This is a Getting "Software" Done podcast, but we made sure to touch onboarding as it relates to the creation of products and workflows.
Today Jeff is joined by Shmuel Bollen, a UX practitioner, and teacher. Shmuel has been creating interfaces for over 20 years- from his time at Bose to his current teaching at Bentley and creating enterprise UX for ACI Worldwide, a billion-dollar payment company. Shmuel discusses how he started creating documentation on the new Internet in the mid 90's, and his journey to where he is now. Shmuel and Jeff spend a bit of time talking about how creating a great product can reduce the need for complicated onboarding and large teams to support that.

Transcript:

Jeff  01:11

Hey, there, it's Jeff from the GSD podcast and I'm going to consider this one A S is for software, although we definitely do get into some services, stuff or things as well pardon my bad English there. So on the podcast is a smooth on who he was introduced me as friend of a friend who's just super passionate about UX, and design, essentially building the right thing for the for the customer. He started back at Bose, you know, almost 20 years ago, and really dove deep into human factors and all the fun stuff to measure emotional engagement. That's actually the name of this course at at Bentley, he's now an instructor at the User Experience Center at Bentley also has been part of a company called ECL, worldwide, with a ton of enterprise customers, and they're helping define user experience for enterprise customers there. So, you know, we tried to pertain the conversation or harness it into to the services realm and the way we kind of did that was focusing on the customer journey. And you know, if we have a bad feature set, a bad onboarding experience, how that pertains to services, as we know, services covers a lot of things these days. And one of them is a lot of sass companies using the professional services team as the onboarding and implementation team. So that was my sort of framework. In my mindset, as we went into this one, I hope you find it enjoyable, and I will talk to you soon. Enough. So here we go. We're, we're rolling. In it's, it's, it's great to meet you small. And one thing that I really love is that we didn't work together, I've talked with so many people that we just have this sort of shared experience. And some people are like, Oh my god, another Endeca story or something like that. But we were connected from a mutual friend, who I considered, you know, kind of kind of really deep into UX. But right before we were about to start chatting, she was like, You know what, I'm just not feeling it anymore. And so she's like, but I've got the guy for you. And so, I definitely checked out your background and said, this would be amazing. Um, so, you know, for you, and for some of the people listening for the first you know, 10 or so episodes, specifically focusing on, you know, services and the customer journey and onboarding and implementation. But as the more and more people that I talked to, a lot of those are tied to the product, and specifically the user experience. And so when I asked when you wanted to chat about and you said, UX and enterprise software, I was like so. So this is a hot topic. So I'd love for you to sort of tell everybody what your background is. And, and then and then we can dive in and start talking about, you know, the good and the bad of enterprise software and user experience.

Shmuel Bollen  04:14

Sure. Okay. So I came out of college in 87 I guess and had a head full of liberal arts education. And really didn't you know, there really isn't anything constructive I think that you can do for a living with that. I agree. I political

Jeff  04:35

science major.

Shmuel Bollen  04:38

Yeah. So I got involved in a number of different kind of entrepreneurial adventures with different friends. And I began to publish my own poetry and short stories and and I love laying out those books myself, right. Wow. Like cutting and pasting things I'm telling Can't get to Kinko's and having them back, you know, printed and bound and, and I ended up taking all of that. Not really knowing what I wanted to do, knowing that I really would like to have some stability and have, you know, a career within which I could grow, and one, which, you know, you gotta make a living person's got to make a living. But I didn't really want to spend a whole lot of time struggling to make a living, I'd like something stable. So I went to a temp agency and ask them, you know, can you get me into a company that does everything, and then I'll look around and figure out what I want to do from there. And they sent me two boats, and I got a job at Bose. Nice. Yeah, 1994 just, you know, processing return speakers. While I did that, I began to realize that there was no documentation for the processes that we were doing. And in this in the same way that I had put together, these little poetry books I put together on on my lunch breaks, I would borrow, there were a few people in an officer that used computers to do technical writing. So I asked one of them, Could I could I just use your computer during lunch, I taught myself how to use a computer. And I put together three separate manuals for the three main processes we were doing, cutting out pictures and sticking them on paper and printing things up. And same thing I did with those poetry books. And then a position was posted on the board for for an information specialists, something something like that information specialists that can't can't remember. And I use those three books that I had produced. And I, you know, went in there and interviewed for that position. And got it. And, you know, kind of overnight went from an hourly employee with to a salaried employee with an office with my own computer with. And I really just just ran with it from then I think, within a year, so I was a material information specialist. There's another term that they use for me. It's escaping me right now. But I made, I was in charge of documentation and that kind of thing. And there were servers, right? Because we hit factories,

Jeff  07:11

deploying servers.

Shmuel Bollen  07:14

So you got to put stuff on the server, make sure that they're gotta get to the server. So this isn't by this, but this time is probably 9596. And my manager told me, Hey, why don't you go to comp USA and take an HTML class, see if you can use the web here at Bo's. So I went to comp USA for three nights, and I learned how to write HTML. And I came back into the office. And I told him, we're going to have to do this, we're going to have to move away from these servers where we store documents, we're going to have to create basically an HTML interface for all of that stuff. Because right now, we have things stored, but it's really hard for people to find them, we have to learn our own hierarchy. And, yeah, go ahead and spend a little bit of money set up a server, do whatever you want to set up a server, the Bose at the time had an intranet that was created by the engineering team, I had one page with just a few links, you know, very, very basic. Within a year, I had created a whole manufacturing information systems, intranet, with a with a web team. One in Japan, one in Canada, one in Ireland, one in Mexico. And one summer in the US in Michigan. So I trained people how to write HTML. And I created this kind of ecosystem where you it was much easier to find things and now these, these manufacturing sites had kind of an identity. So many, the manufacturing division really kind of became something else. But still, to me, it was like, Okay, it's, it's just like a layer over the servers. Again, it can't just be about finding things. It's got to be about more than that. So by 1999, I had hired Human Factors International to come in. Yeah, do some training about HFI. It's not just about the fact that you have a website that works, right. That's very engineering centric approach. Why? Why does it work? What does it do? Why does it benefit people? You're gonna have to do a lot more work with users. It can't just be, but then it was called the intranet webmaster. That was my job title. And that was like, as much as I enjoyed having a superhero job title. I was like, Yeah, but I can't be a webmaster. This has to be for people to use. It can't just be a master slave relationship. And it was not long after that. Early 2000s. I came across a book by Jeff Raskin called the humane interface. And he does quote from that, which is the way that you accomplish tasks with a product, what you do and how it responds. That's the interface, okay. And then he has like a part two, which is, as far as the customer is concerned, the interface is the product. Right? Okay, they don't care about anything that all the hard work you did that they can't see. They don't really care about what's so engineering companies to this day, right Google, you know, you name Microsoft. They're very engineering centric. And the interface is not really the product to them, the user really has to do invest themselves in figuring out how to use their software. My belief, since the late 90s, has been that the dog should Wag the tail, tail wagging the dog. To this day, it's very, very enterprise software is very, very heavily software driven. I think it's, it should be user driven, the experience should be user driven. So that's a phrase that I saw once about six or seven years ago, but in a marketing context, I haven't seen it since I've kind of adopted it as my own mantra where I work, yeah, we're going to a user driven experience, which means we don't tell people what to do with our solutions. We let people do what they want to do with our solutions.

Jeff  11:17

That's awesome. It's funny, because you would think, in the world of SAS, that we're in, that if you don't like it, you're gone. And six to eight months, right, you sign up don't like it, you know, move on to the next thing you would think that people are, would become more user driven and use this term and things like that. Have you seen that? In the SAS world at all? Or, you know, obviously, the standard is Salesforce. So, yes, which is trying to roll out a new interface, and people hate it. But, you know, I, it's funny, whether I'm at been helping out a large company, or a small company, the UX really just becomes it's not really, it's not part of the equation until they want to feature, you know, it's the term that just keeps going through my head is that it's like a buffet table at a Chinese restaurant. And here's some features and it out.

Shmuel Bollen  12:21

Yeah, or even, you know, the Chinese restaurant has a certain number of things that they can make profitably. So they decide, well, this is what we're going to sell and offer, rather than trying to find out from people, what kind of things would you be interested in, what would make you come back, what would make you tell your friends about it, it's just well, we can get all this stuff cheap. And all we got to do is have cut up vegetables and three different kinds of cut of meat, and then a bunch of different sauces, we can just throw it together and people will eat it, they might not come back for a long time. But it doesn't matter, we'll just get lots and lots of people to come in, because we'll offer fairly low price. And now what you're talking, Jeff, is you're talking commodity. So I think many, many enterprise software companies find themselves in that commodity realm where there's low margins and low loyalty. And they just keep churning through new, you know, onboard onboard, onboard lose people. But as long as they keep them for a certain amount of time, you know, then that's fine. And they will make some effort. You know, there's SAP is another culprit, I guess I would say, I met I met the director of usability for SAP back in probably around 2000 or 2001 2002, when when Bose was was using SAP very heavily, I remember asking him, what do you what do you guys even do? Like, what have? What does your group even do? I don't see any evidence of anything in the software. And he had to admit that, you know, the engineering team has just such a heavy voice, that he, him and his team really struggled to make a difference. But they viewed themselves. I think as you know, the plucky underdogs, who have a belief they have, they have commitment to a vision, and they're going to keep hitting it until they get to execute on that vision. And to certain extent, you know, that's a classic UX standpoint, you're, you have a belief and you're going to try to convince people and you're going to so I I myself, you know, fell into that category for many years. But over the last five years, the company met now. You know, various, you know, various things came into play where I was able to establish myself with a tremendous amount of credibility. I was able to do something somebody at Bose taught me once which was compromised where necessary to preserve a working relationship. Okay, so I and I, and I didn't want to do that for many years. But I began to do that I took I took a much longer points of view a longer game. And you know, I picked my I picked my advocates, and I worked strong and leave with them. And I found out who my adversaries were. And I made a special effort to go to them and understand, you know, if I can't understand exactly why somebody is so adversarial to us, I need to really understand their position as well as they do. In order for me to begin working with them, I had to go and do that. And I did that. And

Jeff  15:19

were there any common trends? In You know, I was in a bunch of product roles. And I've certainly worked with very strong dev teams where a dev created the product, and in, you know, they'll you give them some, the UX team would give him some, some designs, and, you know, come back with rounded corners, and everybody, you know, Navy SEALs died and everything else like that. So were there was it like difficulty? Or it was just like, we don't want to deal with this. There are common themes in that, or?

Shmuel Bollen  15:49

Yeah, it was just like you say, right, the engineering team believed that, or the depth that, you know, the development teams would believe that, well, you know, we can do an interface, it's just code, you know, let's put something on top of what we're building the business rules that we're building, you know, the transactional task flows that we're building, let's just throw something on top of it, we'll have somebody look at it later. All best intentions there. But, you know, usually typically unwilling to make any significant changes. So where you run into a situation like that, what I would typically do is a heuristic analysis, where I'll assess the the issues, identify the issues, and assess them for frequency and severity. And then add up those two scores, you score them, and then you end up with a list of high meet high, medium and low issues. And then you then you're able to show Product Management. Look, there is a problem with this task flow, the problem is severe, it's preventing people from completing the main task flow needs to be fixed as soon as possible, here's how you should fix it. Here's the design that you should use to fix it. And here, and here's how the design tested. And now look, I took the score from this to this. So when I when I invested in doing that, I built up some credibility, then I didn't really have to do that again and again. But that was kind of the trend is that something's already in existence. And you're brought in, you know, later on, you have to it takes a long time to get brought in at the very beginning. And teach teach people look, no, no, no, no, the user is the dog. We're, we're the tail.

Jeff  17:29

Yeah, it's so funny. And then by the time, just what I've seen, by the time comes in, that you're brought in, you know, you have to break down a lot of concrete to make things right. Again, it's so built very hard, very hard.

Shmuel Bollen  17:42

So that the thing that I've done lately, that's been really effective. And when I say lately, I mean, like the last 18 months, is there's a guy called Jeff sorrow, who runs a company called measuring, measuring you. And he has, he has uncovered a strong correlation between system usability scale metrics, and improved revenue, and net promoter score and improved revenue.

Jeff  18:12

Speaking all the SAS DC numbers, you know, that's, that's the terms that gets them excited right there.

Shmuel Bollen  18:20

So what I did was I took those two. And where I work, I've, you know, I've been using those metrics for years. So people, where I work understand those metrics, being able to show that there's a strong correlation with improved revenue that opened a lot of eyes. And then I created a matrix, where I plot both of those scores on the X on the Y axis. And now you have a quadrant, which I call the world class quadrant, where if your score is above a certain point, and your NPS score is above a certain point, then you're in that quadrant, your product is in that quadrant. So I can measure a current, you know, a current user interface, show a product manager, you are not in that quadrant, you are not positioned for improved revenue, here's the design changes I recommend. I've already marked them up, I've already tested them. This is where I predict you'll be if you if you agree to make these changes. It's pretty huge.

Jeff  19:19

And I'm sure that you could probably also draw churn and renew, renewing metrics into that higher quadrant as well to be able to be like, if you're a VPS as well, people aren't gonna sign up. They're gonna go for the next. The next one. So yeah, yeah. This is actually probably a pretty good time to segue. Well, we talked about, you have a lot of people that listen that, you know, they're working in, you know, the customers signed up, and now we're trying to get them live. Right. So, especially in enterprise software, especially in SAS these days, you know, sales team goes out, you know, and they sign up, you know, say two 3000 users for your product. And, you know, we're trying to get them live in a certain timeframe. And I've had this perspective for a little bit that if product and UX was perhaps a little easier, we didn't, we won't have to do as much onboarding and training in that, it can be a lot more self service, and then costs will go down for trying to get a customer alive. And so I'm curious when you're how this kind of factors into UX in the whole sort of new customer, new customer adoption, how easy is to use and understand the product? And what are the measures? It sounds I think we touched on about a little bit with that quadrant that you were talking about. But I just want to sort of get some perspectives from you, sort of in this world that we're living in with, you know, it's very easy to sign up and buy software, with a credit card and get going. But how do we make sure that the users can adopt the software successfully? Instead of just saying like, Yeah, screw it, this is this isn't worth the hassle?

Shmuel Bollen  21:04

Yeah, so I think that that's where you would, you would probably look at some emotional engagement type metrics would be interesting, and possibly even measuring some psychophysiol metrics. So one of the things that I really enjoy is measuring emotion. And I teach a course in Bentley University's User Experience Center certificate program, where we actually get into that, and there are two, there are two things that you need to measure, if you're looking at emotion, one is called the arousal and that and one is valence. So arousal is could be positive or negative, depending on the valence. That means you're you're trying to figure something out, or you're experiencing something new, like your onboarding, what's the valence? Is it positive? Is it negative. So there are there are tools like eye tracking, which will show you things like where a user is looking and where they're fixating? Eye tracking, to an eye tracker, engagement and confusion look the same. So I Yeah, so I

Jeff  22:11

like to realize that because we were always looking at like crazy for so long.

Shmuel Bollen  22:17

Yeah, it looks the same, right? A fixation fixation could be positive or negative, that could be enjoying it, or they could be trying to figure it out and getting more and more frustrated. So when you when you begin to measure, valence arousal and valence, which can be done for just a few 100 bucks, little sensors that go around your fingers, then then you're starting to see what is what are the emotions caused by the first few seconds of the experience the first, you know, three to 10 seconds, maybe, which are going to bias the rest of the experience, and they're going to bias the rest of the experience in a number of different ways. You know, first of all, we as humans, right, I think we're heavily designed for survival. So the first, the first half second or second of a new experience, you're typically stressed out, you kind of aroused, you're on alert, you're trying to figure out, am I in danger or not, you go into a cocktail party, you go into a business meeting, you get on an airplane, you get in an elevator, whatever it is. So the more we can do as designers to facilitate somebody getting through that initial phase safely, the better off we are. So what I do is I look I design things with very, very low visual complexity, and very low textual complexity. Or if you want to flip it around, it would be called perceptual fluency. I use regular shapes, format text so that it's an irregular shape rather than a jagged shape. I'll use simple words, I'll use colors that are not too high or not too low in terms of Chroma values. Those provide high perceptual fluency, which leads to high processing fluency. Because you're able to use your cognitive resources to figure out what I'm supposed to do next.

Jeff  24:10

Absolutely. So you're not having to use up too many cycles in your brain to figure out what's going on, I think, is what you're saying. Yeah, yeah,

Shmuel Bollen  24:16

exactly. I can get right into the task itself, rather than, like, you know, take a chair right. You look at a chair, this is Don Norman's thing, a chair or doorknob. They're designed so you know how to use them. When you buy something that's not designed like that. You got to figure out how to use it. You're like, I don't want to figure out how to use it. I want to use my energy to use it.

Jeff  24:35

Right? Yeah, it's like me and Ikea.

Shmuel Bollen  24:40

Ikea is a great example. Like, you get people get annoyed, really annoyed by it company or product or service that they've got to invest a lot in. It's really a you know, I think that that inflection point is just a few seconds of that initial experience before you buy as negatively or positively.

Jeff  24:58

That's amazing. Wow, that's That's some fascinating stuff. Oh, how was that is that must be pulling people into a lab in cooking them up is is that there? There's no browser tools or anything to be able to get those analytics correct the arousal on the Vaillant?

Shmuel Bollen  25:17

Well, there is there is remote eye tracking, right, you can do that, I think through through the camera on on an iPhone. But yeah, it's yeah, there is, yeah, there's a company, I just learned that actually. But yeah, if you're gonna do psychophysiol metrics, it's, it's high touch, you got to get people either to come in or you go to where they are so and you're, you're actually giving me a nice segue there. And so the way, the way I tend to handle onboarding with with enterprise software is, you know, I will go with our marketing our market facing teams, I will figure out who are our big customers, approach them about participating in a design partner program, where I get access to their users. And we start that, even before any design is done, we start understanding personas, you start understanding the tasks, I'll even start validating, you know, very, very low fidelity wireframes. With users. So by the time the product is built, the onboarding is very easy, because a good number of users at a good number of our bigger customers will have participated in the design process. So

Jeff  26:28

yeah, then onboard that sense in the world, and it just drives me absolutely insane. When that doesn't happen in it's, as you know, is because you're in the role, your career. It's it's kind of brought in last minute, okay, how do we make this look a little better? Right? And that's just such the it for companies doing it wrong? It's an afterthought, oh, we'll just bring a contractor in for 20 hours to make this look pretty or something like that. And versus ground up? Let's figure out the best way.

Shmuel Bollen  26:59

Yes. Yeah. Right. Right. I mean, you know, getting over that hump can be really difficult. There are analogies that you can try with people, you can say, well, you know, would you would you ask somebody to build a house for you, and just give them basic requirements. And then say you just do whatever you need to do. And at the end, I'll come I'll just hire a painter.

Jeff  27:21

Oh, my God, this is so funny that you said this. This sounds like a terrible, obnoxious thing that I'm about to say. But like, I literally published a blog post last week. And it was around some of this. And I said, basically, hey, my dad was a plumber. Right? And if you just hired him, in the carpenter, in the electricians and the painters to all show up on a concrete slab one day, do you think you're gonna get the response? You know, so. So it's exactly, it's a very easy metaphor for people to understand that, like, you know, your version of two bedrooms, and one bath is gonna look much different. If you don't have a plan in front, for what you want. Yeah, yeah. And

Shmuel Bollen  28:00

it helps it's a help to do that. It doesn't always, you know, it doesn't always have like, a lasting effect. So you have to have some endurance, right? If you're going to be an Enterprise UX designer, you have to have some endurance. A good example is with color. For example, I'll get into conversations with people about color, and design, as versus art, you know, they will confuse design with art, where art is the creator of the artist expressing how they feel internally, design is not really doing that it's it's trying to accomplish something is trying to impart meaning to somebody so that they can then take that meaning and act on it. And when it comes to color, you know, we understand fairly well, what kind of colors people can see easily what kind of colors have have a kind of semantic meaning to them. And what kind of colors are going to work? Well, when it comes to just human perception? Those aren't always the colors people sort of like to see, right. So I'll get I'll get complaints from a customer and praise from a customer about use of color. And the product management teams will only hear the complaints. And they'll say, Well, you know, we need different colors, for example, because we got complaints and even though I can show them what we got compliments as well, what that tells you is it's a battle you really can't get into you have to choose colors for reasons other than personal preference. So what I did, actually over the last month, was I created a color palette. So I used I use we have a semantic palette now which is, you know, bad, good. Warning, and neutral. And there are those are standard colors and there are six brightnesses of each. You know, when you change the brightness, you'll start with a standard color and then you'll you'll decrease the brightness by 20% and then 50% and then you'll increase it by 25% 50% 75% So now I've given you, it's not just four colors, it's 24 colors you can choose, you know, you can choose.

Jeff  30:08

Yeah, yeah.

Shmuel Bollen  30:09

And then we have, then we have a, another set of color palettes based on our brand, because our marketing team has invested a lot of money figuring out what what attributes we would like to portray, we don't want to be a commodity, right, Jeff, we want to be a premium brand. So premium brands have have qualities associated with them. And marketing teams will, will come along and tell you what kind of imagery and you know, what kind of colors should be using take the two main colors, and again, create those six versions of that of those colors. And then, you know, sort of standardize all of those. And even within that, when I go back to people, product management, tell them here's, here's a whole variety of colors that we're going to be able to use now. You know, I'll be asked again, you know, are those are those not too bright, shouldn't be, you know, for people that are using our software all day long, shouldn't be happy. Are those colors not too bright. So to go back into research into hue, and to Chroma, yeah. But to to understand as long as you as a designer are not making decisions based on your personal preference, that you're doing it based on research. And so

Jeff  31:20

it's so funny that you said that I was working on a project. And the UX was actually really well done. It was a project where my team was the dev, you know, professional services, engagement, you know, customer hired my team to do the development, and then another team to do the UX. And the UX was really good. But when it came time to implement the colors, the designer was putting a lot of their creative license on to it, and it just was off. And that's kind of how it was explained to me, which was that, yeah, it wasn't matching, sort of what the UX was trying to do is more of like, what that creatives person mood was that day. And then when they came back, when I when we said this just seems off, we're not really quite sure. They went back and took a look. And when I talked to the sort of the creative director, after they made these changes, I'm like, wow, that was pretty fast. And we were in, I think it was somewhat along the lines of what you were saying, which was, there was a little bit too much of the personal personality of the designer, in a lot less of the what we're trying to do as a product here.

Shmuel Bollen  32:34

Yeah, yeah, absolutely. And, you know, we can, human beings can detect, you know, millions and millions of colors that are really only, you know, red, orange, yellow, green, blue, indigo, violet, but obviously, there are when you adjust hue and shade and Chroma, you get millions of variations. And they, you know, they should relate to each other, right, you should have a color palette of different colors that are in the same sort of range, same color family and their color palette generators that are out there, that people can use when when that's not done, or when you use colors that are too garish, right, the Chroma is too high, or too muted than those communicate emotional things that we're not always ready for. Yep, yep. So a designer, like you're saying a designer who's more on the art side of things is going to use color to express themselves.

Jeff  33:24

Yeah.

Shmuel Bollen  33:27

Other people will pick up on that, and they might not share those. I'm not here for that I'm here to do my taxes. You know, I'm not here. For anything else. So

Jeff  33:37

this is so let me ask you this, then. Since this, obviously, is very thoughtful, and, you know, researched kind of driven work, how do you balance that against, you know, the classic, deadline driven approaches from enterprise software.

Shmuel Bollen  33:57

So we, we have a design system that we use, so a lot of the research into color, and into, you know, various perceptual cognitive issues, is kind of already it's already done, right. So I'm the, I'm the design lead for, for the whole company, I've got accountability for every single one of our products across our six main solution missions. And that, and that includes a long term vision of sewing it all together into one super capable, you know, ERP for payments, experience. And you cannot do that successfully with a very small team, at least, unless you have a design system. So all of those artifacts are already built, they're already tested, already defensible from a research perspective. And you know, that's what we use.

Jeff  34:47

That's great. And it's fine. As soon as you said that, it reminds me of a large customer that I was working with, and they had the whole design system and in we were brought in for a UX and front end thing and we were just told to Just map up to the design system with creative, or UX, excuse me. And that's exactly what we did. It seems like it's a shared language that people have the professional seem to be able to kind of map up to. So that was great. It is good.

Shmuel Bollen  35:15

And I think it's also easy for development teams to learn enough of that language, so that you're not, you're not imposing yourself on them. You're actually you're actually teaching them and I've got development teams, one actually in Cape Town, South Africa, that I give a lot of autonomy to, because they understand the design system, they know how to use it, they go ahead and use it, we've got a bit of a time zone thing, you know, and they'll share with me, okay, take a look at what we've done. Please, please tell us if it's in line with, with your vision for the implementation of the design system, because you could you could implement the design system in a clumsy way for sure. So that's been amazing. That's that's basically extended the reach of my team, without me adding headcount. So

Jeff  35:57

awesome. That's amazing. I mean, because you want to you don't want to put tools in that's going to slow the dev team down and things like that. So as long as you give them the guardrails, then they're able to produce, that's great. Listen, I just looked at the time. And I know that we're both a hidden mute when the kids run mind. So just wanted to wrap up with sort of like, what's the big thing that you're passionate about that you're working on? Now? You know, for the futures, it's obviously you've got the work and you're teaching are the elements in both of those?

Shmuel Bollen  36:31

Well, I think that the experience that I plan on on designing for our users, is enabling them to almost fuse with the software, and use the software to augment their own understanding of what is going on in front of them. And the example I use this is merchant fraud. Right. So I support dozens of wealthy many dozens of companies that are their retailers, either ecommerce or brick and mortar retailers that have it could be could be 100, or 150. Analysts sitting at any given time looking at software that is designed to try to detect fraud within transactions. And I can only take it so far before, there's too much variety in the way they think about fraud. So they've still got to do some work to take the information I presented them with, even though it's presented in a way that they understand. And it's been tailored and so forth, they still gotta process it further, I want to give them tools, I want to give them tools, so that each individual analyst can say, I'm detecting a trend, I'm detecting that when these three or four characteristics are present within a transaction, that I believe it's this kind of fraud, I want to let them write a rule on the fly, so that whenever a transaction comes in that analysis that they would normally do manually each time, that's already happened, because they told the transaction to do that. And it's, they've been able to create a flag a visual flag that they see it in, you know, probably 200 milliseconds, they already know what to do next, instead of having to take five or 10 seconds, and do that process each and every time. So that's what I'm really that's to me, that's, that's a user driven experience. And that's just one example of many that I have rattling around in my brain, you know, that I plan to implement, you know, within the next few years.

Jeff  38:29

I think that sounds to me, like the logical step after that is these types of tasks that are, you know, able to be picked out by AI instead of a person, and then you know, those types of things. So, but that being able to, you know, still bubble up that information to the next to the next level higher. So, yeah, these years, I don't think we can go on forever. But that sounds fascinating. Well, listen, thanks so much. We don't get enough UX people on here. And the last time we did it was really in the context of like, getting hired to do UX for a customer project. So this was a really great sort of heuristic or overall look at sort of the design process, which I learned a ton. So I'll say we can hang on for a minute after this. But thanks so much for for joining and looking forward to putting the links out there and getting some of the feedback.

Shmuel Bollen  39:24

Yeah, thank you very much for having me. I would love to talk about this stuff. Ya

Jeff  39:27

know, it's that's it that's what our friend said, like, you guys could chat for hours about this stuff. So Exactly. Right. So thanks a ton and hang on for a quick second. Sure.