The 9th Listener phone-in

A transcript of Episodes 156 and 157 of UX Podcast. James Royal-Lawson, Per Axbom and Danwei Tran Luciani take calls and questions from UX Podcast listeners during a live broadcast.

Danwei, James and Per i Studio Axbom

The 9th UX Podcast listener phone-in was broadcast live in March 2017 and covered topics including: UX Book recommendations, UX Analytics, the difference between UX designers and direction designers, share buttons on websites. data scientists, delight, and normalising incompetence.

Transcript of Part 1

Per: This is UX Podcast with me Per Axbom.

James: And me James Royal-Lawson.

Per: Coming to you from Stockholm, Sweden with listeners and a hundred and sixty-seven countries from South Africa to Switzerland.

James: Before we get into this week’s show, it’s that time of year where we have our listener survey, when we ask you the listener to spend just a couple of minutes giving a bit of feedback and information about the show. How we do and what we could do, maybe better. To fill in the survey, just visit

Per: Go do that and keep listening. So today, we have a listener phone in, part one of two. And a listener phone is where we set up set up in Studio Axbom. We include Danwei Tran Luciani as usual in our podcast team. And start a live video feed allowing people all over the world to watch us record live for two-full hours, chat with us and actually call in as well. And this time around, we got a lot of chat questions which was really fun. What were some of the topics, James?

James: Well, over the two hours that we were chatting away, we covered UX Books to read. UX Analytics. The difference between UX designers and direction designers share buttons on websites. Data scientist, agile ways of working. delight, normalizing incompetence.

Per: Yeah.

James: We — and a few more topics to be honest. So let’s hear part one.

Per: This is our listener phone in number nine. And we are in disagreement about what times we’ve broadcast it before, for those of you joining in for the first time on the listener phone in, the way these phone ins work is that we sit here talking for two hours. We’re live on YouTube, and we welcome you all our listeners to join in, and you can chat with us or you can which we encourage more than anything else; also call in. The articles that you guys maybe have picked out, we have engaged on Twitter with people, we have gotten questions there, there’s always something in industry to comment on, or get upset about it. But do call in if you think we are going off-topic which we also have tendency to do.

James: We’ve actually — we actually got one question. Here is a question from Lina Hanson. Which five books do you think every UXer should read?

Per: Oh, that’s — oh, I have my bookcase over there, that’s good.

James: Oh, but that’s cheating. You can’t just read five books. Try just [crosstalk]

Per: Oh, but I know them, I know them.

James: Oh, go on?

Per: Information architecture for the worldwide web.

James: That’s the Louis Rosenfeld classic.

Per: Yeah.

James: Classic.

Per: That’s the first book I read.

James: The Polar Bear book?

Per: Yeah.

James: Uh-hmm.

Per: And Don’t Make Me Think because everybody recommends Don’t Make Me Think.

James: We were on — yeah.

Per: By Steve Krug.

James: The revised edition?

Per: The revision edition when you actually realize that people have to think. The Inmates Are Running The Asylum. Alan Cooper.

James: Of course, yes.

Per: Well, almost the sequel to that is About Face.

James: Uh-hmm.

Per: So, I don’t know how many editions there have been. One more.

James: Well, they have like normally people would recommend is James Jesse Jarrett

Per: Oh, yeah. With the model?

James: That you — yeah.

Per: I know which one… I probably have it over there.

James: You will? I love it.

Per: We’ve interviewed so many fantastic authors over the years as well on UX Podcast. I mean, I — Steve Portigal.

James: The Elements of User Experience

Per: Oh, yeah, the Elements of User Experience. Okay.

James: Yeah. Well, because that’s this.

Per: So these are old school books though.

James: They are.

Danwei: I would also like to add Sketching User Experiences by Bill Buxton.

Per: Okay, yes. I also have that up there.

James: Uh-hmm.

Danwei: He actually has a thinner version, too. I think about 200 pages.

Per: Really?

Danwei: So, it’s easy to digest for people, who’s not interested in the academia part.

Per: Yeah, okay.

Danwei: So it’s just the hands-on tips and a workbook to go with it.

Per: Because honestly, I did not read that.

Danwei: Yeah, okay. It’s a thick one.

Per: Yeah, it is.

James: As regular listeners to the podcast will know, it that I’m legendary for not finishing books, I don’t, I mean.

Per: You never finish books? So that’s just standard.

James: No, I don’t. I never finish books.

Per: Yeah.

James: And but I do like starting them and I read many good book with the first good chapter. I can — I really did enjoy Donna’s book. The User’s Journey. And then it’s “Story, I think products that people love”.

Per: Uh-hmm.

James: By Donna Lichaw.

Per: Yes. She’s coming to Stockholm in.

James: Is she?

Per: Yeah, From Business to Buttons.

James: Of course.

Per: Yeah.

James: That I knew.

Per: Well, I would actually argue if they’re about to read Steve Portigal’s Interviewing Users as well. Because UXers unfortunately they know they have to do it, but the one many other people I met are actually — they don’t interview enough. I think there’s actually a sort of an obstacle to it is that you don’t always feel comfortable doing it in the way that you should be doing it. There’s too much lab study sitting next to users, asking them the questions, but actually going out into the world and interviewing users or people in their own homes, at their office and doing deep interviews with open ended questions, that’s something people really need to work on. So yeah, read the book but actually do the work is really what I’m saying, I guess.

James: I think Uxers need to read few book that are not UX, Daniel Pink’s Drive.

Per: Uh-hmm.

James: It’s probably one I’d say to get more into the behavioural psychology aspect of everything. Coming from that angle, too.

Per: Yeah, yeah, Daniel Kahneman’s Thinking, Fast and Slow.

James: Exactly. Thinking, Fast and Slow, absolutely.

Per: I would say that as well.

James: Yeah. I’ve read half of that one too.

Danwei: Oh, on that note, I also like to just recommend a book called Made to Stick: Why Some Ideas Survives and Others Die.

Per: That’s probably those two guys, I think.

Danwei: Yeah, yeah. Chip and Dan Heath. Recommendation on how to present your ideas in a way that people will remember and care.

Per: So as usual, I realize we’re recommending books in, like, vivid, broad areas and that’s like what UX is so good at, but also at the same time, what makes it so hard.

James: Yeah.

Per: What is UX? And we’re back to square one all the time.

James: We don’t want to do that one again.

Per: No.

James: Yeah. I don’t want to.

Per: But it’s interesting, so, I mean, the recommendation would be actually probably to read one book from all of these different areas from psychology, from sketching, from — we haven’t mentioned like a programming book.

Danwei: Uh-huh.

Per: It would probably be good as well. It’s a basic way.

James: Do they do programming books anymore? I mean, because thinking about the speed dust of changes. I mean, usually case you got these massive manuals, all these kind of guides to various language.

Per: Yeah.

James: But I don’t know if they exist anymore.

Per: They do.

James: They do?

Danwei: They do.

Per: Yeah.

James: Okay. It’s not just being replaced by Reddit?

Per: No.

James: The photos and slack.

Per: Well, you could say the same thing for UX almost.

James: Oh, yeah. I mean, with some — I mean, some of the books you’ve listed, I mean, there are — there are over 20 years old. Some of the books you’ve just listed.

Per: Oh, yeah.

James: And they’re still worth reading.

Per: Uh-hmm.

James: If you’re reading a book, I mean, guys. If you’re reading a book about programming language from 20 years ago, that programming language is probably not in use now.

Per: But it would be really cool to actually have a computer, you could try out those programming languages on, because that would probably help you understand a lot stuff.

James: I mean, this one thing I think it’s really cool.

Per: I think they’re still online.

James: Exactly. And this — and this one really cool — well, there’s several really cool similar to this book. My first computer back in the early 80s, because I’m that old, was a BBC Model B Micro Computer.

Per: Uh-hmm.

James: Acorn. And there is now a JavaScript emulator of that computer.

Per: Oh, a JavaScript emulator? Wow.

James: It’s completely built in JavaScript which is utterly mind-blowing. If you’re 44 years old and brought up using BBC Micro Computers like I was, there were the computers at school, the computer that, I mean, I’ve programmed on for the first time, and you’re first connected to other computers with, they’ve made this in JavaScript.

Per: That’s impressive.

James: And it’s just kind of poooohf.

Per: Oh, somebody’s calling in. Hello? Who are we talking to?

Tristan: It’s Tristan Bailey.

James: Hello, Tristan.

Tristan: The messaging in hangout in your telegram channel earlier today, so I thought I’d join in studio on what you guys are chatting about.

Per: Yes.

Tristan: So, for myself, I sort of transitioned between my podcast in UX and today, let’s say I’ve been speaking to Luke Hay this month on my podcast at Cliff Nights about his new book with Use of Analytics in UX, and I wondered whether, so I tried to keep it quite light because my audience is sort of business managers and stuff. And to introduce them to UX and Analytics, but I would be interested to hear how you feel analytics fits with UX or your experiences, your use?

James: I’m going to come clean straight off now and say that I’ve avoided interviewing or calling or getting in touch with Luke to do an interview. Because he’s beat me to releasing the book. I’ve been working on the same book basically for about two years. And I haven’t had the time to finish it. I’d say I don’t know if it’s the same book because I’ve bought Luke’s book, but I haven’t got around to reading it, because as I have babbled already. I’m not really that good at reading books.

Per: Yeah.

James: So, I felt, Oh, God. I’ve got to read his book before I get him to come on the show so that I can — interview can see how much overlap it is and work out what I’m going to do with the interview and with the book. But I’m really looking forward to reading to, I’m finding time to read it. And I actually hope we do talk to him, irrespective of whatever it overlaps a lot with my ideas or not.

Per: But, yeah, that makes you probably the best person in the room to answer the question about the relationship between Analytics and UX.

James: I think Analytics or Web Analytics and Data is absolutely essential for UX work. And I think it’s quite amazing how little UXers do with — generally do with analytics. And when they do work with Analytics, it’s often over kind of tokenb nature or it’s … let’s put it this way, without trying to UXers down when it comes to Analytics, but it’s very easy to drive blind or at least think you’re driving with your eyes open. But you’re letting things pass by or you’re not understanding what’s going on so you’re coming to wrong insights, wrong conclusions. Now, at the end of the day Analytics is very, very technical. You’ve got webpages, you’ve got web browsers, things get entered into logged files or pixels get collected and sent to Google and things end up in a database. And then you pull the insights out with the database through. say, Google analytics or whatever tool you’re using. On the surface, it can look like this really obvious answers. Look, there’s 60% of our users that bounced. And then you go running off then start designing something based on this insight you’ve gathered. But when you start understanding deeper and looking a bit deeper, peeling the onion a little bit, you may be realized that it’s not that straightforward. I want all UXers to get more involved in looking at behaviour with Analytics, because a lot of what I see is to do with UX metrics. So, using Analytics to try and give a number for UX, a number for their experience what’s happening. I suppose like conversation rate. That kind of final, look, the grand digit you can use to say this is how good it is or bad it is. And the use of analytics for understanding behaviour doesn’t happen as often.

Per: Yeah.

Danwei: Yeah. I guess it’s always easier to measure things but making sense of that number or whatever you’re measuring is the tough part.

James: Yeah.

Danwei: And maybe people are not doing that well enough nowadays or is there not a good enough tool to be able to do that easily.

James: No, I mean, now over half of our websites have Google analytics installed, so the data is there. We’ve got a huge amount of data at our fingertips. But I think quite often we’re focusing on the qualitative side. Which is excellent, I mean, you’ve got talk to people, you got to do the interviews, you got to do the research of real people.

Per: Uh-hmm.

James: And the quantitative side is being used for metrics rather than validating things that people have said, claimed, or shared with you, or even adding another data point, another angle so what has been going on. Or boarder perspective, I mean, you can by segmenting and splitting up your data, you can reveal things that you just wouldn’t get from other sources.

Per: Exactly.

James: And confirm or strengthen what other people have said to you.

Per: Yeah, and that’s what I’ve always say. I always want several data sources that say similar thing so that if I look at the analytics side, it doesn’t say one thing, in the qualitative data says another. I want them to confirm each other and that’s why I always want more than one data source in my testing. But I think the problem with UX of course is that we always talk, it’s about empathy, I think a lot of you UXers are scared by numbers.

James: Yeah.

Per: They don’t want to touch them, because they don’t feel confident enough that this number is actually true in some sense, in the way that the numbers can be true.

James: Yeah.

Per: And that’s why they avoid them or a lot of people avoid them. That’s why I have avoided them in the past. And there’s a lot of things that you need to learn in analytics to be able to draw the conclusions that you’re talking about.

James: Absolutely. There’s a few things that you really –you can’t avoid, I mean, like I said earlier that you can’t get away from the fact that the data side stuff is technical.

Per: Yeah.

James: And if you do try to ignore that, you’re going to generate false insights. Like one thing that a friend Craig Sullivan has always said. And I usually repeat it, I mean, every analytics set up is broken. There’s not a single website I’ve worked with, I mean, I’ve worked with Google Analytics, not in the very first year, I think I started with Google Analytics in 2007. So that’s 10 years. I’ve not yet worked on a website, that doesn’t have data that’s messed up in some way. I mean, sometimes, it’s the easy, it’s the classic stuff that were, it’s not every single webpage that’s got the tracking script involved, I mean, included on. But most cases its stuff that’s more, like, with the kind of complexity of our setup tonight, I mean, it’s really, really weird stuff going on. And you don’t even… the worst things are when you realized that it’s borked and something’s wrong with the setup. But you can’t explain what impact it’s going to have on the data. So you kind of had to flag maybe to the client and said, “Look, you’re Google Analytics setup is screwed. It’s broken.” And they go, “All right. Well, but what does this mean?” And I don’t really know. Let me get back to you. And you end up for the situation where if you’ve burnt your entire analytics or research budget just trying to work out what the hell is wrong so you can say: that thing we can trust, that thing we can’t trust. I’ve had situations when maybe a client had double tracking scripts. So you end up with very, very strange effects, because cookies are getting rewritten and visits are getting reset, and you can’t say whether this one is new or repeating, you can’t say which page have gone to next before the … really weird effects.

Per: But also, I mean, if you work at a large company, do you filter out your own employees, visits, and stuff like that, the editors?

James: That’s another one of the due diligence things.

Per: Yeah.

James: Even if you’re sure that it’s kind of… even when you found out the place which is broken, you then got the filtering, and that’s more the grey area, is it broken, if you’re including internal visits, or is that just a feature. But you need to understand.

Per: Right. That’s a decision, yeah.

James: And some companies, it’s important to separate it sometimes it’s not. So, yeah, I do think it’s incredibly important that UX involves more analytics but at the same time I want this to, I want us to learn how to drive the car before we smash into a wall.

Tristan: So, I’ve got a question, so one I feel you would have to have you repeat that or at least transcribe it, because I feel you’ve beautifully set it up as a segue as we’re building a SAS application that does exactly that. A qualified check of an analytics setup. So it audits large sites, large corporate websites and checks if the data is being setup and following best practice with a sort of a respected figure who was first product manager for Google Analytics and Brian Clif...

James: Brian Clifton?

Tristan: Yeah. So I’m working on.

James: We’ve talked to Brian Clifton on the show before, he’s a very nice guy is Brian we — we’ve known for a number of years.

Tristan: We’re partnering on building this product. Yeah, but we’ll talk about that later when it — when it’s available.

James: But you’ve got definitely the right man to work with. Because he’s a very clever man is Brian.

Per: Oh, yeah.

James: And especially when it comes to the analytics.

Tristan: Yeah. Just to say on the analytic side, I do from a data back, I’m from a developer type about UX, and do a lot of early UX, with people rather than design stuff and stuff where I give off to more trained people is use that data to set up that picture and look for some areas. But then I was wondering, but then when you get actually into testing and into sort of deciding how designs is going to work out, how the interactions going to work. But it does become more of that’s focused testing or it’s more user testing, it’s more subtle. But data gets in there in some way so whether you catch that data by writing down or sorting cards, that is analytics too, it’s just not the web analytics, and maybe you could have some sort of stuff that crosses over between the two, or you just keep those as separate layers.

Per: Yeah.

James: Now, completely, I mean, I have a process diagram that will be part of the book. And basically where you will look around and I have three different starting points. And it’s kind of basically what you’re saying though. I have discovered, validation and evaluation as the three potential starting points for your work with analytics. When you’re working with Discovery that’s when you’re looking for things for the first time. It’s not going to pass through the loop before. You’re generating the insights to kind of work with. And then the outcome of that is that you come with recommendations or hypotheses. Once you’ve got hypotheses or recommendations, you’re then feeding back into your process and this time around, I mean, now we’re evaluating, we’re evaluating whether the hypothesis were true or not. So, the third variation is validation. And I say that’s when we’re you’ve got existing claims or data that you’re checking things off with the data you’ve got. So, the classic situation would be that that workshop at the start of the project where everyone says what they think, “well, our users do this” or “our users kind of never use the hamburger menu” or “they always come from the desktop” or “our main user base is in Sweden”. And these are the kind of things that you can validate using the analytics data that you’ve already got. And from that then you can build hypotheses and build, and come with recommendations which, again, we’ve fed round become evaluations second time around.

Per: Uh-hmm.

James: So, yes, you can validate user testing, you can validate workshops, meetings, and you can generate your own hypotheses, analytic data from a UX point of view for me is a loop that you can fit into any design project.

Tristan: Thanks, thanks.

James: Thank you very much, Tristan, for joining us and giving me the chance to babble about analytics.

Per: Yes, I mean that was … did you bribe him or something?

James: I didn’t. I didn’t do anything at all.

Per: But it’s so true, is when I come into projects and you hear this people, they’ve done a survey or people are happy. And then I ask them what links they click most or what pages or services that they use most, and they just don’t know. And some of them even had analytics installed in but they haven’t checked. They assume stuff and I hate when people assume.

Danwei: Yeah.

James: Yeah.

Per: When you assume, it makes an ass of you and me.

Danwei: So, we actually got another questions in the chat.

Per: Awesome.

Danwei: Now, I don’t know how to pronounce the name but it’s Ay Viberg I think. How would you describe the main difference between a UX designer and interaction designer, if there is one in your opinion?

Per: I wrote an article about this many years ago.

Danwei: Uh-hmm.

Per: Because people were calling me an interaction designer and I was having trouble with it. They still do that today. Now I don’t have so much trouble with because I don’t really care because what matters to me is the outcome of my job or my work. UX designer encompasses stuff that comes prior to interaction design. It’s the easy way to put it, I think. So prior to interaction design, you need to research and you need to talk to stakeholders, you need to understand the real problem, the business problem not the user problem. Interaction designers are really focused on user problems. UXers think about a picture than that. But the problem is when we use the word UX designer as a term we throw around UX often when we talk about this because it’s easier to think about people as working within the industry of UX. But when you use the term the UX designer, you start thinking about okay, so that’s a person who designs stuff. And design to me is problem-solving but that’s not the connotation that most people have when they hear the word design. They think about producing something, producing sketch, producing an interface, whatever. So it’s almost like front end development. That’s what a lot of people think about. So, to me there is a difference but what matters is; does the person you’re talking to understand that difference. So, we can argue or ascertain right here there’s a difference but in the end, if you can’t communicate what your output is or the work you do and the value of it, to the person you’re selling your work to or to the people you work within your team, then it doesn’t really matter.

James: I’ve noticed in the consultant world, at least when you look at the requests that get passed around or we come into contact with. Whenever the agencies that are dealing with contracts to, dishing out … what’s it called in English, can’t remember what they’re called now. They use the phrase UX designer to mean interaction designer. It’s just that simple. There is no difference to them. Now, when you’re hiring a consultant there is no interaction designer anymore under no requests going out for interaction designers.

Per: Exactly, yeah.

James: All the requests are for UX designers. So, we can sit here all evening and talk about what we see as the difference, the ideological difference between a UX designer and an interaction designer. But the market has decided, at least in Sweden, and I get the impression it’s similarish thing in the UK. Someone can ring in and tell me otherwise. But here in Sweden, anyway it’s just been obliterated has interaction design as a thing you’d request from a consultant.

Per: Yeah.

James: And I think that’s sad, because it’s a particular task. It’s a skill. It’s a thing you can do whole university courses about.

Danwei: Yeah.

James: And I think it’s sad when something you can a degree in is basically just blasted away over two years, three years. It’s been wiped away.

Per: Well, I’m actually in a steering committee of an education that called the UX designer. So I should be able to answer this question better then I just did, really. But I’m avoiding to because interaction design is part of the curriculum, of course, but there’s so much, so much more that’s part of being a UX designer. It’s like when we talked about books earlier on today, there’s so much to read up on. There’s so much you could work with.

Danwei: Yeah.

Per: I think being a UX leader, would be something more interesting to talk about because a UX leader would be someone who actually is responsible for the experience of the user or the stakeholder of the product or service. Throughout from from the contact point, from the first touch point to the end touch point. And you’re involved in bringing in the right people because people have different competencies. Even as UX designers, somebody loves doing interviews. Somebody loves doing user testing. Somebody loves doing interaction design. So you want to be someone as a UX leader who brings in all these competencies but when we boil it down to calling people UX designers, we’re not helping anyone.

Danwei: No.

Per: It’s just confusing.

James: Yeah.

Per: But right now, there is no better term because calling yourself an interaction designer is just not cool anymore, like you said, so that’s what people call themselves. And they tend not to do the whole work that would expect from a UX designer that way I see it but many companies don’t expect much more than that.

James: Uh-hmm.

Per: The problem with being a UX strategist or leader or thinking big picture is that you’re actually, you’re challenging some of the middle management what, with their responsibilities. You’re asking, “Are we solving the right problem? Are we doing the right thing here? How do measure progress? How do we measure success?” You’re asking all these hard and tough questions like the meetings we we’re talking about before, do you know the answers? You are assuming that people want this and they visit this area of the site more. But do we have data on that? And you’re questioning stuff, so, really people don’t like UXers like that.

James: You’ve also got the product manager’s side, I mean, we talked to Melissa Perri about that and she doesn’t really see that difference between product manager in the UX and it’s very…

Per: Exactly and that’s the interesting thing. Service design product management, it’s all part of the same bubble. A person who calls themselves a UX designer and a person who calls himself a product manager, they could doing exact same tasks or totally different. You don’t know. You’d have to talk to them.

Danwei: This actually reminds me, makes me think a little bit about a report that was released just a couple of days ago. It’s called Design and Tech Report. I don’t know if you guys have heard of it. Something that John Meda has been doing for the past couple of years. So he is one of the leaders when it comes to design, in interaction design and UX things. And he put together this PDF and he presented at South by Southwest just a couple of days ago. So, is where he put it. And that’s where you will see the slides and there’s the recording to it, and along with a downloadable PDF. And what he has done is to summarize the main points, what’s important in the industry right now and what he thinks might be important in the coming year. And one of the things he talked about was computational design or the computational designer, which is something another title, I guess, that might confuse people even more. But he also mentioned the importance of expanding, broadening your views, constantly learning things, and having a good insight both in business and in tech. And that pure designers that we’re focused in craftsmanship like for example, the architect or the one that only made furniture, that’s not really the skill that the industry is looking for right now. But they are looking for people who can make use of analytics, that can propose something that will change the business, and especially now when the businesses are changing so fast, and the things that we’re making reaches people instantaneously. And reaching people not just a thousand users but maybe million to billion people. So this scale and this speed sort of asks for computational designers was one of his points. So I found that a little bit interesting and a little bit similar to what we’ve discussing so far. So if you have time, I definitely recommend checking out this report. And it’s actually also very nicely designed report, I’d like to say. So it’s to dive and you can just skim through the parts that seems interesting and dive deeper into the stuff that makes sense to you. And the recording is about an hour long but I don’t know if you get more in-depth than just skimming through the report itself. So I would just recommend skimming through it.

James: Thanks to Tristan Bailey for calling in and asking about UX analytics which got me all going.

Per: Yeah.

James: When we’re talking about the books, the recommendations of books to read, it’s one thing when you’re listing just them straight off the top of your mind when you’re sat in the studio answering a question. But I realised an obvious one we missed was Tom Greever’s Articulating Design Decisions.

Per: That’s very true. It’s one of the most recommended books I’ve heard in recent years, actually.

James: And its one of our most listened to episodes.

Per: Yeah, it’s something that people do struggle with a lot, so definitely one to read as a UXer.

James: This has been UX Podcast with me, James Royal-Lawson and him, Per, and…

Per: Danwei Tran Luciani.

James: You got to read her name this time rather than you’re own.

Per: Yeah.

James: Join us next time for part two of this listener phone in. Thank you very much for listening to us.

Per: And remember to keep moving.

James: See you on the other side.


James: Knock, knock.

Per: Who’s there?

James: Recursion.

Per: Recursion who?

James: Knock, knock.

Per: Who’s there?

James: Recursion.

Per: I got it. Thanks.

James: Oh, come on, we got loads more minutes. Come, keep going.

Per: Recursion who?

James: Knock, knock.

Per: Stop.


[End of transcript of Part 1]

Transcript of Part 2

James: This is UX Podcast with me, James Royal-Lawson.

Per: And me, Per Axbom.

James: Coming to you, from Stockholm, Sweden. We have listeners in a hundred and sixty-seven countries from India to Ireland.

Per: Ha-ha, fantastic. Oh, and we want to remind you again that it’s Listener Survey Time which means you get to give us feedback on us as your host on the topics, on the length of the show.

James: Challenges that you face just now as UXer.

Per: Yeah things like that. And fantastic and thank you to all those who have filled in the surveys as well. So it gives us a lot of stuff to think about when continuing our development of the podcast.

James: This is part two of our recent listener phone in. Toward the end of March, we got set up — In Studio Axbom and opened up the phone lines or as we say, the video lines for a two-hour live chat about UX. Some of the topics that we already talked about in part one and we’ll talk about in part two were: UX books that you could read, UX Analytics, The difference between UX designers and interaction designers, Share Buttons on Websites, Data Scientists, Agile ways of working, Delight, and Normalising Incompetence.” And that’s Craig Sullivan who rang in to talk about that. And he’s coming up in this part.

Per: And speaking of incompetence —

James: No what have I done?

Per: No. I forgot to mention the URL to Listener Survey so people are just now sitting wondering, “Well, how do I answer the Listener Survey?” And I’m going to say, it’s

James: No one will ever notice you made that mistake Per.

Per: Thanks. So what articles did you guys bring? What do you want to talk about?

James: Well one article that got me going this week and I shared around a little bit was the article that Josh Clark published on Big Medium about share buttons. And basically they’d — With the work that they’d been doing on a couple of projects. They’ve gathered a lot of data about how social share buttons, are being used. So there was and some parts of they’ve been gathering data from. This is one of those topics that has come up you know, numerous times over the last ten years I reckon. I mean, when, we’ll since we first started adding share buttons in the early days of Twitter. We went crazy adding, you know, Twitter share buttons to every single website. And I remember spending a fair bit of time pointing out to people about how slow it made their websites. ’Cause you had like you know four scripts for all these different, you know, there’s Facebook, there was Twitter, there was all these other kinds of ones. And you know, that you’re on the mobile and the website just kind of like, loading from a floppy disk it’s so slow ‘cause of all these different scripts.

Per: Yeah which we’re also tracking you across the web.

James: Oh, but back then I don’t think it was, that it wasn’t quite as — That wasn’t quite the main thing that we’re doing back in 2008.

Per: Yeah, of course but you know.

James: But it was slowing sites down and as you know, when I did tracking about them I couldn’t see people using them. And as you know, I’ve been an Android user for long time as my main device. And I never click from any of these buttons. I always use the in-built share function on my device.

Per: Exactly.

James: And I know, as the UXers we all know that we shouldn’t have a sample size of one especially not when that sample, that one is you yourself as a designer. You don’t base your solutions on what you do. But God damn it, with share buttons I would do. And clearly everyone does.. uses the share features inside the phone, not the buttons. But Josh Clark in his article is putting forward some data, some evidence to back up some of their theories. And you know their theory was people don’t use them and they prove their hypothesis; they blew out all the water to a degree. So this is connected back to the chat we had which was about analytics.

Per: They blew it out of the water. Does that mean that they agreed?

James: No, they —

Per: Or did not agree?

James: They — Well, they better be — They’d proved their hypothesis to be wrong, or they proved a variation of it to be right. That.

Per: Is this confusing?

James: I’m sorry. Basically, most people don’t use the share buttons.

Per: Yeah.

James: But when they split the data up, they found there was a segment of people who really, really did use them.

Per: What does that mean? How large that segment? Well, who are they?

James: This — The people that used what it was was. It was people coming from a particular social network. They were twenty times more likely to use the share button for the network they just come from. So for example, if you click through from Twitter, to an article, you will twenty times more likely to use the Twitter share button than any other groups of people. If you’d click through –

Per: Oh it’s like “Priming”.

James: It’s I guess you could say that.

Per: You have a serve top of mine and you see a logo for that service. You’re sure to think, “Yeah. That’s what I’m related to right now. That’s yeah.”

James: Or it could be that you’re clearly a user of that service. So you’re very aware of the, of what’s going to happen when you click on the share — Or there’s more familiarity with the process and what’s going to happen when you click on that. So if you’ve come from Facebook, click on the link arrive in a page, you’re more likely to click on the Facebook share because you understand where it’s going to be, what’s going to happen. So it’s less surprising. So what Josh says here is that they — Another thing they’ve done and shown to be a successful is you super-size the preferred network and this kind of tweaking the UI I really like, you — So now we know that, all, they know cause the looks of the data. That if you come in from a social, certain social network, you’re more likely to click on the button.

Per: Yeah.

James: So make the button for that network bigger.

Per: Yeah.

James: So you look and say, “Okay, person’s come from Twitter.” We make the Twitter button big. And everyone’s — If we’re going to provide in everyone’s, we make them, where they make them secondary. In some ways it’s classic call to action stuff.

Per: It is.

James: Now that you found the evidence that says this is the button more likely, most likely from the click. Make that button the real focus. Love that way that didn’t work with the data there —

Per: Great.

James: And then they added to this by saying never more than three. They tried and three was the limit. Anything more and it kills the success of the primed one.

Per: Interesting.

Danwei: Interesting.

Per: It’s always been my theory that, yeah will. It’s also basic psychology. The more options you have, the less likely you regard to choose anything.

Danwei: Yeah, yeah.

James: And another one was verbs do better than that nouns and this actually from a — As far as kind of language UX. I think this is really interesting. That having say using the native language of the network, you’re coming from performing better than generic. So for example, having “tweet” on the enlarged Twitter button for people coming from Twitter worked better than having “Twitter”.

Danwei: I like that one what they did with about enlarging the buttons, depending on which social media they came from because that sort of ties into the thing that you deign for your audience based on their behavior. But how would you do if you will want to change that behavior? Because this way you’re just serving and enabling the continuous behavior.

Per: You’re strengthening something.

Danwei: Yes. That’s already there. What if you want to encourage more people to use sharing?

Per: We have a question in the chat.

James: Oh, okay.

Per: From Robert Luciani. The taxonomy you used to describe designers is very similar to how data scientists describe themselves. How are these two disciplines converging, diverging or related?

James: So by two disciplines, we’re talking about designers and data scientists.

Per: Yeah. You need both and it’s nice if they know a bit about each other’s work I guess.

Danwei: And yeah. So they have a common language. But I think, I sort of understand where he’s hinting at, I think. I think data scientist is their hyped title. And a lot of people assign that title to themselves. Maybe what Robert is alluding to is that maybe it’s something similar to where a lot of people all of a sudden become UX designers.

James: Robert, if you please call in and give us some kind of more background to your question.

Per: Tell us what a data scientist is even.

James: Yeah. Cause I think data scientist as in every one of those just like with UX designers as we were just saying that you get quite quickly to personal definitions and you know, there’s a scientist is it kind of like using data to you know using massive, amount of data to bring up insights across a large data set that wouldn’t normally be seen. You know, if that’s what we’re talking with data scientists then that’s not quite the same thing as what I may be talking about where you kind of like maybe would look at some of the details of and finesses of a user’s journey through data rather than big data as a whole.

Per: Can I just read this from Wikipedia?

Danwei: Mhm.

James: Okay.

Per: Data Science, also known as data-driven science, is an interdisciplinary field about scientific methods, processes and systems to extract knowledge or insights from data in various forms, either structured or unstructured.

Danwei: So I think it stems from statistics.

Per: Yeah.

Danwei: Just like — But now with data scientist they try to make meaning out of the numbers.

Per: Exactly.

James: Which is exactly what UXers do.

Danwei: Yeah, exactly.

Per: So it’s just we’re doing the same thing.

James: Which is interesting you put it like that Danwei ’cause the next paragraph on Wikipedia is, “Data Science is a concept to unify statistics, data analysis and their related methods in order to understand and analyze actual phenomena with data. So you — Did you write that?

Danwei: Yeah, maybe.

James: I’m going to check the — I’m going to check the talk page now and see the edit history and see what Danwei’s been doing. All of the work we do, we’re not islands anymore. You can’t just do that bit and then pass it on to the next person to do the next bit. It’s not production line from A to Z involving you know twenty-six people that hand off between each other. We’re increasingly working cross discipline, cross team. And you know, having feedback loops in an incredibly short amount of time and then there’s also long term feedback loops and so we’re working across time and across disciplines.

Per: It’s interesting because we started the podcast understanding or having the insight about UXers is always talking about that other people are working in Silos, and then realizing “Well, we’re working in a Silo as well.” And this definitely related to that. So what are we doing to help other people understand the benefit of our output? We’re failing miserably, because if UX designers are just in other people’s eyes interaction designers, then we are not communicating our value at all. I always think that I would rather work in a start-up cause then I could actually do all the work from start to finish the way I would want to. But now, I always have to argue for getting into the right meetings, meeting the right people. And people are always questioning, “Well, why would you want to know that?” Or “We already know this, because we’ve done this survey.” But I don’t trust — That’s a problem in itself. I don’t trust the way that other people draw their conclusions. So I think design today, we don’t use — We’d like to talk about the scientific method, most of us touch upon the scientific method, we use hypotheses. We try to verify them but we don’t work with different samples and AB groups and like.

James: We don’t work in —

Per: Like a real scientist.

James: We don’t work in labs. Yes, you can have user research, usability testing. And have a lab that you do that in. But that in itself, is not controlled in the same way you control an experiment for science —

Per: And also, you don’t have a control group. So if you didn’t use ability test, you know we’re going to have one room where you have the site as it looks today and another room the site as if you want it to look and see how different they perform.

James: and a clone of the participants.

Per: But damn it, we’re not answering the question.

Danwei: I know that there are a lot of these courses to educate future developers. So take a crash course in this. And then they usually have a small portion like maybe a one day about UX, or maybe a one hour or two hours about UX if it’s just a couple of days course. What should be taught there, do you think? If it’s become a software developer in two weeks, what do they need to know about design and UX?

Per: I think it’s behavioral science, understanding people, brains, how they work, how habits form. How small things can post huge problems, but also the value of watching real people use their service because I think those are the times I’ve seen the biggest “aha” moments when developers have watched people use the thing they built.

Danwei: I think that what needs to be taught is not maybe the skills of a designer. But like you say, to open their eyes and make them realize the potential value that designers can bring and maybe get them to understand that. Because I don’t think anybody expects someone to learn a whole profession in a couple of hours or a day.

Per: Exactly.

Danwei: So just making them see the potential value I think is what’s important.

Per: Oh someone is calling in. We have Craig Sullivan and —

James: We’ve actually already mentioned you once this evening.

Per: Yes we have.

James: On the live show.

Per: Yeah.

James: We actually, just before you phoned in we were talking about, well a combination of gamification.

Craig Sullivan: Yup.

James: And delight.

Craig: Yup.

Per: Usability and —

James: You’ve got Optimisation —

Per: Conversion rate optimization, and trying to figure out what, how they’re related.

James: And we had a conversation on Twitter with someone yesterday about delight. Well, we delight in the UX world is generally considered to be design details, about micro interactions. And you know, I was putting forward the line that –

Craig: Please. I think it’s creating an emotional states or affecting the emotional state of people.

James: Yeah.

Craig: One of the best test I ever run was twenty people I put through a usability test. Right. I got them all to rate how easy it was to use. The only thing that changed between two groups of people was the error messages. In one group, they are really shitty, horrible, evil error messages. Like, “You’ve done bad boy.” You know, or like “Invalid postcode.” That kind of stuff that you see on a lot of websites. And then around the —

Per: Your name is wrong.

Craig: It was really nice. A nice friend the error messages, right? Do you know what the people who’d had the nice friendly the error message reaction rated it over two points higher out of ten than the other one. And there wasn’t any difference, right? Their perception was easier to use. Was mainly brought about because of the copy treatment not patronizing them and making them feel what they’d fucked up when they were trying to fill in the forms. So even without making a single change can actually affect people’s perception of how delighted or happy they are with something just by not doing the things that act in the opposite direction.

James: But I was saying though that you have diminishing returns of delight that the first time it can delightful. But the hundredth time, the same thing is not going to be as delightful —

Craig: Oh, I’m not so —

James: It wears off –

Craig: I’m not so sure about that. You know, I’m still continually delighted about just how easy to use and simple an app like Buffer is. So I use Buffer for queuing things on social networks and there are occasionally new details surface on that product that delight me. But generally, I’m just delighted because it does things with full little fuss, right? So —

James: But it’s not satisfactorily —

Per: But that’s not delight.

Danwei: Yeah.

James: That’s satisfaction, isn’t it? I mean, I know we’re going to get into semantics of words. But that kind of continue with satisfaction.

Craig: No I think it’s a good issue, and a good question, and it’s one we were talking about today because I’m continually annoyed by this kind of thing where it’s like, “Uh, right.” When we just rummage around in a bag of psychological techniques right? And throw them at people like grenades. I know what these people need. They need some fucking urgency in their lives, right? Or some scarcity; we need to make them feel scared, so that they’re going to lose all their stuff, right? And now it’s just, “Uh, no.” You know, I mean it’s just like you’re trying to pack some techniques without any knowledge with customary scene. “I don’t care what your problem is. The solution is to make you feel shit scared you’re not going to be able to buy your tickets today.” That to me just doesn’t work, you know. So I kind of get annoyed with this over application of psychological principles like it’s something you do to somebody rather than an understanding about people’s worries, fears, barriers, motivations the tasks and goals that have been made been able to utilize that knowledge to change the way that your product is designed so that then it causes people to be delighted with their outcome. That’s the kind of thing I’m talking about. But there’s a lot of people still you know, Viagogo where’s the one website I looked at the other day and there’s this message with flames on it saying, “Our servers are overloaded.” kind of like, “You better buy your tickets soon.” And it turns out that actually, this isn’t true at all. So they are unethically slagging off their own service in order to try and influence customers to buy more. I’m not sure it’s a really good strategy.

James: But this is the kind of the whole thing of, that pressing those psychological triggers and buttons to stress people to worry people to basically make people feel ill. I mean you might short term be able to kind of increase you know the number of people of taking the action you want. But long term, we’re going to hell.

Craig: Yeah. But it’s just the thing that’s missing is the intent and the context. Right. And we don’t have any of those. So we’re basically guessing around them. So a classic example here is right you got a and all these messages everywhere and it’s a lovely service and I really like it. I’m not slagging off. But I sometimes feel when I go to buy, to get hotels or book stuff that I normally book. I get the feeling I’m being ninja’d, right? I’m just — These things are jumping out at me. And there’s no need for it, because my motivation level was high. I book that same hotel every time. They want to know that stuff, so why they’re throwing all this messaging at me? And the answer is because on average, that stuff works, right? So that means it works for some people, but it pisses off some other people. And I think the thing that people miss out when they’re applying these techniques is they fail to measure the qualatitive impact. When I was working at Auto Glass, because we’re able to run experiments and collect the NPS scoring on both sides of the cohorts, we could run an AB test, we could go into you know what. that version B make loads more money, but it’s kicked our fucking NPS score in the teeth, right? So we’re obviously doing something that’s working. We got higher conversion, but at the cost, of how people feel. And I think we —

Per: Oh, that’s fantastic.

Craig: We do measure that. So I was talking to the guys at Google today about voice as a transport. And at future, my analytics data is going to have to include voice stress data is going to have to include facial expression, recognition, micro gestures of my face. That thing has got to know that I’m pissed off or I’m struggling or I’m going through friction. So my house is going to need to actually read some emotional indicators. There are even ways of working out kind of a rough level of emotional state by using Wi-Fi signals. That’s kind of scary. But you know, even if you had the basic device that had you know will send the data back about your heart rate monitor to your house AI, right? That kind of information could potentially be used to actually find out because imagine I’m talking on the phone. I’m talking into an Artificial Intelligence about buying some tickets and it makes me angry and they just shout at it, right? How are we going to detect that? I mean, web analytics is kind of relatively easy. But the stuff was going to be so much more complicated ’cause there won’t be a web page anymore. In fact, there may not even be any` interaction like we think of nowadays. I may ask the house and the house will go off and buy a pair of tickets for me, a concert by going to a ticket arbitrage site, right? So it becomes a platform of the service, right? So then my whole experience is how did my house AI is at interfacing with all these other systems? It’s not about how good their websites are anymore.

Per: So we put works into your mouth sort of. Did you have a question for us, or were you just calling in to be nice?

Craig: I was just calling in to say hi. I was kinda, I did a presentation today at Google. Just sort of about how broken a stuff is. You know, I think if we’re a car manufacturer’s and we are building stuff that’s great then people would be you know dying or really unhappy. We’ve got millions of cars getting recalled. But the thing I can’t understand is that you know, some will all go, “Oh you know, it’s a bit funny that website” Or you know like, “Oh, it’s just it’s really hard to use.” Or “I just have to fill this out and go through one of these form things, right? So you know, they’ll kind of accept these level of defects in a web product, right? But if you had the shit in the car, “Ugh, I can’t read the speedo at all in my car, right?” And the brakes work intermittently. And occasionally, it crashes and the sat nav home and the entertainment system totally crash. But here is a great car. Stop you don’t hear these things with these other products. So I’m almost like we have actually you know — enlured people, we’ve coached them into accepting all level of defects and web products.

James: We’ve normalized that. We’ve normalized the rubbishness –

Craig: Yeah. We’ve normalized shit. And there’s a lot of shit web experiences at that. So like, you know. You wouldn’t go and spend this amount of money on a service or a gym or some sort of service that you’re buying and accept these levels of sort of casual defects tossed into it. But the worst thing is, is it’s not just us pissing off people; it cost a lot of money. So one of the geeky exercises out there recently was workout when you rip the microwave container lid off a microwave meal, and the plastic doesn’t come off, right? So you get a lot of square bit round the edge or it rips into eight or nine pieces. How much time does that need to sort out? And I actually worked out its 850,000 days of lost time, in five EU countries a year. We’re talking nearly a million days of human life sucked out and forever gone packing bits of fucking plastic out of their dinner because the designers couldn’t put a fucking lid on it or a piece of plastic that came off in one, right? So because of their design neglect, they have now injected a loss of a million days of productivity into Western European economies just through that one thing. So I mean, I can’t; but I haven’ even got round to quantifying the kind of defect levels and the frustration. But this stuff must cost a lot of emotion, anger, adrenaline, stress. I mean, what it’s adding to the sum total of human existence is fucking misery, right? But nobody seems to get punished for it.

Per: Craig you have just summarized and answered the most of the questions that we’ve been frustrated about the past hour or so in ten minutes which is fantastic. We love it when you rant Craig.

Craig: So you guys later and have a lovely evening.

James: That’s so great, thanks.

Danwei: Bye.

Per: Let’s talk about takeaways from today’s show. My biggest takeaway was Craig saying we’re normalizing shit. Because that actually aligns very well with what we were talking about previously with settling with solutions that we’re not quite happy with but we can’t figure out how to get mandate or people to, people on board to do it the right way.

James: Yeah. Still disagreeing about what’s the right way; still not surfacing value. Although the value we are servicing is deemed good enough. I hate the whole good enough thing. It’s depressing.

Danwei: Yeah.

Per: But does it have to be?

Danwei: And I agree. I think it’s sort of depressing to settle for mediocre. But it’s a lot of time and a lot of effort that needs to put in to push beyond the eighty percent, probably much more than what it took to even push to eighty percent.

James: And that’s just the quality aspect that we’re talking about. I mean, if we flip it and look at what we’re doing to the younger generations in the way that we’re combating gamification or some of the psychological aspects of stuff you know. We’re constantly producing service and things ad games that are making kids’ lives a misery in long term. You know, we’ve got net hate and bullying that at, at levels we’ve never experienced in the history of mankind. And we — From bullying each other this much and this openly and you know this often ever. You know, you use to be able to kind of hide into the safety of your own home at the end of the day. I mean, now you can’t. I mean you’ve got a device that’s constantly attacking you and we’ve got snap chat that’s giving you, that’s encouraging you to kind of do something every single day.

Per: But it’s not the reason you also are using –

James: Snap streaks and pokemon streaks. All these things that encourage daily use of stuff. Not because it’s good for you.

Per: No, not because it’s good for you because it’s addictive.

James: it’s cause it’s good for the company or whoever’s producing the service or the app.

Danwei: And I just wanted to try to flip it. Because I think all these services are sort of also inspiring. The fact that we can make things so easily, so quick and dirty. I mean quick and dirty doesn’t have to be a negative thing as long as the quick and dirty is not the final version of what you want to make.

Per: Well, you could actually argue that this is a quick and dirty solution for a live show. Well, there’s lots of cables but it’s I mean, it’s not a real studio and a television studio. But we can go live to the whole world.

Danwei: Exactly, yeah.

James: And we’d like to think it’s delightful.

Per: I think a lot of companies start out with the right intent. They see a problem and they want to fix it. They see a need and they want to fill that need. And then they have to make a living out of it, and they panic. And they realized that “Shit, we have to make more money. How do we do that?” And they go to these conferences and listen to these talks about people talking about conversion rates and you think about, “Okay. So I need to do that.” And they look at these numbers. “Okay. So if I do this then more people will spend more money on me.” Excellent; and that’s what happens. I wish people would take a step back and figure out, “Why did I start this business in the first place, what was my end goal? What did I want to give people?” And if we can get back to that thinking about my intent, my intended outcome for other human beings and focus more on that, finding ways to measure that and not just measure the short-term outcome for the business but the long-term outcome for the individual that you’re talking to on the other side.

James: So thanks to Craig Sullivan for calling in. And also to all of you that posted questions via chat during the live phone in and that we’ve answered in last week’s part one, this week’s part two.

Per: Excellent fun with all the chats, kind of hard sometimes to actually be talking and looking at the chat and being with everybody at the same time. But tries…it’s testing. It’s fun, but it’s testing.

James: Yeah. It’s somewhat it’s easy if we present when you’ve got a person’s face on the video rather than a text question to respond to.

Per: This has been UX podcast with James, Per and…

James: Danwai Tran Luciani.

Per: Thank you for listening. And remember to keep moving.

James: See you on the other side.

Per: Knock-knock.

James: Who’s there?

Per: Yodelehi.

James: Yodelehi who?

Per: I didn’t know you could Yodel James.

James: Yodelehi-hoo.


[End of transcript of Part 2]

This is a transcript of a podcast hosted by James Royal-Lawson, Per Axbom and Danwei Tran Luciani recorded for UX Podcast in March 2017. To hear more episodes visit or search for UX Podcast in your favourite podcast client.