Information Architecture for Headless Structured Content

Michael Andrews
Hello, everyone. Thank you for joining. My name is Michael Andrews, and I’m talking to you today from Washington DC. I’m going to be talking about information architecture for headless structured content. Just a little bit about myself, I have worked in the field of content strategy and user experience for the past two decades. I work currently as content strategy evangelist at Kentico content. And Kentico content is a content as a service solution is a type of content management system that uses an approach called headless and mentioning this because I’ll be discussing a bit more about what headless is and how it works. I’m going to be drawing on some experiences our company has had working with headless content, but I want to assure you that what I’m going to talk about should be broadly applicable to anyone who is working with content these days. So, the theme of the information architecture conference this year is around making sense. And I feel the topic I’m going to be talking about really fits in well with this theme, because we need to help users make sense of their needs. And there are changes about how people are working with content. And these changes are affecting the content industry and it is very likely going to affect all of our careers of people who work with content. So, I want to spend a little time unpacking this concept of headless which I mentioned, Now. some of you may be familiar with a concept. others of you may have heard of it but aren’t exactly sure what it Is, and some of it, some of you may find this an entirely new concept. What I want to emphasize about headless is that it is a technical approach, right? It’s not something that is an approach that people who are not technical have been thinking about a lot. But it’s a technical approach that has a number of benefits. And so it’s a case of non technical people starting to learn about this technical approach and understanding what the value of it is for them. And one of the key ideas in headless is that you have a content model, which is a way of organizing all the content that enterprise has. Now this is an internal way of organizing content. It’s not like in sitemap, which is about an externally facing organization for the content. The content model really is an internal way of organizing content. So we understand the relationships between different types of content. Another really key idea and headless is that there’s no one UI that the content is being developed for. There are actually a number of places that the content could be delivered to different kinds of channels and platforms. And this is where the word headless comes from. I know headless may sound like a kind of grisly sort of word. But really, it means that there’s no head, nothing leading the no design leading the content, the content itself is its own, has its own purpose. And then the designs need to match what the content can do. Another really core idea and headless is around API. So this is really a way of allowing the content to get to different places allow the content to be accessed. So these are some of the concepts now the benefits really are a couple of major ones. One is that you create content once. And you can distribute this to many places, we say create once publish everywhere. And it allows the content to be much more flexible in terms of how you use the content. And it also allows you to reuse content. So you may create content initially for one purpose, and then you’re able to reuse it again for slightly different purpose later on. Here’s a diagram that will kind of explain a bit about the relationships that are happening here. And what I want to suggest to you is that information architecture really is happening in a lot of different places. On the far right, we have our users who are accessing content through any number of devices now a number of different channels and this is really exciting stuff. A lot of people at this conference are talking about about all the different ways that people are accessing content, even sort of new areas like virtual reality, certainly audio content and chat box and screen based content, all these exciting ways that people are accessing content and headless enables this type of interaction with content to happen because we’re able to deliver to most any kind of device. If you are curious about who is using headless, it’s being used by all kinds of companies. Everyone from say, The Washington Post to gaming companies are one of the biggest users of headless content management. And so we can do quite a lot with headless and it really the way we connect the content to all these different devices is through API’s. Now API’s are a really interesting area. I know that Jared spool has had mentioned that designers should become more familiar with API’s, because so much of the content that designers are going to work with in the future will be coming from API’s. We have the headless CMS, and this is where the content is stored. And this is something that is not a piece of equipment that you’re going to look at. It’s in the cloud there. And it because it’s in the cloud, it provides a lot of flexibility. But even though it’s in the cloud, we still have a need for authors to be working with this content. And this is where the content management user interface occurs. And so we have authors who are creating content for the customers and they use user interface to create this content. And that’s really, one of the things I want to focus on in this talk is how authors use the interface to create and manage the content that they are going to be developing for users. So we can think of this as a many to many, to many relationship, actually. So we have many different authors who are creating many different content items that ultimately are being used by many different people in many different contexts. And when we think about authors, we think about writing, and we think about how we create digital writing now, and we have many advances in our digital writing. We are creating messages all the time. And we find even that we’re able to leverage all kinds of new technologies where we can write messages where our thoughts are completed for us with different kinds of auto completion and things like that. It feels good. Very simple to create digital text these days, but at the same time is getting to be a little bit more complex as well, because in an enterprise sections setting is not just an individual who’s creating text, there’s actually a group of people who are involved in creating content. And it’s a collective nature of content creation, that is kind of forcing us to rethink about how individual authors need to work with creating their content. And the way that we are starting to work with this need to have multiple authors involved with creating content is through structuring content to focus on specific aspects of a larger message. And this is been around for a while, but it’s starting to get Quite mainstream. There was a survey last year at the Content Management Institute did, which indicated that 40%, over 40% of companies are using structured content. And you can see this happening in all kinds of ways. So on the right is an example from opower, a company that provides information to customers about their energy usage. And we’re looking at an example of an email that they send out to customers and even something as old school as an email, which has been around for, you know, 40 years is starting to incorporate structure to indicate different ways of breaking down messages to provide more flexibility. These changes are happening because first of all, technology is enabling it but also because customers are expecting This customers want content in different ways. They are wanting content on different platforms, they may have different media preferences, they may need different lengths or different levels of detail. And structuring content allows these sorts of variations to be possible. But it also creates a challenge to understand how to do the structuring how to create containers because it’s hard to break things into very tidy containers often, there’s often so many different connections between different kinds of information. So looking at how to make content modular and and ways to connect the different modules together to provide meaningful experiences for people. This move Word modularity is helping to erode this, this very long lasting metaphor of the page. So we have had, we’ve been working with pages since we have been writing by hand. We obviously had pages in the print era, we adopted this metaphor of the web page when the internet was first launched, and is a very, very entrenched metaphor on the page. But we’re seeing more and more evidence that what people are using are not necessarily webpages, we have increasing number of non website, touch points, things like single page applications where the content is continually refreshing. Within the user interface, we have scrolling. We have different sorts of metaphors. Now with tiles and stacks and things like that. And we’re finding that one output. One input can have many different outputs. So we aren’t creating content to exist in one static page. It can exist in many different forms. And I think we work in the information architecture profession understand that these changes are happening. But for many other people, they don’t quite see this level of change. And that includes, indeed, a lot of authors who are creating content. So what I think we need to think about as profession is how we can support information architecture for authors and not just for readers, we spend a lot of our time thinking about how people consume content and the different ways that they navigate through content. But rather than just sort of looking at from the navigational perspective, I think we also need to think about it from a more deliberate tasks. perspective? How can authors best express themselves? You know, what sort of organizations do they need to support their expression and it’s a bit like the layout of a keyboard. I mean, there are multiple ways you can layout a keyboard. In the United States, at least, there’s a one standard the ante standard for how content is laid out. And we have to think about how we can support that. And I had an experience myself where I knew that I rely on a certain information architecture, and my expectations got shaken up a bit. I had an experience where I was about to take a business trip to Europe and my laptop died and I needed to borrow a colleague’s laptop, which had a Czech keyboard, and even with just some minor differences in that layout, I have had to really think hard about how the the keys were aligned with with what I was trying to do. And then it’s kind of an example of how we need to work with people’s mental models to help support their creation of content. So structured content has a certain organization to it. And I want to talk a little bit about how this works. Some of you are familiar with this. If you’ve been working in IA for a while and maybe have heard people like Carrie Hain or Jeff Eaton talks about content models. Some of you may be less familiar with this, I just want to introduce a few core concepts here. So that we have a common basis for discussing about how we can manage this. So on the far right, we have the user interface can be anything it could be at a smartphone, it could be a avoided Spot doesn’t much matter, but it will post different content items that might be displaying or are being presented. These content items provide an experience, especially when there there are multiple content items together in the same context.

Michael Andrews
We asked where do those content items come from? Well, there are sort of a library of potential content items available to be presented in the user interface. And these items have some sort of commonality. So we have a couple of items at the top that deal with events once the IA conference and others world a day. And these are similar things and we are choosing one of them to present in the user interface. Similar we might have content items relating to a location so we might have information about a location such as New Orleans if we were imagining that we had been able to get to New Orleans this year or Washington DC. The content items themselves then are based upon something called a content type, which might think of is kind of a structural pattern for content that is made up of different elements. And this helps to organize the different bits of information that are relevant that need to be appearing in a content item. All the content items together, although content types other together, comprise what is called the content model. That’s a very quick summary of how this works internally. But it will help you give a sense get a sense of some of the key things that we need to be thinking about how they are presented and labeled. So back to the needs of writers. writers have a big task which is to create content if it’s structured content, and one of the opportunities they have is to reuse existing content. So they have a decision, do they need to create new content? Or can they reuse existing content? Now to make that decision, one of the key things that you have to think about is the inventory of content that’s already been created and is available to be potentially reused. Now, reusing content can be efficient if you are able to identify the right content to reuse. You don’t have to go through the effort of recreating something that’s been done already. Obviously, if it’s not very easy to access, it’s not providing that kind of efficiency. But I also want to encourage us to think about reuse more broadly. is not just about trying to save a little bit of time. It’s really about trying to build upon what we have you, if you already have created some content about one topic, you can use that opportunity to build upon it and create sort of additional content that elaborates on what you’ve created. So there are sort of editorial benefits to reusing content. When we look at the task of authors, they have the ability to work with new content that they’ve created themselves or to work with existing content and they have different opportunities. They can create new content that can subsequently be used by someone else later on or maybe be used by themselves later on for a slightly different purpose. They can reuse existing content and they may want to revise content. So That’s another sort of choice people have, do I need to write something new? Or can I just update something that’s been done before, we have a chance to combine both new content with old content. So you might create something new and supplement it with some other things that already are available. And finally, there’s a possibility of recombining existing content. So this is almost like curation, where you’re thinking about a particular audience or circumstance and you think, alright, it would be really nice to combine several things together that are already exist. how this is done, the structuring the content, and making it available to people has been under consideration for a while now. So most everyone will be familiar with Jacob Nielsenand close to 25 years ago, he started talking about how to modularize information or You know content. And he distinguished two concepts. One is what he called weak object orientation with strong object orientation. So when weak, weak approach, we have like just a small number of very big content items. So this is sort of how you think about big long articles. And if you don’t have very many big long articles, you don’t need to describe them in a lot of detail, because you can just find the big long article, but if you’re looking to structure items into much more modular pieces, then it’s important to have much more description about what these items represent. So you’re able to connect them and choose the right ones.

Michael Andrews
One way to think about this is it’s a little bit like selecting a photo if you are needing a photo for something that you’re about to publish. So you first face a decision around you when To use an existing stock photo, or maybe commissioned a new photo, and if you decide you want to use a stock photo, you may have already seen something in the past you thought would be a good candidate. But you need to be able to find that, that photo that you saw in the past, and you might be looking through the photos, and you don’t see the one that you are certain that you saw in the past, it’s just not there. And you’re a little confused where where it is, you know, you see other things that seem to be similar. And what this illustrates is the limitations of sort of a cm point approach to things we would like to be able to just immediately see whatever it is we want to see. But in reality, if you start having a lot of items, it’s very difficult to be able to expect that you’re always going to just instantly find what it is you’re looking for just by recognition alone.

Michael Andrews
It’s a lot like a warehouse where you have an inventory, and you have lots of different parts and pieces that are in these brown cardboard boxes that all look the same. And the question is what’s in the boxes? And, you know, if I’m looking for something specific, where do I have to go? If I’m not sure what I’m looking for, how do you discover what’s available in those boxes? This idea of an inventory for physical things in a warehouse is very similar to some of the issues that the content inventory presents to authors. And enterprises are all about scale. So it’s, you will typically in an enterprise setting for a content management system have, you know, maybe hundreds of users, you can have quite a number of content types and very easily have fat Thousands of content items. So that’s a lot of different people looking for lots of different things. And we need to help authors understand what’s available. And so part of what an author needs to do is think about taking different items and being able to compose them into something that’s going to be a coherent experience for customers. So what parts need to come together to deliver a complete experience for customers. And part of the benefit of using the structure is being able to provide a lot of variation. So if we get this right, we’re going to give the customer something that’s really customized are tailored to their needs. So we have the possibility for variations around what channel the content is optimized for which particular audience segment content Item is meant to speak to our what stage in user journey that the content item is really meant to support. So these are all sort of factors that authors need to be thinking about. And this is a little bit of a change for how authors think. I mean, a lot of authors think about just having a blank webpage that they can just think in what they want to do and create something and publish it and move on to the next thing. This is a bit of a paradigm shift. And with as with anything that’s going to be new for people, there are going to be problems and risk that are involved. One of the things is that names might be ambiguous. So how things are labeled, may not be done well. If we throw every detail at authors, it may seem very complicated for them and Authors may also encounter a situation where they want to revise something that they’ve already seen published, and they need to be able to work backwards to figure out what item is, is supporting that thing that they want to revise. So there are possibilities of errors, finding things if authors mislabeled things that’s going to create problems later on. Some some difficulties either not finding what they want. And if they don’t find what they want, they may end up duplicating something that already exists, and that’s going to create even more difficulty later on. So these are things that we have identified as as, as problems and risks on we’ve been thinking about for a while. One of my colleagues, Peter Augustine wrote This article nearly two years ago now and UX collective talking about this issue, and just to give you a sense that it’s something we’ve been working on, and we’ve made a lot of progress on, but we still are aiming to do much more improvement around because it is a new way of working with content. And there are lots of different variables to think about. So this is some of the UX team, we have involved with thinking about this. And we done a lot of user research with customers. And some things that have come out is that it’s really important for people to have a mental model of what is going on. Now that mental model doesn’t need to be the model that I showed you earlier of how content models connect to user interfaces. That’s a fairly complicated model that, you know, you may understand as a as a professional IA, but may not be the right way to explain it to a, a author, but um, they need authors do need a mental model. And they also need to have some contexts or framework around what they’re looking at. So if we just present a list of all thousands of content items that their company has produced best going to be overwhelming to them. So one of the biggest insights that we’ve gained is that what is it that we’re naming here, so we’re talking about content items. Now these are content items are pieces of content, they’re not articles. So one of the big important things to consider is that articles typically have a title, but a content item, may or may not have a title. You know, it may not or may not have a heading some content items, if they’re just the notification will have a heading. Other items may have multiple titles, you may have a short one that’s being displayed in one context, or another one that’s used in a different context. So we first need to help break this sort of expectation that you name an item based on a title that displays the other things that we’ve been thinking about relate to how we support riders, and there really are several things that we’re looking at. One is how we can support their choices. So we’re trying to make decisions that riders need to make easier and reduce sort of their cognitive load. We’re looking at providing better time scription of things to sort of disambiguate things that might be confused. We’re looking at ways to systematize things so that there’s more consistency in how things are, are labeled and presented, and giving people more context so that they see how what they’re working on or fits into the context of how customers are using it. The other sort of thrust we have is around trying to simplify what authors see. So we’re really looking a lot at their work patterns, in the sorts of things we can expect them to be using and in trying to prioritize those things. Looking at the roles, we have different kinds of people who may need to only do certain things and so on. There’s no point in showing those people everything. We want to provide more sort of options for people because there’s not one standard workflow that all companies use, there’s quite a bit of variation. So we want to give different enterprises the opportunity to customize the workflows to really tailor to support the needs of their riders so that people can be as efficient as they can. And we’re always looking at ways to improve find stability. So on the right hand side, we have an example of making some suggestions to people based on what we know about them and things that they’ve worked on before. And we have the inventory, and I want to give you a couple of examples of how the inventory works here. One is The This is an inventory of content items. And we have a column for the name of the content item, the workflow step, the type of content item, it is some things about the status, but we’re providing different ways of identifying the right content items. So, for example, filtering, by content type or filtering by taxonomy. We also provide different search options for people. Now, the naming is very important, and I think this is where I always have a big opportunity is that they can help authors name things in a way that’s going to be consistent and easy to follow. So we have the names of content items, and we want to help authors name these things. So we can distinguish differentcontent items that might talk about the IA conference, for example.

Michael Andrews
We can’t name all these items, ie a conference, because we need to distinguish the difference between them. We also want to think about the differences in content types and make it very clear to people what the content type is about. And I think these are the kinds of things that information architects can help with. Another area that information architects can help with is around taxonomy. So taxonomy is very important to identifying what the content item is about. And it plays to four roles. One is going to help the authors identify the right content to use, and it’s also going to help deliver that content to customers. So taxonomy is getting richer, we’re finding that it’s useful to identify not just the top But you know, any kind of variance around the topic looking at the user journey, step per stage, the specific audience that the item is intended for. And we are even looking at trying to support that discussion discussions around taxonomy terms that people can use the content management tool and talk among themselves. Is this the best term to use? You know, would you suggest a different one or add an additional one? So, I believe that information architects have a very important role to play in the future of how authors work with content. And I would encourage you to look at how headless is different from sort of more traditional web content management system that Around webpages and and try to understand more about how content items work and how they can be managed. What IaaS can do, I think, is to help their colleagues set up a design system for content. So right now we’ve had a lot of interest in designed systems for front end development and providing consistency and, and certainly in visual user interfaces. By using design systems. I think there’s a similar opportunity with design systems for content. And IA’s can help move standards for these design systems and champion governance of the design systems so that people understand what’s available, how to use it and use it consistently and want to use it because they see it actually supporting their work. And lastly, I just want to say that information architecture really has a critical role to play because it helps to connect authors with the customers that are going to be looking at the content the authors are creating. So just we think about something like notifications, I mean, notifications are just empty boxes. And the experience of that notification depends on having meaningful content in there. But it’s also having the content described well, so that the right notification is getting delivered to people when they need it. And I think that’s really one of the ay ay ay ay ay ay community can help support this evolution and how content works. I want to thank you everyone for listening here. And if you have questions and want to reach out to me, you’re welcome to do that. My email is there. My Twitter handle, and I’ve written a fair number of blog posts. And in recent months touching on some of these issues, so there’s a URL there if you have an interest in reading more about some of these topics.

Transcribed by https://otter.ai