Photo of Donna Spencer

Where did IA go? with Donna Spencer

March 4, 2021 by UX-RADIO

In this episode, we talk to Donna Spencer to answer the question, “Where did IA (Information Architecture) go?” We talk about defining IA, is IA still relevant, the importance of understanding and representing the abstraction layer, and how to present design.

Donna is an independent design consultant. With more than 20 years of experience, she has expertise across the entire design spectrum, from strategy to delivery and everything in between. and she loves it all. She is designed for a wide range of problems across all kinds of industries, with a particular emphasis on information architecture, and complex employee experience problems. Recognized internationally as a leading UX practitioner, Donna is a regular conference speaker and has written five UX-related books. She also created UX Australia.

Show Notes

  • 7:15 What is Information Architecture? Organizing and structuring content and its relationships. There is a large amount of information where we need to understand the relationships, why they are there, how to group things, and model how those relationships work. Then you do the navigation design.
  • 9:00 Where has IA gone? If it went away would it matter? 
  • 12:35 People are over-designing individual screens or pages instead of creating a sitemap and page types.
  • 14:25 The IA outputs are not always obvious to stakeholders in how they add value to the project. 
  • 18:00 Help people think conceptually when sharing the IA 
  • 21:51 Donna’s article on information-seeking behaviors and how to solve for them
  • 29:10 Google broke valuable search results 
  • 31:50 Figure out what we don’t know – Known item search  is when someone needs help finding what they know versus someone who doesn’t know what they are looking for
  • 34:00 Where is IA going next?
  • 38:22 What is your advice for design leaders to better incorporate IA into their process? Help teams understand the abstraction layer and models and representations of things rather than starting to work on design artifacts.
  • 41:00 What would you like your legacy to be? Creating practical resources and I hope that good work doesn’t get ignored just because it feels old. It’s evergreen.  My forward legacy is to continue to help people understand the abstract level of work and build modeling into projects to ultimately build things better. And mostly being nice to people!



And I’m Chris Chandler. And today we are super excited to bring you an interview with one of our favorite people in the world and noted Information Architect Donna Spencer,


thank you for having me, you have to have to have the hang on how do I do this? You two are two of my favorite people in the world as well.


It’s so great to have you here today. Donna, you have such an amazing history and background and experience. Can you share with the audience a little bit about how you got started?


Oh, so I got started in the UX and information architecture world, I was working in the Australian public service and I was working for our statistical agency. And I was actually responsible for publishing some of the survey outputs. And our web team needed a hand because they were putting all of the statistics on the internet, which meant that that website went from, I don’t know, 30,000 pages to a million. And they just needed help. So they pulled me in to help. And this was the very late 90s. And I’d been pulled in to help. I, not only did a lot of migration but a lot of structuring. When you put that amount of content on the internet without a lot of thought, it doesn’t become very usable. So that was where I first did my first usability test, and then spent probably a year or so with them, seeing how we could help people to find that statistical information when they didn’t already know what they were looking for. So a lot of people, a lot of their audience already knew exactly what they wanted every month, but a lot of people didn’t. So we had to figure out how to balance that, that need of expert users who know the catalog and know the numbering system and know the jargon. And people who want to know how many people there are in Australia. And don’t know where that comes from. And, the really good thing about studying user experience and II work then was there was were those resources like we collected, we hadn’t really invented it yet. Usability testing was a known thing, but the rest of it really wasn’t. And so it was great timing to learn something and figure out things from scratch, and builds that body of knowledge collectively, which is entirely different to now where if you want to learn a bit about information architecture, you don’t throw yourself at a million-page website and figure it out. You pick up a book, which is great. But it’s an entirely different way of learning and like, really actively getting things into your head. So I’m glad about that timing. I’m glad that this whole field didn’t exist. And glad that I was around while we collectively invented all the stuff that now does exist for people.


You’re so cute. Do you think people pick up books? Don’t they just watch YouTube videos?


Yes, they do. And people buy books. I mean, I sometimes get tiny amounts of money into my bank from people buying my books.


Well, speaking of that, that perfect segue, I was gonna say, to join that, like you wrote one of the most important early books on an absolute foundational method on card sorting. Yeah, and I, I mean, I, you know, I don’t know anybody who’s done any amount of card sorting who isn’t in love with it as a method. And I mean, I feel like I probably overuse it and over-rely on it myself. I’m curious something about the experience and sort of how often do you find yourself doing card sorts these days?


I haven’t done a card swap for a long time. But it also because I mean, lots of reasons I, we’ve probably had this, we probably talked about this at the time. The reason I wrote that book and the reason that Lou as a publisher approached me to write it was that I was negative about canceling. Like, I was like, This is not a good technique. Why on earth are we just using it, to create our information architectures, because people were using it as I’m going to do a card source. And out of that, there is the I and I, I talked to people at the time who were doing CAD taught, their users would kind of make 14 categories and they go, Well, that’s 14 categories we make and it’s like, no, put your brain into this as well. combine it with other things that you need to know. And then come up with information architecture. So one of the reasons I don’t use it as a technique a lot is because my experience with information architecture and kind of understanding concepts and being good at figuring out what goes together means that I am more likely to put something together and then test it, then necessarily involve users in doing those, like doing those groupings, I find is not universal, that I can learn just as much about how people think about concepts, and what goes together by talking to them understanding them. So having done it for a long time, but also I had a bunch of years where I didn’t really do any work because I was doing other things like running conferences and, building design names. And other things that didn’t involve quite so much. I’ve got a project now that I think we’ll do one later in the year because it’s a situation where I probably can’t just take a guess at it where I do need to ask people more deliberately, why things go together.


It’s interesting when you’re talking about card sorting, that they would come up with categories and then say, Well, this is IA, and of course, at the IA summit, and many other places we have discussed defining the thing. So I just have to ask, how would you define information architecture?


So I think information architecture is all about organizing, and structuring content and its relationships. Then, in our defining the conversations we used to get tangled up, and people still do this with their roles get tangled up with what is the core concept here? And what are the things that I do when I have a job title of a theme. So as somebody with a job title of information architect, I don’t think that exists anymore, except for me and a couple of people do other things. We do research, we make prototypes, we work with different kinds of people, we do screen design, etc, etc. There’s lots of stuff that we do. Not all of that is information architecture. So the concept, I think of information architecture, as an idea is you have some usually large amount of stuff. And do you need to understand the relationships within it and why those relationships are there? You need to figure out how to how you’re going to group and present things. And you often need to kind of make, like, model out how those relationships work. And then that often turns into something that we present to the public. So then you might do navigation design, but navigation design isn’t necessarily only it as well.


Yeah, I think that makes it I mean, makes a lot of sense. We all met at the IAA conference, yeah, in the early part of the century, and watched as the term UX sort of took over. Right? That Yeah, that where we’re now information architecture is seen as a subset. Right is a lot of the time for you x. And I want to get to sort of the question you posed when we were talking about having you on the podcast and I really interesting discussion, which is, where has the IAA gone? As you said, I do know, I know one person, maybe two people in my LinkedIn network who have the title of an information architect. One of them who works here in LA for a big financial services company writes a lot of time working on metadata and organizing the content for their sites. But I do think the issue of why we don’t see that as a job role as a specific discipline, I mean, if you even touched on it, Write that I feel like UX research has emerged as, like a solid discipline, get by information architecture has pretty much gone away. As it is. Yeah. and of itself.


Yeah. And then and that’s why I thought it would be fun to talk about because sometimes I think, Okay, well, how would we bring it back? And then I’m like, well if it went away, didn’t matter, what we making it up when we said that this was like, you know, the information architecture of a set of content was fundamental. And it was the blueprint of the service. And like what we just alluded to? How could it be? How can it be that something that that we kept, we kept saying was super important, just seems to have gone poof. And so I don’t see, like, if I see a job advertisement for something like a UX or service designer, really would say knowledge and information architecture is important. If I see people talking about their processes for going through a project, rarely Will I see a stage where there is some focus on the structure of the content. So we don’t even need to necessarily hire people with information, I could tip two titles, but the concept seems to have disappeared. It doesn’t get covered in things like boot camp programs, or might get covered like super briefly. And I talk with people, designers coming out of boot camps, and they’re like, Oh, no, I didn’t know anything about IAA, we covered it in the kind of half a day. How are they designing anything? Like how I didn’t understand how somebody goes into a project and doesn’t think about the content, its relationships, and structures? Are they just doing teeny weeny like teeny, weeny projects with teeny, weeny content sources? Or they like messing it up? I certainly see a lot of projects that seem to be project seems to be messed up, because nobody thought about, like structure navigation relationships at the beginning, I certainly see people trying to swap things onto a page without understanding how one page might relate to another. So I’m, like, I’m talking myself in a circle young guy doesn’t really matter. Of course, because it matters. It does. Of course, it matters, people are just designing really, really. So a lot of what I see happening is people over designing individual screens, so producing volumes and volumes of high fidelity individual screens for for for something like an application or a site, when actually what would be useful, more useful, would be having a really good, like sitemap structure model and then saying, okay, we, we’ve got five-page types here, we design the five-page types, not the 70 pages, but I’m seeing a lot of just like, designers doing 70 pages, and then hoping that they will link to each other somehow, that just takes a lot of time. And I have seen this happen with a couple of teams I’ve worked with, and I’m like, Oh, you’re so slow, and then realize they’re so slow, because they’re designing individual pages. But you just do a model for those pages, and you make some templates, and then you represent them and you write some annotation saying this one’s slightly different, because but this is how it seems to be happening everywhere I see it’s almost like we get up, I’m trying to organize the content.


Yeah, we can’t give up we can, it’s too important. You have to have that holistic, systematic understanding of what this is how the pieces fit together, how they make sense. And I think we absolutely can’t let it go.


Okay, so if you can’t let it go, and I’m completely with you, because I do not understand how anybody could design something larger than about five pages without abstracting it without thinking of like, what is what is here? What’s the model? What are the kind of main styles of the pieces? What are the main topics we’re dealing with? I don’t know how you design something large without that abstraction layer. So of course, we can’t let it go. But I think I have I read some, some theories, some ideas,


I thought about it. So So that was my first one, which is we just gave up because it’s hard. But I think one idea is that, right? Like the outputs of IAA are not tangible, are not always obvious to the stakeholders, like how they’re adding value to the project. So in a way, I could say it’s, it’s sort of like it’s analogous to me to the like, Why don’t people spend time in the problem space of research in the solution space? Yeah, and or, or, more specifically, right, How come I’ve I’ve probably pitched ethnographic research 100 times and And had one client decide to do it. And so there’s like a tenuous relationship, I think between or understanding between the value of the work and how it relates to everything else. Yeah, and maybe another way of saying that, right is I could say to a team, right? Like, hey, go ahead and start designing pages yet. And within a couple of weeks, we’d have some pages designed. And if I say to you, okay, we’re gonna do the site. So first thing we need to do is work on the conceptual understanding and do some IIA, right, that is the other leaders of the stream, what they’re not, they’re not sure, what am I going to get? Yeah, what do I get?


yet? Right? And like, I have theories on, like, Why all this has happened? And where, like, why? Why did we do a thing that we thought was really important? And now that thing’s gone. And suddenly, I think one of one of the reasons it’s disappeared is because of agile, because exactly what you said that focus on tangibility, and out because you can, you can write into a two-week, sprint planning session to create a draft up some wireframes. And at the end of it, somebody who has been given that task can say I did this thing, here it is look at these, and then you can talk about them and figure out your next sprint tasks. It is less common to say, Hmm, let’s actually spend what might be more than like, a two week or one week sprint, really understanding what we’re working with, and doing kind of modeling it out and doing that, doing that conceptual thought, because it doesn’t look like very much. So yeah, all the time that I mean, and this isn’t necessarily agile in itself is the problem, by the way, that agile has been implemented by a lot of people in organizations to focus on outputs, and to focus on being able to identify tasks that can be done in a really like, small timeframe. So you go, Okay, what am I tasks, this sprint, we have to, we have to knock down everything. So it’s under half a day. And I kind of need like four or five days just to think about this, to like, dive in deep into the content and to understand it and to swim around in it and figure out like what other relationships, but you can’t write that on a ticket that has a, you know, story size of medium. So think it’s a Why doesn’t mean that again, it’s this, the hard thinking is unimportant. But I can see why teams have lost some of that focus on planning.


Well, to your point, Chris, around the tangibility of what you’re providing to the client, like, I have been in numerous scenarios, where we go over the sitemap first with the quizzical brow of the client, and then when we show them the prototype, they’re like, Yes, I get it now. Yeah. And so it does seem like it’s a hard thing for them to grasp and to understand.


And that just means that we need to be better at helping people think conceptually when they do need to think conceptually. Because if you can’t help people think conceptually, they don’t know what to think about. And you’re showing them a sitemap, and they’re like, Oh, yeah, I can see some boxes with words in them. But they don’t know how that’s going to translate into anything, they will make decisions that they have to get unmade later on when they finally understand it. And certainly, that is something that, oh, it’s hard to get people to think conceptually when they want to see the screens. But if you make the screens first, and then try to kind of shoehorn some structure into them, the project won’t work. The client might be like, yeah, I get it, there’s a navigation bar. But they still don’t, they can’t see the relationship and the complexities under the surface, they still might need to know about, because it’s going to, later on, cause either, like difficulties in managing the content, difficulties in working with it, or difficulties in actually getting the project built. Because like that’s where things can fall apart. If you haven’t done your planning up front, as you go to try to, like actually do a technical build, which needs structure. It needs structure and rules around it, not just a pile of screens with a navigation bar. And so it can trip over in that part of the project if you haven’t thought about it at the beginning.


Oh, that happens all the time. Yeah, obviously, we’ve all seen that all the time. And I do I mean, I think that again, it’s hard to quantify that value in advance and explain. I mean, I think we touched on a couple of things. They’re always the idea of thinking about the website or your app. abstractly, right. That’s an obvious one where we professionals, right, we’re used to, you know, sort of like manage to keep all the different layers. I go back to that Jesse James Garrett diagram right? Like we can see like six different abstract pieces and kind of squint at it and understand how that will eventually manifest as an experience. And I definitely think a lot of the stakeholders don’t have that experience and so they have a hard time valuing it. I haven’t another maybe. I’m curious what you think. Do you think some has of this work gone to the content strategists?


Definitely, some of this work has gone to the content strategist, definitely like because they’re the people wrangling large amounts of content. So definitely someone that works on there. And definitely, they talk about information architecture. When I look at the way that that field writes and talks about their work, I don’t always see a lot of kind of structural thought, I see a lot around like, you know, brand and maintenance and content planning for the future. And I don’t want to say it doesn’t get done, because like, I definitely some of the work is happening in there.


Well, or the sort of micro-content, right is the like, hey, how can I plan to put pieces of content across different levels? Right, and, and, and different experiences, but I, but I do agree that that doesn’t seem to be an emphasis or a lot of work on that high level, the conceptual relationships.


I’m curious, Donna, how do you approach that?


How do I approach it on projects? Yes. I approach it in the same way that I did 15 to 20 years ago, I do the work in the same kind of way, I still say no, we’re going to, I’m going to spend some time understanding this modeling the relationships, making sure that we understand what goes with what and why and how this is all hooked up. And I also put careful time into making sure that people who need to make decisions around it, actually understand it. So not just say, here’s a sitemap, we’re going to do the next thing, but making sure that decision-makers really understand what the consequences are. And so on a project, I’m working on at the moment, which has a big complex classification that underpins it. It’s actually taken, I’ve needed to spend quite a lot of effort and time just helping people understand both how important this thing is, and what they need to think about before they make decisions on what goes together and why. So not only, like, what are the decisions, but how do you need to think about this, this is homework for meetings, read these things and set aside time in your calendar to do this before you walk into a meeting because you can’t, you won’t be able to absorb it and approve it at the time. And so do things like that so that we make good kind of structural decisions early on that we won’t have to try to undo later on. When we start getting into complex work. Look, I do I work on complex projects as well. Not everybody does this not everybody needs something like that level of difficulty to be dealt with. If we’re working on things that are a lot smaller than those, you know that those structural decisions aren’t as critical. I don’t know if that answered your question.


Yes. Well, I had a follow-up question, which is, you said something, it’s like the homework before the meetings. What are you talking about? Hey, I’ve produced a document that explains to you the conceptual or are you saying, hey, go read this article I wrote 10 years ago, now online?


Oh, actually, occasionally I do that. I tell you the time when I really say go read this article I wrote a really long time ago, is I wrote an article called information-seeking behaviors and how to design for them. And in that I explained, like known item, information seeking and, exploratory information seeking. I also talked about refundability. And when you don’t know what you need to know, but really, a lot of times when people say things like why do we need to do any of this? Can’t we just use search? Or why do we actually need to give people a way of browsing through this information, I will send them to that because it’s a really good resource for the difference between, I already know exactly what I need, and I can dive in and grab the thing I need. Because I already know it and I know where to go. I don’t know what to call it. And I’ve got to figure out this thing that I don’t know anything about and, and I might even my starter knowledge isn’t enough to give me some terminology or anything to go with. So I pointed out things but usually on homework for meetings, or for important decisions. I will usually like right up For people, you know, write and illustrate and do diagrams, so that and look, I’ve just written a book on presenting design work. So all of this goes together, helping people get from where they are to where they need to be to, like, approve something or make a decision. That’s actually the designer’s work to do the education work to help them understand what matters and what to think about and how to think about it. So that when people go into a situation where they need to give feedback, or actually make a decision, that they’re making the decision on, with all the resources that they need. And this, I think, is one of the things that derail projects, and that designers don’t always do well, is bring their clients along on the journey in such a way that when decisions get made, they get made with really good understanding and knowledge and all the background that they need. So that a good decision gets made, and it doesn’t get have to get reversed later on, because people all of a sudden go, Oh, is that what that means?

Book cover image of Presenting Design by Donna Spencer


That sounds like an amazing book.


It’s really super skinny. It’s, it’s like it’s 1000 words, it’s published by a book apart, you can read it in, in the nonexistent plane rides that we don’t do anymore. You can read it on your couch.


Yeah, I’m wondering, like, how can we be better advocates of building that into the Agile process.


The thing, The curious thing about like, a lot of this, and again, something I’ve seen in working with, like growing designers, is if you don’t understand it yourself, and you don’t really know what needs to be done, because you haven’t experienced or you haven’t worked on something this hard, or you haven’t had success, like a successful version of really anything in the past, like if you haven’t, if you haven’t run a successful ethnographic study, and it’s going to be hard to convince anybody, then this graphic study is an important thing to do. If you haven’t run a successful complex piece of information architecture work, then convincing people early on in an agile process isn’t going to work. So we need to understand our work well like we need to understand what needs to be done, what we’re good at what our like what our scope even is, and I think I see a lot of people who don’t know how to communicate their value and what they can do. So there’s a chicken and egg with that you need the experience in actually, you know, understanding the work so that you can explain the work. And I suspect that that chicken and egg comes by doing like small pieces, and growing into bigger, but not saying, or I’ve heard that there’s this concept of this thing that we need to do. I think we’re going to need a chunk of time at the beginning, but I can’t really explain it very well, because I actually don’t understand it myself.


That’s another part of the chicken and the egg problem, right? Since we’re not ta well, right? We don’t Yeah,


yeah, we’re not teaching it well, then people also don’t understand how they would be able to tackle it. Because that, you know, people can figure out the work, that we’ve got resources, and they can read books, articles, YouTube videos, on how to actually do a thing, right. But if you don’t understand the thing before you do it, it’s hard to say, you know, in a planning session, I need this amount of time to do this thing. And it’s going to look like this is the way I’m going to do it. This is what’s going to come out of it. You don’t have that experience. It is hard to do that. And that’s why when I when people ask me how I do it, I’m like, Well, I, I know how to do this. So I know, for an individual project, what’s needed, because I have the experience. I’m like, Okay, this one’s really complex. These are going to be the things we’re going to trip over. Let’s plan not to trip over them. But it’s experience. It gives me that snow kind of abstract set of rules. But yet, if we’re not teaching anybody, there is another thing that’s happening, which is the recency kind of effect of resources. And Google has broken these, we collectively and those of us who have been in our field for a long time, built a really good resource base. There are very good books, and there are very good articles. But nobody falls over them anymore. And they don’t return in search. I went looking for the other day, some articles, I was just searching for some stuff on FAQs. I was probably trying to just shortcut a thing and send a client something on my FAQ site. And I know that I’ve written stuff, and I know that other people have written stuff about FAQs and why we shouldn’t use them because they’re really not organizing information. They’re like they’re anti IA, and that’s usually my perspective and I was trying to find the article, and when I went searching All I was finding were recent articles that clearly have been written only for search engine optimization and boring word search. And they are poor quality. But they recent so they appear. So we know that there’s a bunch of resources, we know there are books, we know that a lot of this has been done. But people are falling over it to learn it to then even do it. So not even, like even if even though this isn’t this topic isn’t necessarily taught officially in boot camps, people also unnecessarily stumbling across it. Because the internet is awash with recent poor quality content. A lot of it written by those poor people in boot camps, because they’ve been told to write some medium articles as part of their assignments. And so the internet’s full of people who have written articles on things they didn’t know about, as assignment topics. And that’s now our resource base.


That’s incredible. I was, I mean, what an insight on I was gonna say, I was gonna mention Google before as one of the reasons why IA isn’t as popular. And you alluded to it too, right? Why can’t we just use search? So I do think, right, the idea of like a search engine that actually works, which Google, the old Google that credit, right, certainly changed. I mean, I’m one of the people, right, I have not created a subfolder in my email. in over a decade. I literally do no organizing of email, right, because I just assume I’m going to use Gmail, and I’m going to search and find it.


But that’s known knowns, known items. Such maybe I know, I’ve got a theme there that’s not that exploratory. I need to figure out something I don’t know. I need to figure out what, what am I meant to do on this project? You can’t just search for that.


Thank you for bringing up known item searching again, because I think if there was one concept that I probably have said, right and explained to clients more often over the last decade, that’s that is pure information architecture is that one, right? That simple difference between someone who knows what they’re looking for, and just needs help finding it? versus someone who doesn’t know what they’re looking for? Like, that’s probably the most powerful explanation of like, now you get it, right, of like, what was the differences and why that’s important. And again, I think we, we all say, write a legacy, people are surprised to know about the legacy of library science, in the field of UX writing, and always surprised to learn that like, Oh, hey, there were librarians who are dealing with this exact problem. How do I find something either when I know, you know, I know what I’m looking for. And you as the library know that you have it? So how do I put those two things together, and the non-exploratory search, but also to layer in the fact that we’ve now also got 15 years of people sort of scamming the Google algorithm, and all the learning that’s gone into, right, how do I do keywords and push my topics? And I mean, there’s just basically it’s a joke in the sort of, like, cooking recipe world.


So every recipe now has got, you know, it’s like, clickbait headline, right, like some pithy thing. Eight screens of story, ya know?


And then, where is the actual recipe?


How long do I cook these chickpeas? I can’t remember if it’s half an hour or an hour and a half. Yes. And you have to, it’s right in the bottom of the thing somewhere. Yeah. That was what I had to do two days ago, though. Not that great, because you know, that the rest of you that I found, told me to cook them for an hour and a half, and luckily, I tasted them after 40 minutes. And I’m like, oh, they’re done. Yeah, those rates of those recipes are Yes, yes.


Where do you think AI is going next?


I don’t know. I don’t know. And that’s what I’ve kind of been using. This is why I kept thinking doesn’t matter. Like if it went away, it didn’t matter that it went away, because how would we bring it back. And if we’ve got a large pool of designers who don’t know, really anything about, like the planning and structuring of content, and they’re the ones who are then describing to their clients, how a project will run, and they’re the ones who then get, you know, promoted into more senior positions and describe how the projects will run. And then the way projects run is by people making lots of lots and lots and lots of screens. It’s fine UX is really slow and expensive now, compared to being able to kind of make a model and some templates It’s just it’s getting big and slow and expensive. And I think I think all of that could be like it could be more efficient. With a bit of a thrown-in. I noticed that, Lou Rosenfeld, I don’t know if it’s part of one of these conferences, somebody’s doing a workshop on, like designing with like pieces and templates, rather than designing whole screens. And the premise of this workshop, like, when I read it, I was like, Oh, yeah, that’s exactly what I’ve been thinking. Like, everything’s kind of blown out and got inefficient. But if you can, you know, if you can abstract, you can be more efficient in your work, I just don’t know, I just don’t know where, where, where this is all going to end up. And if it matters, many people are happily throwing that kind of money at designers on projects, to create lots of screens that I’ve funded it, there may be my wish that, you know, things would be more efficient is irrelevant. I certainly know where I’m going. And that is more working with kind of folks in data science to merge manual kind of classification methods, and, you know, thoughts around how, you know, categories, concepts, structures work. With automated approaches, that’s certainly where I’ve been focusing, and what I’ve been working on, I think I’ll will continue, you know, individually, that will be kind of my focus on work, I like it, it works. Well, I’m good at the data science stuff, because I am good at math and things, which is not necessarily what most information architecture kind of people would be like. So I’m just kind of doing my thing that works for me. And, and, you know, thinking and talking,


you hit on another interesting angle, which I, which you brought up for me there, which is I think that structuring the concepts is often a larger problem than the online experience that you’re trying to build. Yeah, in that way, right? Like we all know. And it’s, it’s talked about a lot the way that design and product design right can be sort of subversive in an organization, and kind of bring things up that they’re in, or being user-centered, right can sort of like customer-centered can like, really shine a light on things internally that people maybe don’t want to talk about. And I think that that is like the what makes the AI hard is exactly that right to get that. It’s one thing to create a system of relationships between concepts. And, and we focus a lot in this conversation about how that relates into a design process for an experience that a customer is going to have. But it also directly relates back up to our technical partners. And where’s that data coming from? And how is it structured? And how is it merged and managed? And I think that that is something that completely just escapes the sort of like, understanding of roll, right, like, Hey, I’m just trying to figure out how to get customers to the right product detail page on my website. What do you mean, I have to have a data governance commission? That is, you know, managing an entire data team on the back end, right, like people have a hard time putting those things together. Yeah.


So Donna, in thinking about design leaders, what advice would you give them to incorporate AI into their process?


What the advice I would give them is to I don’t know how I don’t quite know how they, I mean, again, people are varied and difficult, different experiences. And something that design leaders can and should be looking at is how they can help their teams understand that abstraction layer, and how can they help their teams understand and work with models, and templates and, representations of things, rather than working directly on artifacts. So as a design leader, or a team leader, somebody coaching designers, helping them understand that if I design a, you know, a template for a thing, then that can represent a lot of things that they don’t need to then like draw out every single thing, which also then means that designers need to be able to understand that value of abstraction and modeling as well. And if there are super good resources around for like, where you would kind of learn about that necessarily, I think even our all of our resources are still fairly practical as well. Except for books, of course, but even they’re fairly practical. I mean, they’re practical. They like to do the steam. They’re not necessarily Oh, I know the one that really is useful at teaching abstraction and that’s Steven Anderson and Karl Fast’s book “Figure it Out; Getting from Information to Understanding”. It’s about understanding and modeling and thinking. Yeah, look, it’s a hard question because I’m still musing on what the problem is. So I don’t know what the solution is. And I don’t know what the problem is even a problem, or if it’s just something that bothers me.


I think you’re right. I mean, I mean, I love what you said there. I mean, I do think, right that like, there’s so much emphasis in the sort of design ops world, right, like, what is the design system? And again, it’s very artifact-focused, right? It’s all about workflow and pipeline, and how do I make sure that the different people have access to the right thing, and we’re not reinventing the wheel. But what you’re talking about is in order to really make that work, and it’s like anything else, people need to have a conceptual understanding of why they’re doing it that way. And how those pieces relate to each other. There’s not that’s at a level up, that is abstracted that isn’t about the model. And I think that is extremely good advice to leaders, right, which is you need to share, and figure out how to deepen the conceptual understanding of why you’re doing that how the pieces relate to each other at an abstract level, not just the, hey, how does somebody updated design component in react? To ensure that we always have the right, you know, the font on the page?


Now let me ask you another question. Which is, what would you like your legacy to be in our field, Donna?


Look, I’ve already built some chunk of my legacy. And that is producing very practical resources, books, articles, and all the workshops that I have taught over time at things like conferences, I’ve done a chunk of that legacy already in teaching people the things that I think matter around this topic. So I built some lit some metal legacy already, I hope. I hope that that good work, doesn’t get disappeared and ignored, just because it feels like it’s old. Because a lot of you know, a lot of the things that I’ve done and work I’ve contributed is evergreen. So I hope it does, I hope it doesn’t disappear. I’m not up for kind of rewriting a book, just because it has a date on the cover that was a while ago, it’s rewriting a book is a lot of effort. And my forward legacy, I hope to continue to help people understand how to kind of do this more abstract level of work, and how to build like that kind of modeling and abstraction and thinking into our projects so that we ultimately, you know, build things better and more efficiently effectively. And, and, and I hope my legacy is also that I was mostly nice to people.


Yes. It’s been wonderful having you on the show. Thank you so much, Donna.


Thank you both.

UX-radio is produced by Lara Fedoroff and Chris Chandler. If you want more UX radio, you can subscribe to our podcast on iTunes and Google Play or go to, where you’ll find podcasts, resources, and more.

This episode is brought to you by Philosophie. Philosophie helps entrepreneurs and organizations validate and develop their promising ideas through agile design, rapid prototyping, and software craftsmanship. To learn more, visit

More About Donna

Donna is an independent design consultant. With 20-ish years of experience, she has expertise across the entire design spectrum – from strategy to delivery and everything in between – and loves all of it. She has designed for a wide range of problems across all kinds of industries, with a particular emphasis on information architecture and complex employee experience problems. Recognized internationally as a leading UX practitioner, Donna is a regular conference speaker and has written five UX-related books. She created UX Australia and ran it for 9 years.