Bram Wessel and Gary Carlson are the founders of Factor, an information architecture and experience design consultancy. In this conversation, we discuss their practice of helping organizations gain alignment by defining the information layer that underpins their digital systems.

Show notes

Show notes include Amazon affiliate links. We get a small commission for purchases made through these links.

Transcript

Jorge: Bram, Gary, welcome to the show.

Bram: Thank you, Jorge. It’s great to be here.

Jorge: Well, it’s a pleasure to have both of you here. We were talking earlier that it’s actually the first time that I’ve interviewed two folks on the show. But I know you; I’ve known you all for a while. We’ve had the privilege of… well, I’ve had the privilege of collaborating with you in the past.

Bram: The privilege was mutual!

Jorge: Some folks tuning in might not know who you are. Do you mind please introducing yourselves?

About Bram and Gary

Bram: Sure! So My name is Bram Wessel. I’ve been in the field since the late nineties. I started out helping organizations, mostly with supporting users learning digital products like Photoshop and Illustrator. And then I moved on to design. And I’ve been involved in digital design in various capacities since then.

Once I started to get serious about forming my own practice, I started running into a challenge, which was that the information component of the projects, which were mostly UX strategy projects I was involved with, was often neglected or underserved. And my background had come out of the information architecture world, so I could recognize this, and a lot of the design agency folks that I was working with would not be able to answer the question: “where’s the data going to come from, from this service that we’re writing an experience for?” And so, one day, an old college friend called me up, and that was Gary.

Gary: And my background is similar to Bram’s in that I have been working in the space of information management for a long time. My focus has always been on taxonomies and metadata and what it means to make them work for an enterprise as opposed to just one system or one experience. So really taking this holistic view of what they are and really trying to understand how they can meet the needs of a wide range of people.

So as we’ve been doing this work, and when I went out on my own, It became really clear that companies needed to work on these information structures — these taxonomies, media models, and ontologies. But every time I asked them how they were going to get used or what the end users were expecting, or what the experiences were going to be like, no one had any really good information for me.

It’s really difficult to create a taxonomy if you don’t know who the audience is. They really need to be targeted toward end use cases and with a good understanding of what those people’s goals are, what they need to do, how often they’re going to get updated and changed, and stuff like that. So early on, after I’d gone out on my own, I reached out to Bram, and we immediately started bringing each other into each other’s projects, with him being able to provide me with a lot of those insights as to how end users were going to interact with this work and me being able to help fill that gap in terms of, well, if you want to build this experience, this experience is dependent upon this type of information foundation, in order to truly support it.

And after we pulled each other into each other’s projects for a year or two, we then started Factor and built this practice around the idea that information needs to be treated as a corporate asset. It needs to meet business goals. It fundamentally needs to support user goals, and it also needs to live in these really complex environments with multiple systems and governance processes and perhaps regulatory environments and stuff like that. So that’s really what our focus has been at Factor, and it’s a reflection of both of our backgrounds.

Jorge: And correct me if I’m wrong, but Factor either turned ten or is turning ten this year. Is that right?

Bram: Factor is turning 10 on February 15th, 2023.

Jorge: Wow, that’s amazing. Congratulations!

Bram: Thank you! It’s been a lot of fun.

Jorge: You used the phrase “information component,” and you said that they were neglected and underserved. And I think I heard you, Gary, say something similar where like, there’s this lack of awareness of the information involved. But, I think that when I hear “information component,” it feels a little vague to me; a little ambiguous. What do you mean by information component?

Information components

Bram: Yeah. I was being asked to design experiences that were information dense. And so I was an interaction designer and a user researcher, and I would go through what in the mid-two-thousands or late-two-thousands was a typical design process, and very quickly start asking questions like, “okay. You’re creating an e-commerce environment to launch a new phone service.” I can’t remember what they were called, but I was working for a company called Amped Mobile, which was an MVNO. That’s what it was! A Mobile Virtual Network Operator.

And there were all of these objects that the experience was designed to help people acquire. So you need to purchase service, you need to add things to the service, you need to maybe purchase digital assets. And my questions were, “Okay, so where is this information coming from? If you want to purchase a service, where are all the attributes about what that service is, coming from?”

And so, I would start asking questions about the technology and the data, and what I found was that I really needed to talk to stakeholders across the organization to be able to get those questions answered. So something as simple as two different service plans. In the experience, it seems like a binary choice, but to be able to make those plans available for purchase and provisioned for the user was this vastly complex information challenge. So, that’s what I mean by information components.

But I think what Gary and I started to realize as we did more and more of this work is that every organization has what we call an information layer. And that is — as we have come to describe it — the layer that is between what we call the technical layer, which is all the technology and the software and the data that underlies a product or a service or an organization’s offering, and the experience layer, which is all of the screens and surfaces that actual users of whatever the product or service or offering is interact with directly. And tying those things together has become Factor’s principal line of business.

Gary: Yeah, I think what we found was that if a company is going to provide an experience that includes products and advice about those products and reviews and availability. All of that information is probably coming from multiple systems. And the information layer is where we rely on an organization to explicitly define what they mean by product and what they mean by support content, and what they mean by all of these different things that people need to interact with to have a seamless experience on their website.

And because that information is coming from many systems across multiple workflows, often across different business units, there needs to be a coherent orchestrated approach to managing those definitions or that, you know… really the creation of that information layer. What is the organizational perspective on the product model? What is the definitive set of geography terms for our organization? What countries do we do business in? Which ones do we not?

By being able to model that outside of any individual system, we’re then able to create this and then have it be available to all the systems that are needed almost as a service. And that’s where we’ve found the organization has been able to make a big step forward in the digital capacity by being able to stitch these systems together, at least at that semantic level, so that they don’t have to do mapping tables between all these systems and that they can say that, you know, this is a “this,” and this exists in these five different systems across these business units so that we’re doing analytics or providing personalization or providing new experiences on our website. We all know how to get that information and we can provide that consistent information across the experience to our users.

Jorge: What I’m hearing there is that most people when they think of software, they interact with an app or a website or whatever. They might think about it in two ways, normally. One is the design of the thing that they’re experiencing, right? And they’ll think of things like, “Well, this is what it looks like. You know, it has a certain layout. It has particular navigation bars,” or whatever, right? And then there are people who might think about it in terms of the code, the technical infrastructure that makes that software possible.

Gary: Right.

Jorge: And what I hear there is that there is something in between those two things, right? So it’s not just about what the thing looks like — what the front end of the experience is like. And it’s not just about the technical underpinnings that make the experience possible. There’s this middle layer that somehow bridges the two. Is that fair?

Information defines organizations

Gary: Yeah, and it not only bridges those two, but it provides an organization the opportunity to really define who they are. Because your product model is a definition of who you are. Your navigation hierarchy; your product categories —those are all an expression of your worldview. And you want that to be consistent across all of your systems. A big part of this as well is making sure that that organizational view is represented in all the systems that are going to provide the information for all those different experiences.

Bram: I think an example might be helpful. If you have an e-commerce organization and you’re really good at describing how your product should be used, the challenge that you have is: how do you connect product information with content about how the product should be used? So, if you’re REI or Backcountry — both of which are customers that we’ve worked with in the past — and you have all this information about your products that is embedded in your merchandising software, all those backend systems that handle your supply chain and where you get your products that you then resell. And then you have your content management system, which really provides the experience that users interact with, and that might be more sophisticated than just a CMS. It might also include a digital asset management system and a product and service catalog piece of software. And those are the pieces of software that directly impact the experience.

Well, the mistake that some organizations made when they first started down the road in e-commerce in the two-thousands was, “Oh! Well, doesn’t it make sense? We have all this product information; let’s just cut and paste that.” And that was really confusing for users, and it was a really poor experience because that information was all structured for the purpose of managing a supply chain or managing a merchandising organization; it was not designed for navigation. So, when we think about the information layer, that’s one of the places where an organization is going to want to manage the core concepts. And that allows them to do things.

So, for example, if you’re really good, like I mentioned at the beginning of this example, at describing how your product should be used, then — in the experience — how are you going to tie non-product content, which is just content about say, skiing or snowboarding, or fly fishing, to product content or to product information if you don’t have an independently managed set of concepts, to be able to link those two things together? You’re going to run into challenges.

Just about every e-commerce company we’ve ever worked with has had this challenge. They tried to cut and paste their merch taxonomy onto their navigation, and things went sideways, and they couldn’t do any kind of non-product marketing of their products in a digital environment because they couldn’t tie the product content to the products themselves.

Gary: And then you can also throw on what you know about each of your customers and how do you have those same concepts reflected in your customer relationship management software so you know what things they might be interested in. So, as you go down this path, there are more and more core business capabilities. Merchandising, your content management, your customer management — all of those things generally live in different systems, and if you’re going to start stitching that content together across those, you want to make sure they’re describing things the same way as much as possible. And that’s what the information layer allows you to do. It allows you to say, “this is what’s important to us, and we’re going to use these terms, these concepts, to describe in our information across all of these systems.”

Alignment and ontology

Jorge: What you’re describing here sounds to me kind of like an ontology. Is that a fair reading?

Gary: It is, although it’s funny. It sounds like it should be an ontology, and in some ways, it is. But when we’re working on these really large projects, just getting companies to agree to a common set of terms for what their products are or a common audience and those things is really the first step. So ultimately, they could greatly benefit from an ontology. But on day one, they just need to get this alignment across the systems and across their workflows.

When you have this much complexity, the first step in this is just getting this alignment across the systems. And then that way you want to have it, you need it to be as simple as possible. Then once you want to build on additional capabilities, and you’ve got that level of maturity so that you can create these concepts and govern and manage them across these systems, then you’re able to add additional complexity to your models themselves.

And that idea of governance is a really important one because when we’re working on our projects, one of the biggest inputs is what’s the organizational capacity for managing these structures? Do they have the technology capabilities and how to propagate these terms — these concepts — across the systems? Are they able to keep them up to date? Is their marketing team disciplined enough to be able to give people a heads up in which direction things are going to be going six months from now so that they can reify those concepts into the models ahead of time? So that’s the understanding that governance process and maintenance needs are a huge part of the models that we create.

Jorge: What I’m hearing there is that while these relationships between concepts might manifest in something like an ontology, the real value, or an important part of the value here, has to do with gaining alignment around what those definitions are. And you said earlier, Gary, that you provide the opportunity for the organization to define who they are. I noted that because it sounds almost a little philosophical, like you’re consulting with them to help them come to understand themselves through this process of articulating information in a structured way. Is that fair?

Gary: Absolutely. You know, your categories — your taxonomies — are your worldview. That’s how you see the world. And the more aligned a company can be across all their business units and systems and stuff like that, the easier it’s going to be for them to interact with the world and gain people’s trust because there will be consistent messaging across all of those things.

Bram: We’ve been talking a lot about the tactical realities of these complex environments that most of our customers have. But there’s something, Jorge, that I think you’re getting at that’s even more fundamental than that, which is you need to define the things that you possess and that you control and that you need to share both within your organization and with the world.

So, let me give you an example. If you’re a marketing organization that spends 2 billion a year on marketing — and we worked with an organization like this — and we can walk around a cafeteria where the whole marketing organization is in the same building. Obviously, we did those walks pre-pandemic, but we’re starting to actually meet with them in person again. We’re going to do another offsite in a couple of months. Or onsite!

You can ask seven different people, “okay! You work in one of the most legendary marketing organizations in the history of mass marketing. Tell me what a campaign is.” Well, you might get seven different answers, and then you might talk to somebody who scratches their head and says, “wow! I work with campaigns all the time, and I can kind of tell you what I think a campaign is, but we don’t have a common definition of ‘campaign’ in this organization that we all can faithfully articulate.”

And that causes a problem when you’re in a complex technical environment, and all marketing organizations at that scale are in a complex technical environment. They have all of these systems, and they have all these processes and different functions and different groups that need to share information about these fundamental concepts. That’s what the information layer really refers to. It’s that abstract notion of… I’ll use the word “semantic.” You know, it’s what do these things actually mean? And then how do their definitions get instantiated into this complex digital environment?

Jorge: And obviously, if group A over here has one understanding of “campaign” and group B over here has a different understanding of “campaign,” it’s going to make communicating between the teams difficult.

Gary: And it made doing analytics on their marketing campaigns almost impossible. Well, it was impossible. And that’s actually how we started there; they were unable to run analytics across their campaigns to see how they were doing while they were in flight. And then it would take three or four weeks after a campaign was over to even get an inkling of how it had performed because even though they had a best-of-breed technology stack that was actually running pretty well, there was nothing stitching those systems together, semantically.

Jorge: Bram, I think that you used the phrase “an independently managed set of concepts.” And I’m wondering about the phrase “independently managed.” The way that I understood that was: there is some kind of system where these conceptual relationships are formally articulated. I don’t know if it was you, Gary, who used the word reified, right? So that implies to me that there is a source-of-truth system. What does that look like?

Bram: I think it’s important to point out that the source of truth really needs to be the concepts themselves. There are systems that are available to… you know, digital products — software — that are available that can manage these really effectively and in a way that is governable and accessible to non-technical users.

But what I mean by “centrally managed” is that we would go into organizations where the product taxonomy would live in six or seven different places. There’d be one version of it in the content management system; there’d be another version of it in the CRM; there’d be another version of it in some form of ERP, which was really fundamental to managing the product hierarchy.

There are organizations that use PIMS — which stands for Product Information Management System — that was set up to be the source of truth for the product information, but yet there were many different versions of that proliferating. So, the notion is that there needs to be something that the organization owns and controls instead of pieces of the organization independently managing their version of it.

To truly provide that as a service to the entire organization, it needs to be centrally managed and governed. And maybe centrally is a little bit more accurate than independently. By “independently,” I just meant independent of a group or a piece of software, or a business function. So I think that clarifies what I provide there.

Jorge: Yeah. And the way I’m hearing that is that the way that something like this would work is that you have all these disparate systems, right? You mentioned CRMs, CRPs… I would expect that content management systems would also be a part of that as well, right?

Gary: Absolutely.

Jorge: And they all have various means of storing taxonomies of terms and/or semantic relationships between terms. And what you would want is some centrally managed system that talks to those different instances and says, “No. This is the official term for this concept.” Right?

Gary: Yeah, and it really depends on the capability of those systems. Ideally — and with our more sophisticated customers — they’re able to actually push these models into those systems. Some systems aren’t quite as far developed; there’s a bit more of a manual process. But at the end of the day, you want complete line of sight between what this core concept is and how it’s represented in these different systems.

Information as an organizational asset

Bram: I think it’s also really important to remember that the reason that the information layer is independent is that… well, let me tell you the history of how we came to think about organizations as having an information layer. In the dawn of the contemporary web or internet, which really started in about 1998 — that’s when content management systems really started to become widely adopted because organizations quickly realized that they needed to manage this in a dynamic and they couldn’t be building individual pages for everything.

And then, for the next really ten years, I think organizations spent a lot of time and energy, and money on the first wave of digital transformation, which is: “let’s implement a CRM. Let’s implement a DAM. Let’s implement a PIM. Let’s implement enterprise search. Let’s implement all of these vast, complex, expensive systems.” And they got to the end of that decade, and they realized that “Oh! Now we just have proliferating notions of everything in all of these different systems.”

And the world changed in 2008. Smartphones became a mass consumer product, and social media started to condition users of information and users of digital experiences to expect information-rich, information-dense experiences. So for the next ten years, the experience layer was the focus.

And I’m sure, Jorge, you remember what it was like in 2008 when the smartphone and direct manipulation and surfaces became widely adopted, that it felt like waking up. “Oh, where has the world been all of our careers? We’re suddenly in the center of the universe as human-centered experience designers and interaction designers.”

And what we think has happened now and maybe been accelerated a little bit by the pandemic is that organizations have come to realize, as digital transformation has become even more of a mandate, that they really need to tie all of this stuff together. And they don’t just need to tie systems together. They need to tie experiences together. They need to have experiences that are coherent and seamless, and that represent the same notions and concepts and objects and things across a really complex firmament. So the notion of the information layer is that you need to think about that as an independent corporate or organizational asset.

I don’t like saying corporate because every organization, whether it’s a business or not, faces this. You need to think of that as an organizational asset and focus on really defining all of those concepts and managing those as assets themselves instead of just as things that live in a system or in a business unit, or in an experience.

Jorge: You said every organization needs this, and in hearing both of you talk about it, I’ve been wondering, is this something that is more applicable to large enterprises just because of the nature of the beast?

Gary: Certainly, the approach we take is geared towards large enterprises because our focus has really been on addressing these really complex problems. Smaller organizations clearly have this issue, but they’re less likely to have a wide range of systems. They can get everybody in a room at the same time to hash this stuff out. So, from an implementation perspective, if they’ve got a good, clear vision of where they want to go as a company, they can start to address this stuff. I’m not saying they all do, but they could because they’ve got… there’s less fuss in the environment.

Every organization has to decide, like, who are they? What makes them different? What is it that they’re truly trying to sell? What services are they truly trying to provide? How do they articulate that to people? There’s no magic in anything I’m saying there. That’s been known for a long time. Companies have to have a vision. And for large organizations, that vision ends up needing to be implemented across multiple systems and across multiple experiences. And that’s why we’re here. And for smaller companies, they absolutely need to do that work, but it’s of a different scale. Although, at the highest level, it’s the same type of thing.

Jorge: Well, and to your point, just because of the nature of the fact that they’re smaller means fewer people, which makes it easier for them to get in alignment, right?

Gary: Right.

Alignment matters

Bram: Alignment is never trivial.

Gary: Yeah.

Bram: And the information layer has really become a key instrument in our success in getting organizations to align. Once they realize that there is this other thing that they need to concentrate on, then suddenly they can have a conversation about that. Whereas before, it was a series of fiefdoms, and there was a lot of, “No, we own this concept. No, we own this concept!”

And that still exists in small organizations, but in large organizations, there is also a very large, vast, and complex heterogeneous digital environment, both inside the organization in what faces their audience or their constituents or their customers. So we focus there because the problems are more of the scale of the kind of solutions that we provide.

Gary: But I think one of the things you said, Jorge, just triggered for me is we really view this as a lens into helping a company or an organization work better. You know, working through the use cases, working through the problems of getting the right information to the right people at the right time, really is a great lens for helping an organization clarify what does it mean to have the right information? And what does it mean to have the right people? And what is it we’re even selling?

So, it can be a really great lens for an organization, a) to get some things done immediately, but also start to coalesce around these ideas of who they are, and you know, we’ve had projects where… We had one project where this huge organization couldn’t agree on what a product was. And we spent three days in a workshop just going through it to help them get to the point of, “okay. This is what a product is, and this is what it isn’t.”

And you know, it’s frightening when you realize the size of companies that can’t tell you what their products are. But that’s really the lens that we bring to this is that information management; that information lens provides organizations to glean those sorts of insights about themselves and how to move forward.

Bram: Just sort of jumping off of what Gary just said, I think it’s really important for organizations to understand that if you don’t know what a product is, or if you don’t know what a campaign is, you can’t do anything. You can do a lot of work, and you can be really busy. But fundamentally, once you define what a product is, then you can have a product model or a product hierarchy. Then you can start implementing that notion of product in all of the different places where it needs to exist. All the experiences, all of the processes, all of the governance regimes, and all of the systems. And the information layer addresses the kernel of that problem of what are the things, who are we, and what are we offering.

And once you start diving into that, suddenly organizations realize, “Wow, we haven’t really thought about this in any way that is going to work in a complex digital environment. This has all evolved organically. We acquired this company and brought in a bunch of their notions of what our product is.” Or, “We implemented this system, and just the architecture of that system itself forced us to change what we thought of as a product, but that doesn’t really align with the notion of product-hood that is fundamental to who we are.” So all of these things force organizations to really think deeply about defining concepts and owning the concepts that are fundamental to their offering, essentially.

Jorge: What it sounds like to me is that as more organizations move to become digital, we are being forced to grapple with the ambiguity of language.

Bram: Yes.

Jorge: And the act of becoming more precise in what we mean when we say the things we say has value that goes beyond building these digital systems.

Bram: Exactly. Yes, it has organizational value, it has business value, it has cultural value, it has brand value, has value in communicating with your constituents, your customers, your audience, and your users. It has value in all of those areas.

Gary: Any company that’s going through some sort of digital transformation or becoming digital and stuff, their products are… people don’t get to pick up and physically hold their products before they buy them. Everything is mediated through information in these digital experiences. Everything! So being able to be as targeted and concise as possible with your information and be able to use it as powerfully as possible is what is necessary for digital transformation. Because, again, everything is going to be mediated through it, and the way we need us to need to be more concise and accurate about the information is only going to help that process.

Bram: Yeah, what Gary said. He used the word “mediated through.” Information is the medium of business in a digital environment.

Jorge: Well, I don’t think I can top that. That seems like a fantastic place to summarize and wrap up the conversation. Where can folks follow up to find out more?

Closing

Bram: So by the time this podcast airs, we will have just done a workshop with Rosenfeld Media on this topic. And we did one of those workshops in the spring, and it sounds like we’re getting into sort of a spring/fall cadence. This workshop seems to be reasonably well-attended. We also are about to sponsor a conference in our field, which is called KMWorld. And the sub-conference that’s part of that event, which is in Washington DC in November, is Taxonomy Bootcamp.

And you can always find out more at our website, which is factorfirm.com. Our name is actually Factor, but we were able to get the domain “Factor Firm.” And so, a lot of people call us that, but it’s factorfirm.com. And you can find out about the information layer there. We have a maturity model for organizations to help them self-diagnose where they are in what we’ve seen over our many years of doing this as several stages, about five stages, of maturity. You can take a quiz about that, and that’s one of the things that we do in our workshops is take people through that exercise.

And, you know, we’re on LinkedIn and Twitter. It’s always @factorfirm, or “factorfirm” is the handle in whatever environment we’re in.

Jorge: Fantastic. Well, I’ll include links to all of those things in the show notes. It’s been such a pleasure catching up with both of you.

Gary: Yeah, it’s been great to talk, and hopefully, we’ll get to see you in person again one of these days.

Jorge: Hopefully soon.

Bram: Yeah. Perhaps in New Orleans next spring.

Jorge: Yes. And what Bram is talking about there is the Information Architecture Conference, right? Spring of 2023. Thank you both for being on the show!

Gary: Thanks for having us.