My friend worked in documentation for a very large software company for many years. Every few months or so he would update a single product warning that would take him 3 days to update because the warning was in 14 places (pdfs, web help, etc.) 14 countries (print documentation for example) and 14 different places in the software (mobile app for example) plus training and internal knowledge base for internal support personnel. By the time he had finished revising the document someone would say that the document had already been revised.
Yes, this sort of waste is not normal and is very common. It has even become a genre.
So even if you have “structured content”, this does not automatically mean that you are dealing with multichannel delivery as well. A workflow to publish the structured content to the required output channels is needed as well to manage the required output formats. So the next question is: How do you manage the needed output formats in order to publish your content in all channels?
What “structured” actually means in practice
The content is structured so it can be published to many channels. But it does not matter how many channels there are: one HTML page or a hundred PDFs, a website with a lot of pages or a mobile application. Structured content is very suitable for this kind of multi-channel publishing. But structured content is not the same as multi-channel publishing. Structured content is prepared content. But it still has to be published somewhere. And that is not easy.
The content then flows or is piped down the structure or the metadata attached to the words. This then means that there is no need to reformat the content in order to re-purpose it for different channels. This is also why it is stated that changing the pipes changes the flow of the content. Also changing the pressure of the pipes changes the flow of the content. And finally changing the faucets at the ends of the pipes changes the way the content is published.
Structured Content is very important in Multichannel publishing. Content that is written in a structured fashion and has been authored with a solid component-based content model, allows authors to focus on the writing rather than on reformatting the same information for distribution to different outputs. For example, when publishing a document, all of the steps to complete a particular process might be written and stored as individual components or as parts of a topic, which can then automatically be reformatting to create a set of tutorial steps that are published on a company’s intranet,
As was previously explained, the output format of the structured content also varies, depending on the channel. The HTML and responsive layout of a web portal are typically the preferred form of content presentation. PDFs are often fixed page with specific typography. In addition, there are the various types of mobile apps, that may require their content to be delivered in the form of very light weight JSON. In some cases even in the form of processed ‘snippets’ of content, in the form of API calls. In addition to all of the above, Learning Management Systems, need the content to be packaged up in the form of a SCORM package, or in some other format supported by the specific LMS.
So how does one manage to get content to all of these channels without creating a whole new set of processes for publishing to each of them?
So to summarize: publish from a single source and let your toolchain transform the content to the required output format. This requires a few things to be true, though.
- Your content is genuinely modular, meaning topics or components can stand alone without relying on surrounding context
- Metadata is applied consistently so filters and conditions can do their job automatically
- Your team has agreed on what “done” looks like for each channel, not just the master document
- Output templates or stylesheets exist for each target format before you need them urgently
But again, there is one element of content planning that is virtually impossible to plan for in advance of creating the content itself. The Template. Well, I suppose this is one area of content planning that can be done well in advance. I mean, who hasn’t seen a neat and shiny template delivered to a writer or author the night before to create content against? Ah yes. There’s nothing quite like delivering a pre-designed template and then demanding that it be completed on time against a tight schedule, is there? Nope. Not for me. Thank you very much.
How a real multichannel workflow actually looks
Let’s walk through the process of a software release and how the release notes for that release get published to 4 different destinations: 1) A public documentation site 2) An in-app tooltip system 3) A printed quick-start guide 4) An internal knowledge base for employees. All of this content was created by the content team as structured content, as topic-level components, and used conditional tags for both the audience and the output type. The publishing pipeline for this content would look something like this:
- Authors write and tag content in the component content management system (CCMS)
- Conditions filter out what each channel doesn’t need (internal notes stay private, tooltip content strips out lengthy paragraphs that would never render properly anyway)
- Automated publishing pipelines generate HTML5, PDF, and structured data exports from the same source
- Each output is pushed to its destination, whether that’s a web server, a delivery API, or a print vendor
For teams who produce high volumes of documentation to be distributed across multiple platforms, tools like MadCap Syndicate help to make the process of centralized and allow for much more control over how content is distributed without having to create a separate source for each distribution channel.
(And keep in mind that the nicest plumbing analogy only goes so far. If you’re putting really dirty or really poorly structured content into your system, you’re going to get dirty or poorly structured output, regardless of how elegant your pipes are. And that’s why it’s so, so important to make sure you’re putting really great, structured content into your system in the first place.)
Metadata is doing more work than you think
A good content model, based on a well-structured set of topics, which have been very thoroughly tagged for conditions, for different audiences, for different purposes, can automatically publish 12 different channels of content from 30 topics of content. A content model that has not been well tagged for such conditions and attributes will publish 12 different channels of content, but each will have to be written and published by a separate writer, in a separate editing project. The writer’s life will be a poor thing if such is the case.
| Channel | Typical format | Key structural requirement |
| Web help portal | HTML5 / responsive | Topic-based navigation, search indexing |
| PDF / print | Fixed-page layout | Page sequencing, print stylesheets |
| Mobile / in-app | JSON or lightweight HTML | Short-form content, no embedded images |
| LMS / eLearning | SCORM, xAPI | Assessments, tracking hooks, module structure |
A word on governance, the part nobody wants to talk about
Most organizations have not established sufficient governance around the creation, tagging and approval of content before it is published. This means that all their money spent on tools and templates is essentially being used to create a very efficient garbage generation system.
So although the structure may be established in theory for generating content across multiple channels, the absence of good governance in practice will bring a publishing team’s work to a halt.
Yes, content is a large part of delivering information in multiple ways. But, the content is only one aspect of delivering information. The information architecture (IA) to support the publication of the information to the various channels is a much greater challenge. And, as before, the less glamorous governance of information is critical to supporting a scalable publishing system.
Now do that well and you can publish from a single source. This means that if your structure and all your metadata are in place then the rest of your work of publishing to different channels will become so much easier than it is today. The pipeline will do the hard work for you to turn your well-structured blocks of content into whatever is required for the channels you need to publish to.
