Lessons on readability and bias, and demonstrating the impact of human-centred approaches — Reflections from the UCD Gathering conference

Back in October, I had the opportunity to attend the UCD Gathering conference, a new virtual event for practitioners of user-centred design in all its forms.

Showcasing our research into staff and students’ needs around accessing course materials digitally

At UCD Gathering, I had the opportunity to present about our service design approach to improving student experience, showcasing our collaboration with the Learn Foundations project.

Blog post: A service design approach to Learn Foundations

The presentation was brought up to date with an explanation of how our work has helped schools move their teaching to a hybrid model in the wake of the coronavirus outbreak.

Three themes from ten talks

But better still than presenting about our work, was the opportunity to attend sessions from a wide variety of leading user-centred design practitioners.

UCD Gathering acted as a virtual amalgamation of a number of separate physical conferences, including UX Scotland and Service Design in Government, which I have attended a number of times.

Despite the virtual setting being new territory for everyone concerned, UCD Gathering had the same high quality of thought-provoking sessions that we have come to expect from the organisers.

I will highlight ten useful sessions around three themes:

Being aware of bias, and other cognitive considerations
Improving readability of content
Showing the business impact of human-centred approaches

Being aware of bias, and other cognitive considerations

Rebellious creativity — What kids have taught me about designing experiences

This session from Laura Yarrow was of particular interest to me, given that at the time I was anticipating the imminent birth of my daughter!

Adults are at the mercy of social judgement. So there are big differences between what kids and adults produce. Kids’ brains are better than adults’ brains for creativity.

If you set adults a creative task, they are constantly jockeying for power in a group. They are riddled with uncertainty, hesitation, inefficiency, and competition. Adults are preoccupied with status management.

Conversely, kids aren’t afraid to look stupid. They aren’t afraid to ask the right questions.

But education and business tends to reward quick answers over imaginative inquiry. Often, questioning is barely tolerated. This is why so often people start with solutions.

You might have heard about the five whys, the idea that if you keep asking “why” something is the way it is, you eventually get to the root cause of the problem. But kids don’t have five whys. They have 5,000 whys!

Adults also suffer from a sort of cognitive presenteeism. We believe we need to think really hard. But our best ideas often come when we let our minds wander — in the shower, on a run, or doing anything apart from thinking hard about it! Procrastination is a vice for productivity, but a virtue for creativity.

Kids are also more inclusive. They include people who are different. They don’t get stuck using just one tool or method. They avoid groupthink.

Finally, we should aim for volume, then quality. Mozart had over 600 pieces, and Edison had over 1,000 patents — but you only know a few of each.

Design for cognitive bias: Using mental shortcuts for good instead of evil

David Dylan Thomas is an expert on cognitive bias. There are over 100 biases. Knowing about these is important to making good human-centred decisions.

But it’s useful to note that biases aren’t inherently bad. In fact, they are a crucial set of shortcuts that help us get through the day. If we were to remove our biases, we would never get anything done. But that mechanism sometimes gets things wrong, which makes us cause harm.

He showed how this has an impact on how we design our content. One bias is cognitive fluency. This is when if something looks easy to read, we assume it’s going to be easy to do.

For example, a densely-packed recipe may make it look difficult to make pancakes. But using big images and small snippets of text makes it look easier.

We like things that are easier to process.

If something is easier to read, we think it’s more true.

If something rhymes, we think it’s more true.

We can think creatively about how we can inform readers how much stock they should put into a piece of information. For example, if something is a rumour, make it harder to read. Make facts easier to read.

David Dylan Thomas’s points about how bias informs our content design brings us neatly onto the next theme I want to highlight from the conference.

Making our content more readable

Helping users feel valued and supported

NHS Digital’s Sara Wilcox presented a practical but powerful session about the importance of “creating respectful, caring content” for people seeking healthcare advice.

She noted that some people have an anxiety that digital will remove human contact, removing the humanity in healthcare that people value so much. But the purpose of technology is to extend humanity, not replace it.

Her advice on creating accessible content in a healthcare context was striking.

NHS Digital’s content style guide advises that you should use words like pee and poo — because more people understand those words than urine and stools.

This attracts complaints from people who see it as dumbing down. But these complaints are not representative. In actual testing, NHS Digital have found that people appreciate the clear and direct language that everyone can understand.

Pee and poo and the language of health — blog post by Sara Wilcox for NHS Digital

Sara Wilcox also noted that we should be wary of using reading ages as a way of measuring the readability of content. It is possible to write gibberish and still get a decent readability score.

Reading ages are in fact meaningless for adults. The reading age methodology is based on research from the 1940s on Australian childrens’ vocabulary. It’s not very relevant for adults in the 2020s, and it only measures vocabulary, not actual readability.

This was the subject of an episode of the BBC Radio 4 programme More or Less. Hear the discussion from 20 minutes onwards.

More or Less — 7 February 2020

The most overlooked U-word in user experience: Understanding

Readability scores were also tackled by Candi Williams.

She noted that reading age scores are not an indicator of intelligence. Our reading ability can change based on our situation and emotional state. It’s really about speed and comprehension.

But she said that readability tests are a good way of checking ourselves and finding improvements.

It’s useful to consider that the age of 9 marks a significant milestone in our reading development. At this stage, we stop just reading, and we begin understanding. We begin to understand meaning, reading between the lines to pick up on things like irony and tone. With 9 year olds, you can begin to cover up words, and they are able to accurately guess the missing content.

This led us to some readability tests I hadn’t come across before.

The cloze test asks users to fill in the gaps from every 6th word. If people get 60% of the blanks right, it shows a good understanding of the meaning.

Interestingly, we can use this to understand what language people use to fill the blanks. So if you want to know whether you should say ordinary v normal, or cashpoint v cash machine, you can use a cloze test.

Another is the highlighter test, where you ask people to highlight which parts are clear, and which could be clearer.

Showing business impact

The material of outcomes

Linn Vizard capped day one with a thought-provoking session about how we understand outcomes of service design work. This is a topic that has been on my mind a fair bit recently.

Linn Vizard hypothesises that designers’ ability to “make something out of nothing” gives them a false sense of control. But, she said, “it was never about the thing.” There is no inherent value in products. Their value comes from their use and exchange.

This idea comes from a concept from economics called service-dominant logic. This shows that all economic activity is ultimately driven by services, not products (Mauricio Manhães 2018, This is Service Design Doing, pp. 29–31).

One way of understanding this is through different types of chairs. A chair could be an ergonomic support to help you work in your office. A different chair enables you to eat your dinner at the table. A step stool helps you reach the top shelf. Some chairs are primarily status symbols. Wheelchairs help people be more mobile. Other chairs are designed as works of art.

Each of them are chairs. But they all offer people different services, so their value is derived in different ways.

Value is co-created through exchange. For Linn Vizard, outcomes are co-created value that occurs throughout the service design process. She described it as “total co-creation of value”.

How do we know that service design is working? Good outcomes might be changes to what people think and feel. We want to demonstrate a change in people’s feeling over time.

Outcomes like this of service design work are often intangible. So, says Linn Vizard, designers need to let go of their attachment to control.

Designing decision making to deliver better experiences

If the idea of losing control of the outcomes of our work seems scary, a pragmatic approach to demonstrating impact was outlined by the AA’s Angela Arnold the following day.

Like Linn Vizard, she noted that measurements cease to be useful once they become a target. Organisations often default to using metrics that are readily available — things like traffic, conversion and velocity.

If you measure traffic, you create more content.
If you measure conversion, you stick banner ads all over the place.
If you measure velocity, you just build stuff more quickly.

These all amount to a bad user experience. So we have to change the way we measure things.

This is a problem because quantitative measures are often not attributable to everyday design decisions. Meanwhile, qualitative approaches are good for understanding design decisions, but often not attributable to shifting business metrics.

Angela Arnold said we need to start correlating our metrics. So instead of having random sets of metrics flying around, start to tell stories in a more connected way.

Our job as designers is to facilitate better decision making. This is key because some people don’t like user experience, and don’t like words like “discovery”, because it sounds intangible and it sounds like we’re not going to deliver something. When you’re working in a business that delivers outcomes, you need to help teams focus on delivering the right thing and delivering value.

Designing decision making to deliver better experiences — session synopsis

Design and strategy — Why and how to mind other people’s business

This leads us on to the session from Morgane Peng of Societe Generale. She talked about how different types of stakeholders can be blockers to good design.

While it can be tempting to point the finger at stakeholders who don’t “get” design, Morgane Peng pointed out that it’s important to realise it’s not their fault. The root causes are down to issues around culture, the organisation and business.

Among her tips to tackle these issues:

Ask our stakeholders questions
Highlight good examples from other departments to create a “fear of missing out”
Clarify methodologies and explain procedure
Speak the organisation’s language, by “hacking” your explanations (for example if your organisation is risk averse, talk about how your approach helps avoid risk)
Present ourselves as providing a business solution, not just “pushing pixels”

A blog post and video version of this talk is available — I highly recommend it:

Scaling UX impact despite dwindling resources

Similar themes emerged from Roy Opata Olende’s session. He spoke about his experience of demonstrating the impact of user experience when he worked at Buffer. He said that fewer resources doesn’t have to mean less impact.

He highlighted the importance of allies. They don’t just come from nowhere. We need to cultivate these relationships. Who can help you achieve your goals, and how can you work with them to further your progress.

The main thing on senior managers’ minds is not user experience. It’s improved business position. So talk in terms of business goals.

Easy documentation for user research

Kyree Holmes from Spotify shared her tips for giving research a voice in the decision-making process. She says we need to consider the user experience of user experience — a thought I’ve often had. She said: “We have formulas or an established way. But how often do we evaluate that?”

She shared a number of creative and engaging ways of presenting research. She said thinking creatively about how we bring our documentation to stakeholders helps make it more sustainable, and builds in meaningful sense-checks to monitor effectiveness.

Easy documentation for user research — session synopsis

Conclusions

In summary:

Don’t be afraid to ask questions
Do the hard work to avoid harm
Do the hard work to make content easy to read
Be creative about demonstrating impact
Speak the language of your organisation
Do the hard work to make your processes inclusive