The way we access content is changing. Our content needs are not limited to web pages anymore. We’re also interacting with content via apps, searches and voice technology.
As our content needs change, we need to be smarter about how we produce content to meet these needs and so, we've started looking into the structured content model.
What is structured content?
Structured content means separating the content you’d put onto a web page into smaller parts or components, treating it as data.
These separated components can then be re-used and organised in a predictable way as they are not tied at the page level, and they can be stored separately.
Components can then be classified by a set of pre-determined metadata. This metadata, in turn, can be easily processed by algorithms and robots, making the content findable on search engines such as Google, and accessible to voice technology like Alexa.
There are many advantages to using structured content, including:
- it’s easier to maintain — you only have to update once to change the content on all the channels it’s published on
- writers know what information they need to provide
- it’s easier to translate
- it allows consistency of information
- we can tag content components for search and voice, using markups
- robot algorithms work with the data.
The structured content model
Just over 11 years ago, Nathan Wall came up with an idea about breaking the content down into parts that would help us create and manage it.
We were both working in the web team at Inland Revenue and he wanted a way to ensure that content was structured consistently, without gaps.
Nathan included me in a project to try to work out exactly how we could break up content, and how we could then use it to define and create better content. We’ve both been refining and using the model since then, both together and independently.
This work provided the foundation for the structured content model in our programme.
Different influences for content and duplication
When we create content to suit our users’ needs we need to consider the following influences:
- user attributes — who are the users of this content and what are their circumstances?
- operating environment — this includes legislative and policy requirements
- interaction channels — web, phone, face-to-face, etc
- customer activity — the user journey through the information.
It’s important to note that we can have multiple pages of content to describe a service or a process, depending on the user attributes or other influences. These pages will usually have some duplicate information, so we want to avoid creating it on every page. Instead, we can re-use a component containing the information and deploy it on the relevant pages.
When someone is applying for NZ Super, what they need to do will depend on if the user is, or has been, a previous client of the Ministry of Social Development.
This means each client circumstance has its own page on the Work and Income website. There is some duplicated information on each page, so it would make sense to create common components once, then re-use as required.
What we're working on
We’ve been working with writers and creators within DIA and beyond to test that the model will meet our structured content needs. This work has identified what we need to work on going forward.
Currently, the model helps identify what content is needed, but we need to do a lot more thinking about how that content is presented.
Co-design the model with us
We’ll be looking for volunteers shortly to help us with the next stage of this work. We want to co-design the model with stakeholders from agencies and vendors so that it meets all our needs going forward.
If you’re interested in helping us test what we’re doing, please get in touch at firstname.lastname@example.org
This work is commissioned by Government Information Services at the Department of Internal Affairs. The programme team is based at the Service Innovation Lab.