topic-wise question that we had through earlier, to do with laddering of evidence. So Kate, how would you kind of approach this when you start to see some information come through, perhaps you start to see some patterns emerging? If you've had a few sessions with different participants already, what's kind of your next steps there?

[00:18:03] Kateryna: So we would either bring it to the discussion in our meetings with product and see if we see any potential here. We would apply different techniques where we do affinity diagramming, if we see that there is some appealing topic for further research. So for example, recently we've seen that our parent persona has mentioned something about their kids, like in different subjects on our platform. So four people out of 12 mentioned this, and we had a discussion and asked the team, should we probe further whether there is some cross-selling potential? So we've conducted separate research on that, because we see that it's a complicated topic and we need to be 100% sure that we need to work on it further.

[00:18:46] Kateryna: And we would do retrospective interviews to probe for pain, whether it exists, what are the patterns there. And in the very end, this is my favorite part of research, to suggest to them that we already created an MVP course, would you like to be signed up for it right now? Therefore we can understand that they are actually interested right now, at the moment. And we can make some conclusions if not, just asking why. So this was the recent situation. It works pretty well.

[00:19:17] Louis: Okay, great. And so Mall, are there any perhaps examples you could tell us from some situations that you've had of where some specific stories have come out, or where you've developed types of evidence through a series of different interviews with different participants?

[00:19:35] Mall: I mean, we're always building on the knowledge base we have of how people are thinking about money and banking and their lives. We had, some time ago, an idea about what a bank could be. It was kind of a conceptual idea, and we brought it to customers and they didn't like it at all. They didn't understand it. They almost felt like it wasn't for them. It felt silly. And that was very powerful, to take that and say to everybody in the organization, this is what we saw. There I would have wanted to have videos of them to show to management. Now we just had quotes, and that's not as powerful. And then we took that and made it into one piece of research and made a concept, and another piece of research and made concepts.

[00:20:34] Mall: So it kind of evolves with time, and the thing is, this is two years ago. And I think that two years ago people's perception of what a bank is is different from today. So we learned some things, but that's not true today, because the world around us has changed so much. So we need to pay, as we are paying, attention to that as well. But essentially that was just seeing those of the people wanting to understand more, have more control of what their life is about. So we need to understand the baseline of what's really behind, but you also need to see, okay, what changes when things around people change?

[00:21:11] Louis: And for you, when you're seeing these kinds of patterns emerge and this information coming through, when do you think it reaches the point where something is validated and you can really start to derive an insight? What stage does it get you to where you think that's significant, and where you can start to base decisions on the answers that you're receiving?

[00:21:38] Mall: It really depends on how big the decision is. Is it a small decision? I can make a decision based on if I get the majority of my answers. Is it a large decision that's going to kind of move the entire company? We need to go and be very sure and look in many places. So I really think it depends on how much is at stake. What we do is to take that and we move it into goals, and we kind of make it into intentions. What are people trying to do? And then we can validate if our designs meet the goals. So that's how we try to go about it: is the research actually changing our main goals, or is it changing just the design?

[00:22:21] Mall: So, maturity. It's the same usual research thing: maturity, getting the right users, knowing that you ask the right questions, being very sure about your research techniques, and all the things that Gianni and Kateryna talked about. Not probing too much, really understanding (inaudible). So, basic research in a way. Yeah, I think the main point is to get everybody on board with the finding. And as you said, you need to have your product manager in the research, you need to have your stakeholders really there, and you need to have the product designers being coached by a researcher. That's when the magic is happening. That's when everything goes faster and you can do your things much better. So I think that's the key. And I think our last speaker was also about that, that we need to have them inside of the research, and not the UX designer sitting on this side somewhere.

Telling the story to stakeholders and executives

[00:23:14] Louis: Yes, definitely, and sort of painting that story. Gianni, is this something that you've found now works well with most of your research moving forward? Is it becoming part of the enterprise there also, or is it still something that you consciously work on to make that successful, about that buy-in and that involvement from the extended kind of stakeholder group, I suppose?

[00:23:36] Gianni: This is absolutely always something that's maybe not a work in progress, but certainly a focus area. And there's a lot of PMs. There's a lot of product designers, and as we're growing there's more product designers and more PMs. And depending on where the PM's going to come from, they may be the type of PM who really likes to get involved in research and be in those interviews and be a really great collaborator and partner with the product line. And sometimes PMs have come from places where that's not what they're used to. So it's certainly something that we always focus on: how we as a product design team get the best partnerships with our PMs, but also our engineering partners.

[00:24:17] Gianni: I think that follows what Mall was just saying there. I think the final piece of the puzzle, what she was talking about, is, I find that we can't get our senior leaders, our executives, involved in all the research. Of course, our VP is involved. But it's how we tell that story as a team back to them. Tell them, without going too far into the weeds and telling them so-and-so said this or that, but how can we extract good solid insights and be able to create clear and easy numbers that they can kind of digest? Out of X number of people, why did they say this, and give them a good steer.

[00:25:02] Gianni: And I think that really helps us for the slightly bigger decisions. Maybe not ones that will totally shift the direction of the business, but the slightly bigger decisions where we need to call in more people. So we have this evidence on, hey, we've talked with all these people. I'd say it's something that was based on some quant, there was a survey (inaudible), that can give us really easy data that we're sharing, first of all, to our executives, our stakeholders who may need to be involved in those decision-makings. And then across, I think it's important for product designers to be sharing that information with other designers, with engineers, the other scrum teams. If your team is doing weekly or monthly check-ins and sharing all of the different things that are going on, I think it's important that design is sharing those kinds of research findings, giving people a taste of what their customers are saying.

[00:25:50] Gianni: So even if anyone, say an engineering manager, or even a PM on a different part of the product, is not involved in that research, they're getting a flavor for what their customers are saying. There is overlap there. It's important that they're getting the learnings and everyone's getting the max value out of that investment in research. And just at the end of the previous session, I heard there was a question about the investment in research, that it can be trickier in smaller companies. I know that to be true from a previous life I did. And it's all about being able to show how the research gives you value. So when you do the research and when you have those strong outputs, make sure that they're being communicated, and the value has got out, that it's not six folders deep in some vault with only the immediate team having access to it.

[00:26:50] Louis: Yeah, exactly. I think that can sometimes be a difficult thing, actually, when we've used the research. It's often the place that the senior management and some senior stakeholders, they really want to have the tangible. Okay, so you're suggesting these improvements, these changes, whatever it may be, but what is the kind of bottom line on that? How is this going to impact the business? Can you give us something? They want tangible things, and it often can be difficult to have it that directly. But, Mall, is this something that you've encountered, and how do you go around that perhaps? Maybe it's not such a difficult point for you, being in quite a large organization also.

[00:27:24] Mall: I think it's all about teaming up with product managers who are good at turning it into business, and therefore making sure that we are having goals for the company, for the teams, that are customer centric goals but also are business goals. So if they go well together, if it's about retention, if it's about churn, things like that, that speaks volumes for doing usability and research. So if you can get some kind of a metric of things that is good for the user, of course, long-term retention, things like that, then you need to have a really good product manager who supports you and is able to talk about the payback being much, much longer. And they need to think about the brand perception, et cetera, et cetera. So it's teamwork. That is really teamwork.

[00:28:25] Louis: Yeah. And so, Kate, you had mentioned about taking the qualitative approach at Preply. Is this something that you've had experiences with, or that's been challenging, of trying to get further buy-in for research and really shouting the benefits of user centered design and user research more broadly?

[00:28:41] Kateryna: Yeah. Well, when I joined Preply three years ago, we didn't have any research in place and there was no habit of talking to customers. That's why it was really difficult to show the value in the very beginning. With our baby steps, having some research findings and tracking what impact do we have on product development, what market and product solutions were based on our insights, whether the experiments that were launched were successful or not. So even monitoring these stats was very visual and broad, like increased engagement inside the team. Also, engaging all the stakeholders in conducting the research also helped a lot. And yeah, after that, the demand for qualitative research is that high right now, so it's really hard to prioritize and plan our quarters.

Q&A: what if the product is already succeeding?

[00:29:35] Louis: Great, thank you very much. As we're actually approaching the end of the session, we've just a minute left. We did have a question sent through from Brendan, and this was in relation to quite a unique situation, perhaps if you have a wildly successful product, and then what's the approach that you have there for further research? You don't want to be adding problems and making new initiatives for the sake of it. Gianni, is this anything you've experienced, almost too successful products and the research team is scratching their heads?

[00:30:11] Gianni: I don't think that's a problem. And I don't think that by asking the customer and by actually understanding their problem, are you making more problems? You're just finding them, to understand your customers better. If you're in a predicament where you don't know where you should be looking, what's a really good way? I know we're living in a funny Zoom world right now, but getting out there and trying to get designers, researchers onsite with customers, shadowing customers using your product. Again, really getting an idea of how they're using it, why are they doing certain things.

[00:30:45] Gianni: So if you shadow a person using your software, you'll see all the different, unusual ways they have things configured, the different shortcuts they may be taking. And by understanding that and truly understanding their workflow, you could understand, maybe not finding new problems or creating these new problems, but you could understand different ways that you can kind of optimize your product better, make it better. And no product is ever perfect, and we can always be improving them. That's got to start with an understanding of how it's being used. How are your power users using it?

[00:31:19] Louis: Yeah, exactly. I think that actually is quite a good place to end. It loops around to what we were discussing at the beginning, of the problems and solutions. But I agree with Gianni about finding where you are strong and you can continue understanding what is appreciated, (inaudible) and for your research development down the road. Okay. So yeah, I'd really like to thank you all for joining, Gianni, Mall and Kate. Thank you very much for that, and for all of the UXDX hosts. Yes, so that's now our panel session done. If you'd like to follow up with any of us, for the audience, please do connect on LinkedIn, and all our details are there, shared within the Slack meeting group also. Thank you for joining.