Michael Andrews: Content Metadata Strategy – Episode 74

Podcast: Play in new window | Download

Subscribe: Apple Podcasts | Google Podcasts | Stitcher | TuneIn | RSS

Michael Andrews literally wrote the book on metadata for web content (actually two books).

Metadata puts the structure in structured content. It helps both humans and computers understand what your content is about and how it relates to other content. It ensures that your content is always up to date, easy to find, and able to be tailored to your customers’ unique needs.

In today’s hyper-connected and ever-evolving digital media landscape, every online publisher needs a content metadata strategy.

Michael and I talked about:

• his work as a Content Strategy Evangelist at Kentico Kontent and his prior work in agency, consulting, and in-house content strategy roles
• his definition of metadata
• the range of benefits of attaching metadata to your content
• how metadata helps search engines find and display content
• the benefits of metadata standards like schema.org and Open Graph
• how to attach metadata to content (, which )typically happens in a content management system)
• the differences between internally used metadata which is used for workflow and other administrative purposes and externally facing metadata which is published along with the content
• the emergence of “unbundled” content and how metadata helps reconnect content components, permitting new practices like omnichannel publishing via APIs
• how accessing content via APIs permits modern business practices like content personalization
• his approach to metadata strategy: taking a holistic approach, thinking across the content lifecycle to account for all possible metadata scenarios
• how a metadata strategy (along with good governance practices) can help span organizational silos
• the role of taxonomy in metadata – it permits teams to align on a common terminology
• how metadata can help align content to your customer journey map
• how getting started with metadata can help you start thinking about modularizing your content

Michael’s Bio

Michael Andrews is Content Strategy Evangelist at Kentico Kontent. Over the past two decades, he has advised organizations in half a dozen countries about content strategy and user experience in diverse industry sectors. His previous roles include working as a Senior Manager for Content Strategy for Publicis Sapient, one of the world’s largest digital agencies. Based in Washington DC, he’s a frequent speaker at industry conferences such as Lavacon, Taxonomy Bootcamp, and the Information Architecture Conference. Andrews has published two books on the role of metadata in content strategy. He holds an MSc, with distinction, in human-centered computing systems from Sussex University in England.

Connect with Michael on Social Media

• Twitter
• LinkedIn

Michael’s Books

• No More Silos: Metadata Strategy for Online Publishers
• Metadata Basics for Web Content: The Unification of Structured Data and Content

He has also published an e-book at Kentico Kontent, The Complete Guide to Content as a Service (CaaS).

Video

Here’s the video version of our conversation:

Podcast Intro Transcript

Digital content is evolving to guide customers through unified experiences across a variety of channels and devices. These new publishing practices require unbundling conventional publications like documents and manuals into smaller chunks of content. You then need a way to re-assemble those chunks into new content formats. Metadata provides the connective tissue that makes this possible. Michael Andrews can help you understand why and how to build a metadata strategy to revitalize your organization’s content.

Interview Transcript

Larry:
Hi, everyone. Welcome to episode number 74 of the Content Strategy Insights podcast. I’m really happy today to have with us, Michael Andrews. Michael is a content strategy evangelist at Kentico, which is a CMS company. Well welcome, Michael, tell us a little bit more about your background and your role there at Kentico.

Michael:
Well, thank you, Larry. I really appreciate being part of your podcast series. I know you have a lot of very interesting people who have appeared on your show and I’m very happy to be part of the series. As you said, I am a content strategy evangelist at Kentico Kontent, which is a headless content management system. And my role there really is to help content strategist really do more with their content, to try to help them think out ways they can be more effective with their content. So I do a lot of speaking and writing about the content strategy field and giving people ideas about what they can do when they manage content. So I’ve been working in the field of content strategy and more broadly, in user experience, for maybe the past two decades and have been working in different roles. I’ve worked in consultancies and agencies, I’ve worked internally in some large organizations and now obviously I’m working for a vendor, but it’s given me an opportunity to look at how people work with content from a variety of different perspectives and different sectors and indeed in different countries too.

Larry:
You might be the most well-rounded professional I’ve had on the show because you’ve done everything, agencies, consulting, in house and now with a vendor, like you said, that’s great. Hey, I want to kick off because I really want to focus, there’s a ton of stuff we could talk about, but even within that, there’s a ton of stuff we could talk about, about metadata. And that’s what I wanted to really focus on today. I guess first we should start with, for folks who might not be familiar with the term, what is metadata? How would you define it?

Michael:
Yeah. So metadata is one of these terms that sounds very abstract and it is in fact, an abstract term. It’s been around for quite some time, but not everyone knows what it means. There’s a standard definition of metadata, which is that it’s data about data. And I’ve never really found that definition very helpful because it’s just recursive and you don’t really know what does that mean. The way I explain it formally is that it’s data and you can think of data as like a fact about an aspect of a content item or something mentioned in a content item, would be a piece of metadata. So that has two parts to it, one is the idea that you’re talking about the content. So we’re talking about something, it’s like, all right, who wrote that content? That’s an item of metadata there, the author is. Or if I write a content item about you, Larry, you’re mentioned in the article, so that’s also metadata, that it’s mentioning you in there. So it has those two aspects to it.

Larry:
Right. There’s both visible metadata. If you looked at a restaurant listing, it would have the address and the menu and that’s all stuff about the restaurant, but then there’s the behind the scenes stuff. Like, when did we write this review or that kind of thing.

Michael:
Exactly.

Larry:
Yeah. And I think for a lot of people it’s like, “Oh boy, another detail to worry about.” I guess showing people the benefits of caring about something, I think, can help. So what would you say are some of the biggest benefits of attaching metadata to your content? What are some of the classic uses and benefits of it?

Michael:
Yeah. Yeah. So there really are a range of benefits. So if you are familiar at all with the idea of a content lifecycle and maybe some of your listeners have encountered that phrase before, but if you think about the content going through different stages of it’s life. It gets started, someone has to create it and then they will deliver it somewhere when they publish it and then it has to be maintained and then maybe people assess it at the end to see whether it needs to be updated or whether it needs to be retired. That’s a lifecycle of the content and metadata can help with every one of those stages. So it can help you when you are creating content because of one of the first things you want to do is make sure that you’re creating content that is something you haven’t already done.

Michael:
There’s no point in duplicating what you’ve done in the past. So if you have metadata for the content that you’ve already created, it’s going to help you locate that content and then you can say, “All right, we have or haven’t already covered this topic.” And then you can use metadata for delivering the content. So probably it helps with the pushing it out to the right channel. We talk about different channels of delivery. You can say, “All right, I need to put this content out in social media channels, or I need to put it out so it can be found by search engines.” And metadata can help with that, it can help people discover the content and then the people who use your content can use metadata to access it. They’re able to navigate through your content using the metadata, find the specific information that they’re looking for. Back on the other side, when you’re having your content that you’ve already been published and you need to maintain it and manage it, the metadata helps with that.

Michael:
It helps with your content inventory, understanding what you’ve published, how much in one direction you’ve gone as opposed to another direction in terms of the kinds of content you’ve published, do you have a lot more one content type than another, or are you more about one topic that another, metadata can help you get a picture of that. And then as I said, with the analytics piece too and doing the assessment, it can really help you drill down into the details of your content performance.

Larry:
Yeah, well I got to say, I love that you just rattled off five key benefits of it and never once mentioned SEO, because I think to the extent that there is awareness of metadata, that that’s where it comes from. And so there’s also, I would add to that list you just rattled, discoverability aspect of it. But let me ask you a little bit, because so many people come to metadata via one of the specific tools, schema.org and search engine optimization, I’m always looking at leverage points to get these really good practices into content operations. Is that a common entry point for understanding and implementation of metadata?

Michael:
Oh, absolutely, yeah, because people obviously want to have visibility in Google and so it’s hugely important. And let’s put it this way, SEO has been around for a long time, but it hasn’t necessarily been that focused on metadata, but certainly in the past five years with the rise of schema.org, as you mentioned, which is a particular metadata standard that Google and Microsoft and DuckDuckGo and any other search engine will be using that to help when they do this crawling of content, the web crawl, they index the content and they look for this metadata in the content and that helps them know what is there. And that’s how they will not only look at the search results, but even pull out very specific pieces of information that you might see in your search results page, these little panels of information that’s coming from the metadata.

Michael:
So most organizations get it, that it’s super important to be visible on search engines and so that’s often one of the first catalysts for starting to really think about metadata. But as you’ve alluded to, it is much broader than just search engines. So it’s a great place to start, but it’s not where you have to necessarily end, you can go much further.

Larry:
Right. And I’m thinking of many other places. One thing that occurs to me there, well, I guess and this is backing up from the benefits we were talking about, talking about schema.org and the fact that it is a standard. I think some people, when they first learn about a concept like metadata, they get a little worried that it sounds complicated and there’s lots of details to be attended to, but a lot of those details have already been created and codified in these standards and you’re just adopting a standard. There’s not, apart from whatever your technical implementation and procedural implementations are, you don’t have to do a lot of the heavy lifting. Is that one of the benefits of standards like schema?

Michael:
Yeah. I mean, definitely with schema in particular, but there’s also another standard called OpenGraph, which some of the social media platforms use, in both those cases they have figured out some of the key pieces of information that are useful to pull out in metadata. So you’re right, you don’t have to be thinking up everything on your own. Having said that, I mean, there’s a limit to what standards do and standards are meant to be something in common for everyone to utilize, but you may have some very specific needs in your organization that the standards don’t really go into in great detail and you may decide that you want to extend, build off the standards and do some of your own metadata work. And that does involve a little bit more thinking on your part, a little bit more responsibility on your part to figure out what to do. But I guess what I also just want to get across is the standards are there to help you, they’re not there to limit you, so if you need to go beyond what the standard has, you can do that.

Larry:
Right. It’s a baseline, it’s table stakes to be in that game, to be discovered by Facebook or Google or something like that. One thing I’m thinking about, it’s helped me to understand, the way I discovered metadata was by building HTML pages 20 years ago and there’s that meta section at the top, the meta description, the meta title tag. So I think that’s a common way that people discover this, but I guess the expressions of taxonomy [I misspoke here, jumping ahead to a later topic, actually meant to say metadata] would be, in HTML, if you do a view source on a document that has schema.org information in it, you’ll see that, you’ll see this specifically formatted information with schema, but there’s also things like RSS feeds and you mentioned the OpenGraph standard. Those are the expressions of it, but I’m curious, backing up from there, how are people working with it in their workflow? You write an article or a piece of content and then assign metadata to it, how does that typically work? How does that metadata get there?

Michael:
Okay. Yeah. So what happens really is most people are developing their content within a content management system and the content management system, as the name implies, is there to help manage all these little bits relating to the content. So what we can do is we can separate out two kinds of metadata at a very broad level. One is internal metadata, which helps to manage the content internally, make sure all the pieces get connected together. And then there’s external facing metadata, when you’re pushing the content out on a platform like Facebook or Google or some other place like that. So the metadata that’s used internally really is part of the content management system and so it’s already knowing when the date was published and things like that, it’s taken care of those kinds of details.

Michael:
When you’re pushing the content out, you’re actually publishing your metadata along with the content. So think about it this way. There’s some metadata that gets published because other outside parties need to consume that metadata and then there’s some metadata that’s really handled internally and it really helps the user of the content management system, which are authors and writers and content creators, help them just get their work done. So it manages things like workflow and the status of things and things like that. So we call that internal metadata, we often call that administrative metadata and that’s a term that a lot of this goes back to the days of libraries and books and librarians needed to catalog books and they started thinking about metadata back in the 1960s and 70s, when you had the old mainframe computers starting to catalog these books.

Michael:
And so they started thinking about different kinds of metadata and the administrative metadata is like one of those categories of metadata that evolved at that time. Thinking about what’s happened with the book, when was it acquired? That would be a piece of metadata for a book. So now we are in the digital content era and we started thinking about, all right, when was this content last revised? That’s a piece of administrative metadata. We may or may not actually publish that out to the outside world, but it’s certainly useful internally to know when something was last updated.

Larry:
And that gets at one of those internal benefits, content management benefits of having both, but there’s also the usability benefits on all the external, the externally facing metadata. There’s some other, just going back to thinking about the benefits of it, I’ve heard people talk about metadata as a, well it’s an expression of, I’m thinking, for example, like Carrie Hane’s and Mike Atherton’s domain modeling, or a lot of people talking about content modeling and there’s these conceptual models where you have entities with attributes and with metadata and other things associated with them, which is good, that you’ve got things clearly identified, but it’s really the connection between those entities that’s the real power. And I think we’re seeing that just exponentially grow in importance, the way the web and digital media have evolved. Can you talk a little bit about the importance of metadata in connecting things to one another?

Michael:
Sure. So really content is starting to get smaller, rather than it used to be people created these really big, long documents and expected everyone to read these big, long documents and now we’re starting to unbundle content. It’s a way you can think of it and think about how to make it where you’re just talking about very specific things and getting information to people as quickly as possible. When you’re unbundling it, it’s great, you’re breaking it apart, but we still want the ability to connect it back together again so that people can connect things that belong together and metadata helps identify what each part of the content is about and then it also helps to relate these different pieces of content together. So that happens in different ways. One of the key ways this is now happening is some people may be familiar with the idea of publishing across different channels. They call it omni-channel and you’re giving information to people in whatever channel they’re in, depending upon their context.

Michael:
And metadata really has a big role to play in this omni-channel publishing, where people are jumping between different devices and they’re moving between different situations. And a lot of that content now is being delivered through something called an API. So it’s really delivering very specific content, generally to a screen, although it doesn’t have to necessarily be a screen, it could be a voice interaction device. I know you’ve had some people on your show talking about that fascinating area. And the metadata really is what helps to deliver it. So if you have heard at all about something called APIs, the API is really delivering information based on metadata. So we have something like the content model, that’s what the API is talking to, the content model is breaking up the content, is expressing the metadata.

Larry:
That gets into, I don’t want to go too deep into these weeds, but I think it’s an interesting thing to talk about because so many people think of a content management system as this monolithic thing that goes from authoring, to content management, to content distribution and display. We’re in this world now where CMSs are decoupled, the back end from the front end and there’s this whole new world, there’s the JAMstack and other ways of expressing this, but you have something like Kentico Kontent, like a headless CMS and then you have people writing JavaScript or React or Vue or whatever front ends and the APIs are the things that connect them and it’s the metadata that gives you what you need to make that connection. Is that an accurate way of summing that up?

Michael:
Oh wow, you summarized it quite nicely. Yeah, that’s exactly right. So it’s allowing you to deliver something very, very specific to someone, so they don’t have to get everything because when people have a little tiny screen often that they’re looking at and they need that very specific bit of information. And then on the other side, it gives the user the ability to say, “This is exactly what I need.” And it can pull in exactly what the person needs. So that API is really talking to the content, which is being structured with the metadata and allowing for the right information to get to the right people when they need it.

Larry:
Yeah. Hey, the title of this podcast has strategy in it, I would be remiss if I didn’t ask you about metadata strategy. Can you talk a little bit about that or is that a thing?

Michael:
Yeah, it is, well I’ve proposed it as being a thing. And I honestly am not the first person to think about this. Actually, Christina Halvorson, I think a number of years ago, mentioned metadata strategy in passing and I thought it was a really interesting comment because I really hadn’t seen many people talk about metadata strategy. But what I like to do, I’ve actually written two books on metadata, my second book is on metadata strategy, the first book is really on the nuts and bolts and metadata. But my idea here is to try to think about metadata holistically. So as you mentioned earlier, people can get very focused on a particular aspect of metadata. Maybe they’re interested in SEO, which is great, as I said, it’s an important area, but we want to be thinking about the whole content life cycle and all the different uses of metadata.

Michael:
So metadata is useful for authors, it’s useful for the users who are using our content and it’s also useful for the machines that are managing the content and delivering the content. So all these different parties can benefit from metadata. And the purpose of metadata strategy is really to look at it in a more holistic way and think about what are the different ways that we can use metadata and can we start planning so that all these different scenarios, these use cases, support one another. So rather than seeing this as, oh, it’s just another bit of work I need to do and it’s only going to have this one benefit, actually there could be multiple benefits and you get much better return on your investment when you start thinking about the multiple benefits.

Larry:
You’re reminding me, one of the really common themes that comes up in this podcast and in every conversation I have with content strategists, is the notion of silos and of different parts of an organization doing things. And you do much broader content strategy work than just metadata stuff, but it sounds like a metadata strategy could be one of those unifying practices that could help people in marketing and support and sales and the web team, stay aligned because they agree on the details of what they’re talking about or agree on the principle that this should be easy to connect. Has that been the case in your work?

Michael:
It is. I mean, it really is a foundation of breaking down silos. So I think there are a lot of different things you can do to break down silos and governance is a very important thing to do and even metadata has governance involved with it. But more generally, we have to think about how we are describing our content and metadata is one of the key areas, it’s related to how we structure our content, things like content models and having an enterprise content model is very important so that the different pieces of content can fit together and taxonomy as well as, is very important. So taxonomy is really the values that you use in metadata. So if you are looking at a particular aspect that you’re trying to describe with metadata, like what audiences is for, the organization wants to have a clear sense of how they’re describing their different audiences and everyone needs to use the same terms and be aligned on what this terms represent. So all these things need to come together for the organization as a whole to be collaborating and supporting each other.

Larry:
Right. And it sounds like that’s a really, I mean, you can see how that would be an important tool in aligning people. It’s so interesting that there’s always fiefdoms and silos and things going on, but fundamentally, everybody at least says they want to get along and connect and agree. And I think there’s this growing also … And actually, let me ask you quickly about this, there’s this growing awareness, I think, of the customer journey as the guiding thing in putting together your content or whatever other interactions you’re creating – your service design even. But that customer focus though, is the thing that drives that and it seems like this is just one more tool to like go, oh yeah, but our customers label or articulate or whatever things this way or that and metadata is just one of the ways that that’s articulated. Is that a way to summarize?

Michael:
Yeah, well I think you’ve raised an interesting thing. So we have metadata that will be managing what gets delivered to customers in different situations. So in my work, I think about what stage of the customer journey someone’s at and you try to describe that and then you describe what customer segment you’re talking about and then you say, “All right, this content is intended for this customer segment at this stage of a customer journey.” And you come up with ways of describing that. That’s all an internal management for creating the content, managing it and delivering it. But as you mentioned earlier, there’s metadata that customers see. So if they’re navigating through your content, you probably want to have the content labeled in a way that makes sense to them. It’s using their terminology and they can go, “Oh, that’s exactly what I want.” If they’re looking at a series of choices there, they know that’s what I want.

Michael:
So you can talk about your content internally in a slightly different way than you talk about it externally and you just need to make sure that those different expressions are connected in a way that you can differentiate them and then be able to track them.

Larry:
Yeah. You’re making me want to convene a panel with you and Gerry McGovern and a taxonomy person. Hey, I just noticed Michael, we’re already coming up on time, these conversations always go so quickly. But I always like to give my guests a chance, is there anything last, anything that hasn’t come up or that’s just on your mind about metadata or content strategy, that you’d like to share with the folks today?

Michael:
Yeah, I encourage people to get interested in this sort of thing. It does seem a little daunting and I totally understand that. It’s a new set of terminology and once you start digging into it, you may feel that there’s way more than you can learn quickly, but to start with very simple goals of just wanting to learn a little bit more and see where it could be useful to you, think about your challenges, talk to other colleagues. I think if you do have people on your team who work with SEO or some of the developers maybe on your team who maybe understand something like databases, can help you think about how content is managed on the backend. Once you get an understanding of that, then it opens up the possibilities to say, “You know what, if it’s being managed that way on the back end, that gives me new possibilities for how I deliver it to people, or maybe I want to break it up.” So we’re not thinking about just these big monolithic articles so much anymore, we’re actually thinking about content in a somewhat more modular way.

Larry:
Right and that’s a whole other conversation. Well, thanks so much, Michael. Hey, what’s the best way for people to follow you or stay in touch? Do you have a preferred social media or other ways to stay in touch?

Michael:
Sure. I mean, people can follow me on Twitter, I’m at story needle. They’re also welcome to connect with me on LinkedIn. I’m not the only Michael Andrews on LinkedIn, but if you look for Michael Andrews at Kentico, you’ll be able to find me there.

Larry:
I’ll include links to those in the show notes as well. Well thanks so much, Michael, for taking the time. This is really a great conversation. I really enjoyed it.

Michael:
Well, I enjoyed it as well, Larry. Thanks again for having me.