We’ve been writing and talking a lot about content operations (ContentOps) at GatherContent recently.
Our Co-Founder and Product Director recently presented a webinar introducing ContentOps and sharing some of our thinking on what it is, and why its important for organisations so they can repeatedly produce effective content. What follows is an edited transcript from that webinar (or you can watch the recording).
Today’s session is about some of the research we’ve been doing as a company over the past seven or eight years. More specifically, creating effective content every day and introducing “content operations“. People can be wary as it just seems like a term for the sake of having a new term, but what is important, is this idea of the trenches of an organisation’s content production and what happens behind the scenes on the inside of successful organisations to enable them to create effective content.
A lot of discussion that happens around the strategy, about what’s involved with coming up with a plan, or architecting really great things. In terms of how we deliver things to our audiences, there’s a lot of discussion around products, CMSs, headless CMSs and multi-channel publishing systems. But not many people seem to share stories from the daily production side, about what happens in between this strategy and these publishing channels. What tools and what capabilities do you need to execute on these things?
So, content. Before I began building websites, I never really heard anyone describe the information that exists on the internet as being “content”. It was information, as you would expect, and it had a huge amount of potential to be connected and digitised.
Books like Restructure by Deleuze and Guattari convincingly compare the Web to this big, organic rhizome, saying how it would saturate human knowledge. Which, however you interpret it, sounds pretty powerful. This is a bold quote from Information Foraging Theory by Peter Pirolli.
Modern mankind forages in a world awash in information, of our own creation, which can be transformed into knowledge that shapes and powers our engagement with nature
– Information Foraging Theory, Peter Pirolli
His research really backed this up, showing how humans were starting to adapt some of our skills and behaviours that we’d taken from hunting and gathering and applying them to finding and discovering information on the Web. Jakob Nielsen described this theory as the most important concept to emerge from human computer interaction in the last decade.
The above is a visual map of the internet. All the points that surrounded these intense visualisations from people like David Hand were convincing, arguing that this ‘information fungus’ would change everything from how we interact with government to health care and education and everything in between.
I remember doing my first real website project and hearing the project manager referencing the “content” of the website and for some reason we were no longer talking about information on the Web in the same way. Now it was about content, and content was annoying. It was always in the wrong format. It took ages to process. It delayed things. And at the end of the day, it just wouldn’t fit into those beautiful Flash menus you’d spent ages creating. And content ultimately came last. It was an afterthought.
Working in an agency, I realised that even when people did care about content, it was just hard work. When you’re building out websites or big projects and you’ve got deadlines and content design and technology all developing in parallel and being juggled, it’s difficult and content did, and it still does, delay these projects because it’s an afterthought. It just isn’t delivered on time, or it annoys designers and developers because it’s submitted on floppy disks or crumpled up bits of tissue paper or whatever else. And ultimately just makes things hard to get done within time or within budget or even for a lot of businesses just to make money and survive.
While some agencies and organisations just chose to ignore content, this wasn’t a luxury for other companies that started out online, like eBay. Their audience had to be able to interact with them solely online. Their business depended on it.
This is eBay’s help documentation from 2001, which I accessed via the Internet Archive Wayback Machine.
Content wasn’t a novelty for eBay at this point. It wasn’t a case of having a GeoCities site that you could talk about your local birdwatching club. Content was a component of their business’ operations and there’s a serious amount of content on that page. It’s not huge by today’s standards, but it still involved a lot of thought to produce, organise, publish and maintain. eBay did some amazing things to handle this exponential growth of content, because they had to. eBay adapted XML that was being developed in universities like Harvard and Stanford in the 1990s, and they started applying XML’s structured markup to their documentation to help manage it more efficiently.
It was partly because of a lack of resource, and the rapidly increasing scale of content, that they had to come up with these ways to manage things more efficiently. This really summarises the emergence of structured content, first of all, but also content governance and looking at how we maintain large bodies of content. But most importantly, this is an acknowledgement of content as a business asset and as a functional part of an organisation.
Another good example is from National Public Radio (NPR), which is a huge network of over a thousand radio stations based in America, and they’re publishing so many different formats of content, from articles to books and everything in between.
The above is their model for creating multi-channel content, content that can be adapted to work on different devices or delivery channels. They wanted to be able to create a piece of content once, and deliver it to different formats, or in different formats to different places. They elaborated on this Create Once, Publish Everywhere (COPE) model which allowed them to distribute books and articles in multiple digital formats and adapt them so they worked on different devices.
This is a great example of infrastructure, which is a big component of content operations. But it suggests how much had to change to create content that reaches an audience. And it illustrates that a large amount of work behind the scenes has to happen if we want people on their phones to be able to access our content or interact with our organisation.
An even more recent example, which is probably even more referenced, would be GOV.UK, the site for the UK government. I find it amazing the scale of organisational change and political navigation that had to take place to make the Government Digital Services (GDS) team, possible.
They adopted Agile. They started talking about meeting user needs with content. They created new internal infrastructure and tools to create structured content. They created a whole reporting system which was open sourced, so the world could see what they were doing. They created new roles for people like content designers. They created a whole new pay structure for teams in GDS, so the people working in content had more authority – literally putting content higher up the chain. And they created one of the most referenced content style guides ever.
Along with this they implemented training systems to get people to write better for the Web, teams had to train for up to two weeks. You can’t really see any of that from their homepage. You can’t really see how much internal change had to happen in the trenches to make that possible. This is one of the greatest examples of organisational change and showing how that kind of transformation can have such a huge impact on the organisation’s capabilities online.
According to an article published on GOV.UK in 2015, this initiative saved the UK government 3.5 billion pounds over the first three years, which really shows the impacts of this effective content on the organisation.
A final example is Deliveroo. It’s a good example because it’s a fast food delivery service, so not something a lot of people might imagine requires a huge amount of effort in terms of content. But they’re doing some amazing stuff when it comes to product design and content. They written about how they approach content and product design. They’ve totally changed the structure of their design team to make content the leading principle in their design process.
Their new team is called the content, research and design team and it’s a team of 38 people broken down primarily into content designers, content design managers, user researchers and product designers.
Successful organisations are choosing to, or having to change, the way they work internally with content in terms of process, or even product design process, in order to be successful, to compete.
Whether you’re an agency building websites for clients, an online marketplace like eBay, a government, a radio station or even a takeaway fast food delivery service, you need to have certain internal capabilities if you want to create effective content. That being content that reaches and resonates with your audience on whatever device they’re using or whenever they want it, and ultimately serves a purpose and is of value to your organisation and the people that interact with it.
A lot of this comes down to how organisations work internally. From changing people’s roles and responsibilities to coming up with new approaches to infrastructure for structured content, to adopting Agile, to implementing governance models for content to changing the way services are designed in the first place. Or to just creating new teams and training programs. There’s definitely a lot to be juggled behind the scenes.
From investigating so many of these organisations over the past few years, we’ve seen there are thousands of different ways to do this, but there’s one theme which is present across all of them, and I think it’s best summarised as this investment in content operations.
Our ContentOps Venn diagram shows three main components. We’ve got people, process and infrastructure, which make up content operations, and these would be the three bases you want to have covered, and the three things you want to consider.
This sits after, or autonomous from your plan, or your strategy, and similarly, it sits before your delivery tools or publishing channels like your CMS, your help centre, social media, marketing. Even like non-digital channels like print. It’s very rare for the system to really solve any of these content operations problems. It’s very rare to see a CMS solve people problems, or process problems or often even infrastructure problems.
You might not have a plan on how you’re wanting to communicate with your audience online or you might not even know who your audience is or what on earth you’re trying to do. I’m sure we’ve all worked in organisations where that’s the case, where there maybe isn’t a very clear plan or strategy. But it’s very rare for there not to be some form of these three things.
Every organisation has some form of content operations, regardless of whether or not you’ve got a plan or a beautiful strategy. That you’ll have someone somewhere, a person or people, that are creating content in some kind of tool. It could be a Word document or an email and they have to do something in order to get that published, so you consider that a process and it could be something like emailing a Word document to someone that knows how to use the CMS. And that is what’s required in order to get content delivered.
You might not have deliberate content operations. You might have deliberate infrastructure or some deliberate processes or some deliberate things about people, but it’s important to have deliberate content operations in relation to all of these things, and about thinking about them as pillars to approach.
The idea of content operations isn’t brand new. There are a few definitions of ContentOps in existence.
This one is from Deane Barker.
Deane is the author of the O’Reilly book, Web Content Management and is a definite expert and real pioneer when it comes to content management infrastructure in particular, and content strategy in general. His definition is a bit technical and clearly states the idea of this void between content strategy and content management.
The next definition from Colleen Jones is brilliant:
It’s so inclusive because Colleen takes into account the people and the process required to create content. And importantly, it isn’t just specific to marketing content, so it could include more informational content such as support documentation.
If you’re really to think about this full picture, this holistic view’s really important. I also really like how Colleen talks about what’s happening behind the scenes. Because I definitely feel from our exploration into this that these kind of activities are often in the shadows, almost a secret.
Now we can look into a more visual way with defining content operations which involves a lot of dots.
This is a way we’ve been explaining the need for content operations for a while, where each of these orange dots represents a publishing channel. So you could imagine they could be anything like your website, your app, your help desk, emails, a blog and so on. Essentially somewhere your organisation publishes content.
We’ve seen organisations that have up to a thousand CMSs alone, so that’s just websites. It doesn’t take into account all the social channels and so on. There are an ever-increasing number of publishing channels for organisations to utilise these days.
The scale of content that we’re talking about is so vast. IBM have 18 million help and support documents to manage. Organisations don’t realise how much content there is when you summarise it all together and this needs to really invest in managing it in a different way.
This is a pretty reserved ring of a few channels. And then come the humans. Each of these grey dots could represent a person or a team of people. It could be subject matter experts that are responsible for approving content or submitting some core content. It could be stakeholders that are involved in the creation of it or approval. It could be people that involved with the publishing and structuring of it. They could be interface designers or managers giving brief, legal teams or whatever sort of wildcard content producing characters exist in the organisation.
And again, a definite spectrum. Could be a few hundred, could be a few thousand, could be hundreds of thousands of people that are involved. But regardless of where you are, these people could be pretty dangerous, in the nicest possible way because I think often it’s how they interact when it comes to content, that it can end up looking a lot like a grey, spaghetti chaos.
Somewhere deep in the depths of this grey matter, there’s some kind of plan, or ideally some kind of content strategy. And it’s just sort of screaming to get out. It’s held back from your audience by this tangle of people, or of infrastructure and a process. This is the problem with bad content operations, or un-deliberate content operations. Which raises the question of what good, or what deliberate content operations would look like?
What does good ContentOps look like?
It looks beautiful. It’s looks quite organic. That neat, purple structure would be those deliberate processes, the people and the infrastructure inside the organisation connecting the strategy neatly to the audience, and to the delivery channels. A nice way to put the objective of this is to make the organisation more healthy.
We’re going to look at one example from higher education of how this might really look in your organisation, using one of these content ecosystem visualisations that we created.
Orange represents a publishing channel Purple is some form of infrastructure. That could be a Word document or a tool, like a marketing automation tool or your content management system. Blue is the actual content. Red is a process. That could be approval or structuring or content modelling or publishing.
The diagram above is from a university in the UK who are currently undergoing a big digital transformation project to improve the way that they produce and deliver effective content. This diagram represents the before as they’ve become more deliberate about content operations, and this is the process they’re looking to improve.
It looks pretty complicated and this is for only one content type, email marketing. What we were mapping out here was their CRM email production process.
This one starts with a marketing service manager, so a grey dot, having a plan for an email and suggesting some content.
The first thing is that he or she sends it off to the subject matter expert and to the marketing officer, so those two grey dots work together via email and they work on some content to start off the process of getting something together for an email that they’re going to publish. The marketing officer then goes away on their own, and they design the content for user needs, start thinking about personalisation and they start doing some initial drafts, these are all processes that they’re going through. They’re starting to produce something and they do this in Google Drive.
Google Drive emails are then put into Mailchimp, to structure them into HTML templates. So this process would be the email being designed up into a template. You’ve still got the marketing email and this is done by the marketing officer. The HTML is exported, so Mailchimp isn’t actually used to deliver the email, it’s just used to build the actual templates.
The problem is that theHTML and the CSS are two separate files. They have to go through this process of opening the file in Notepad, running it through a CSS inliner, then save it in Notepad as HTML. And they then move that file into Google Drive.
From here, content is pulled into their actual CRM. At this point, it’s purple because it’s not really a publishing tool here, it’s just being used to organise the content. Sometimes the HTML goes directly to here, depending on who’s using it.
The CRM technical admin formats the content of the marketing email and finally gets it into the CRM, where they define the audience and publish it. That’s impressive in terms of the amount of steps to go through, and it doesn’t cover the process of working with student ambassadors across email, using Google Drive to create videos, put them on YouTube and embedding them into web pages. Or the process of working with photographers that are commissioned to create photographs or media for these emails.
What it really illustrates is if you break down the amount of steps and processes involved in creating something as simple as an email, it can be a complex process.
From our research, we’ve summarised some of the things we’ve seen companies do that worked well, that enabled them to become more efficient with their ContentOps.
People, do we really need them?
Do we even need people? I think we probably do. The first thing we’ve seen people do, in terms of the best way to start out with this, when people join your organisation, is onboarding and training and it’s one of the best ways to instil a culture that content’s important.
Show, don’t tell, is a really nice way of putting the value of content to appeal to different people. Try to avoid the politics of navigating people and trying to convince them about this process change or the strategy. Just change something and then show them the results, show them how it saved you time or resulted in better quality content and had more of an impact.
Having clear owners for content, different processes and parts of the journey is important to clarify who’s meant to be doing what, and acknowledge any gaps or overlap that might exist. Overlap can be more dangerous.
Avoid jargon. The quickest way to get a plan rejected is making it incomprehensible by focusing on things like explaining the theory of content operations or content strategy to them. It’s probably best, again, to focus on plain language and get them to be aware of the outcome of any work.
You’ll need processes
The first thing with process is the idea of iteration. Getting carried away with doing too much means the first step becomes massive as people try to bite off so much. It’s an outcome of trying to imitate what other companies are doing and get there instantly.
Even if it’s to do with theoretical content production, like creating a structured content type for every piece of content in your organisation or modelling your entire process. There’s really no need to do your entire organisation at once. Start with one thing, like an email, and then prove that that works. Get the value to your customers and to your business and then move on to something else.
Breaking things down, like style guides, is essential too. We’ve seen a huge amount of success in some organisations when they’ve broken down big style guides like a huge PDF document. You don’t need to tell everyone everything. For example, you don’t have to make someone read your entire style guide before they write a staff profile. Just give them some guidelines for writing a staff profile.
A breakdown of the necessary infrastructure
One of the biggest considerations is infrastructure, the tools that you use to actually create the content. One of the biggest decisions is platform versus stack. So that would be whether you go more for a custom stack of connected tools like a headless CMS with a platform like GatherContent, or an authoring interface combined with content quality management tools like Acrolinx.
The stack option is far more flexible. It can be easily set up under a budget in terms of piece by piece. It can be extended iteratively in terms of piece by piece, so it sort of goes … in line with what I was saying about iterating it would be improving one thing at a time, one part of your infrastructure at a time.
Every organisation is different and everyone’s needs are different so it’s often better to have a set of custom tools versus a single platform. You ultimately need an authoring tool in some sense. You need somewhere to create and write content. You need an inventory, a central one, so everything’s in one place. You need asset management and you need some form of project management, a way to schedule things. Finally, you need publishing, an actual way to get your content to your audience. And you’ll need analytics and reporting, a way to measure the performance and impact of that content that you’re creating.
One of the most important things is to consider resources. What suits you? How many people have you got in your team that are able to do this stuff if you’ve got a very minimal development resources then it’s worth considering maybe something that works more quickly out of the box. And if you’ve got different needs, or you want to start small and do different things then consider where you’re going to start and what suits you.
That’s the three things we consider to be the components of content operations and hopefully illustrating why it’s important and why it’s real. It’s a lot harder than it looks when you’re on the inside of an organisation. It can involve a lot of work and a lot of change, and if we’re to succeed, we need to create content that is effective, that being structured and essentially having value to our audience or our organisation. And creating effective content, even at a small scale, involves organisational-wide changes to the way we work, which is also hard. But by being a bit more deliberate about content operations, it can be far easier to communicate effectively on the Web.