From late summer 2018 the work and planning around our new website gathers pace. A supplier and integration partner for the new CMS will be announced in the coming days. This is an incredibly exciting moment as it means we can now push forward and begin to firm up on our plans.
One of the biggest challenges we face is how we go about migrating web content from the current to the new website. You could say this is a bit like moving home and, like every house move, if we want things to go smoothly we have to spend a bit of time preparing and planning to avoid any unexpected surprises.
Where to start?
With a website of 90K pages indexed by Google (we’ve managed to reduce this from 540K 😮), the migration process could be daunting to say the least. An obvious place to start might be the content that delivers the most value to our users whilst also playing a key role in meeting the University’s business objectives. If we follow this logic there’s a strong argument for migrating either staff profiles or course pages first. Both rank highly in our top tasks survey and play a critical part in student recruitment/income generation and promoting the teaching and research activities of our staff. From our design sprints with staff and students we also know that people like to think of courses and profiles as being the common denominator which connects all our web content together.
Whether it’s courses, staff profiles, or something else that we tackle first, content migration should be as much about improving content and producing results as it is about just moving stuff. After all, we’re not simply replicating the current website (with all its flaws) in a new CMS and with a new design.
Where to put it?
If you allow me to labour the metaphor a bit further, when we move house it forces us to think about where our stuff belongs. Sometimes that’s fairly obvious. For example, the cheese grater goes in the kitchen not the shower (unless you’re a bit weird). So we box it up with a load of other kitchen paraphernalia and label it ‘kitchen’.
In the case of web content, where something belongs is often a much more difficult question to answer than where to put our grater. In the past we’ve grouped items of content together to reflect the structure on the University. This approach produces a website based on silos; we miss the opportunity to make connections between different types of content across the University and the user experience is often characterised by confusion and frustration.
A better starting point for deciding where content belongs is an understanding of the user need it’s meant to fulfil rather than how it relates to our internal structure. Once we know this we can begin to group related items of content together – or even better ask users how they think these items should be grouped.
For example, our research with users has shown that they see ‘Campus’ as a natural grouping for top tasks like getting details about campus facilities, opening hours for buildings, room bookings, and travel information.
Similarly, users expressed a preference for top tasks like paying tuition fees, ordering a new student ID card, applying for a visa, and registering with a doctor to be grouped under ‘Support’.
There’s still work to be done defining our site navigation and architecture but the evidence and best practice suggests that when someone reaches a page we should strip away the things that are unconnected to the task they are trying to perform. Web guru, Gerry McGovern hit the nail on the head when he spoke about this at the Event Apart 2018 conference:
“When someone is on a page, trust that they want to be on that page. Trust that the person is in the place they want to be. And then it’s your job to keep them moving forward. With Amazon, when you choose musical instruments, the navigation changes and is reduced to only things related to that choice. If you want to simplify, you must strip away everything that is not connected to the task at hand. Also, be sure to establish clearly and immediately where they are.”
There’s another factor that comes into play when we migrate content and it relates to quality.
Going back to our house move and our much-cherished cheese grater, let’s suppose you decide it’s looking a bit tired and has grated one wedge of parmesan too many. Putting sentimental attachment to one side, you make the executive decision to buy a beautiful, new grater. This means there’s less chance of anything diminishing your experience when you move into your lovely new house. Like moving our possessions, migrating web content from the old to the new website will give us the opportunity to address the issue of whether it’s fit for purpose.
As our website has grown over the last 25 years or so, in a very haphazard and unpredictable way, parts of it which worked perfectly fine a few years ago now struggle to meet present day standards. Moving substandard content into the new website would be self-defeating and something of a missed opportunity. We need to have clear understanding of our content quality standards before we begin moving it. This is where our wonderful content style guide comes into its own, helping us make informed decisions and consistent approach to things like voice, tone, writing style, and commonly used terms.
Defining content types
As we start the process of editing and reviewing content for migration we should begin to see patterns and similarities. These patterns allow us to produce content types. Think of content types as predefined structures or templates in the CMS which break our content down into chunks or fields.
Structuring our content has many benefits. Firstly, it prompts us once more to think about purpose (why is it there and what user need is it addressing?).
Secondly, it gives users a consistent experience when reading similar types of content on the website. This builds reassurance and trust, no bad thing for our brand.
Lastly, by breaking content down into small parts we increase its potential to be reused in different ways. For example, we might have a content type called ‘student profile’ which we chunk up into various fields such as name, country, course, and quote. As well as displaying the quote on the profile page we can be really clever and display it automatically on the appropriate course page. This makes our content work harder for us and move aways from the idea of content belonging in only one place to a model where chunks of content can be published across multiple locations.
Structuring content also opens up the possibility for it to be read more easily by search engines and voice recognition systems like Alexa and Siri. Through structuring content we’re future proofing it and making it more robust and flexible.
Getting our content ready
Over the last two years we’ve made a concerted effort to improve the quality and structure of our content as we’ve launched new versions of school, subject, department, and service websites so these pages are easier to migrate when the new CMS is ready. Below are just a few examples how we’ve given structure to our content.
Examples of content types
|Case study||Telling a story about an aspect of our research, student experience, staff activity, and industry engagement.||The implications of EU withdrawal
|Guide||Explaining how to use one of our services or perform a task.||Print from own laptop
|Facility||Providing supporting information for a place, amenity, or equipment that supports student and staff activities.||General Workshop|
|Publication||Providing links to University documents and associated metadata and supporting information.||Annual school learning and teaching enhancement report template and guidance|
|Staff profile||Displaying contact details and profiles of university support and academic staff. Listing academic staff research outputs and publications.||Jacques Hartmann|
|Corporate information||Displaying policies, regulations, procedures, and other information relating to the governance of the University.||In development|
|Project||Demonstrating the aims, phases, and outputs of research and campus projects.||Zinc House|
Let it go
Like our cheese grater, we should also be prepared to admit that some content has had its day and, for the grater good (sorry), it’s best to consign it to the bin. It would only be an unwanted and unused burden if it were moved to a new location and could potentially hinder access to better content. Given the time and energy that may have been invested in creating content it’s understandable that some people might be reluctant to remove it, even if it is past its sell-by date. Difficult but constructive discussions are needed perhaps in this situation. We can also use the persuasive powers of data and user testing to help content owners get a wider view of how a single item of content fits into a user’s journey and, together, we can make decisions based on evidence.