Digital Marketing & Communications

We've seen 1s and 0s you wouldn't believe

Tagged: content

Caption my caption

  , ,

📥  Show & Tell, Style, content and design



I'm glad to say that my cohorts here in Digital are – like me – sticklers for style. Of course we all dress well, but punctuation, grammar, and spelling rules are what we work by.

That's not to say there's never room for improvement, though. Iteration and incremental enhancement is key. To this end, the Image Use Guide has been updated to include a passage on best-practice caption writing.

'Why?' you may ask. Well, many of the images we were seeing published included captions that read like alt text. They did little to deliver information, and this is a missed chance at elevating engagement.

A student daining experience in a professional placement

The good: this caption adds context and info to a Placements page

Why captions are important

Almost every image on the website needs a caption. Simple as that. There are exceptions of course: hero images and background images which are a part of the page design don't need a caption. Similarly, caption writing for staff profiles and Location pages is tough. But be creative and you can add information like when a building was opened, how many Learning Commons it has, or its opening hours, for instance.

Captions need to be 'added extras' to the image. They need to tie the image to the content. Captions are entry points. Like crossheads, pull quotes, summaries and headlines. These all offer users a place at which to enter and engage with the content we offer.

Users look at imagery first, then the caption under the image. If the caption intrigues them by providing context and information, they will look back at the image and see something new. This is a positive loop.

Writing a caption that describes the image misses an engagement opportunity.

Writing a caption that describes the image misses an engagement opportunity.

Captions, not Alt Text

It's important to know the difference between image captions and alt text descriptions.

Alt text descriptions need to describe the image. This helps visually-impaired users with screen readers, and in instances where the image doesn't load.

"A caption should normally explain what readers cannot see for themselves in the picture (_President Karista appealing for the education bill at a teachers' conference in Washington on Wednesday_) and should omit the obvious (_A man licking an ice cream cone_)." – The New York Times Style Guide.

Captions need to elaborate and add to a piece of content. Captions must quickly tell the viewer what the picture itself cannot say – people, places, dates, significance.

The Content Publisher makes it easy to define a Caption and an Alt Text description, and gives an example as default.

Use the Alt Text field to describe the image. Use the Caption field to add a relevant captio

Remember the 'Five W's and an H'

Writing a caption is like writing a headline:


Use these prompts to add information otherwise missing from the image and you can't go far wrong.

Bad captions baffle the user and present a barrier to engaging further with content.

Bad captions baffle the user and present a barrier to engaging further with content.

Some dos and don'ts for best-practice caption writing:

Try to:
- use present tense to describe the image and any action
- keep your captions active (where possible)
- think in headlines – punchy and pithy is good
- explain the unknown and/or unshown - what's happening in the image that the reader can't see or know
- provide answers your user might have about the image: use the 'Five W's and an H'

Try and avoid:
- giving readers information they get from just looking at an image
- beginning with an indefinite or definite article e.g. 'she' 'it' 'a', 'an' or 'the'
- writing 'above' or 'pictured'
- repeating information contained in the title or summary

Some images on the website don't require captions. These might include:

- background images used in the page design
- banner images used at the top of a page
- logos

Finally, don't rely on bots to generate captions for you. They might be able to identify an image for you, but they remain pretty hit and miss... might claim to be able to recognise and caption an image for you, but it's still not as reliable as a human eye. might claim to be able to recognise and caption an image for you, but it's still not as reliable as a human eye.


Early review of research ethics content performance

  , ,

📥  Beta

In October, I worked on improving the information we provide about research integrity and ethics. To deliver the new section, I worked with subject matter experts in the Vice-Chancellor's Office and the Office of the University Secretary.

When we started, the content was basically a single page with multiple tabs, many many links and subheadings which were generic or duplicated. Together, we set out on a quest to make the process simpler and important tasks easier to spot and complete.

We shipped the new content in mid-November, so it's still early days. But we can already see from the analytics that we've made a huge improvement.

Making things simpler

Our biggest aim was to make it easier for users to understand what they need to do to conduct ethically responsible research. For this, we reworked parts of the original content into a plain English guide.

Plain English is about writing in a straightforward manner so that the content can be more easily understood by a wider audience.

Looking at the analytics, this guide is by far the most visited individual content item, with 517 pageviews compared to the next item with 344. It has an average time on page of 2:29. These stats suggest that it's being both found and read.

On that note, the fact that we can even start to draw comparisons between individual items of content is a huge improvement in itself. With the old page, all we had was basically an overall number of visitors to the whole page and our best guess.

Coherent user journey

One of the biggest changes is that the new Collection is making it easier for people to move on to the next step in completing their task. The bounce rate has gone down from 64% to 30%, which suggests that people are finding the content relevant.

The main purpose of both the old page and the new Collection is to point users to the relevant information. The change in the average time on page (old 4:32, new 1:15) suggests that it's a lot easier for people to find what they're looking for. This is supported by the fact that 80% of the people are moving onto another item of content compared to 63% on the old page.

There is also a big difference in what content they're moving onto to. From the new Collection, users go to the main items of content in the section - the top three items are two guides and our statement on ethics and integrity. From the old page, users were moving onto a more random set of content scattered across the website with no clear indication of any shared top tasks.

Celebrating success

So yes, it's still very early days. But after nearly three months since going live, the analytics would seem to suggest that users are finding it easier to navigate and read the content we’ve created.

In a transition project of this size, these little successes are worth noticing and shouting about. By sharing them, we are able to not only keep our stakeholders informed but provide a useful example for colleagues in the wider Higher Education community.

I compared data on the new content from the launch date to present to data on the old content from the same timespan one year earlier.