Takeaways from UX Scotland 2019

1st takeaway: “The current complex (digital) environment calls for a Debbie Downer”

Cheryl Platz told us her terrifying story about her encounter with an elephant during an African safari. The reason for this encounter? Unchecked optimism. During her walk, she constantly was thinking about “happy paths”, and not about how things could actually go wrong. What does this have to do with UX design you might ask? Well, her answer was clear: EVERYTHING!

According to Cheryl, we all are addicted to impact with disruptions, transformations and paradigm shifts hitting us in the face everywhere. This impact, however, brings an increased complexity with it. And how do we counter this complexity? By being opti-pessimistic.

Don’t know how to be opti-pessimistic? Don’t worry. She gave 4 rules that you should follow:

1. Consider the human context
2. Design for the best case
3. Plan for the worst case
4. Be ready to adapt in the moment

But how can we put these rules into practice? Let’s use Amazon’s Alexa to illustrate this.

Rule 1: Consider the human context

When it comes to any app, website, or in this case Alexa skill, you should not ask yourself if you can build it, but whether if you should build it.
An example a completely useless Alexa skill is “Fidget spinner”. It just tells you if an imaginary fidget spinner is spinning. Another example is the ability to shopping online. Apparently, the reality is that only 2% of the people who use Amazon’s Alexa have made a purchase by voice. And the once that do make a purchase, often just do it once. If more contextual research would have been done about the interaction of people, they probably would have seen that. Meanwhile ordering stuff via voice has been a huge part of the abilities of Alexa. Time, money and effort that could have been better spent.

Rule 2: Design for the best case

When Alexa was built, it probably was built with adults in mind. Kids however quickly gravitated towards Alexa to be able to play songs and ask questions. This led to some hilarious cases such as a child asking Alexa to play a song. He gets a list of porn channels instead.

If they would have thought about the best case – children using Alexa – they would have immediately thought about a kid-friendly mode which they implemented later on.

Rule 3: Plan for the worst case

Let’s think about buying stuff online via Alexa. This has gone seriously wrong in some cases. The case of a girl buying a dollhouse with her parents Amazon Echo for example. She ordered the dollhouse without her parents knowing – could have been avoided with a kid-friendly mode. After her parents realized this, they added a code for future purchases. Apparently, this wasn’t set by default. But it gets even worse! This news was broadcasted on San Diego’s CW6 News, which in its turn triggered Alexa in households all over San Diego watching the news. If Amazon would have planned for this worst-case scenario, they would have known that a user should probably set a code for future purchases by default.

Rule 4: Be ready to adapt in the moment

It’s always possible that you can’t predict everything, or that there are uncertain elements in place. But you do need to be able to track and tackle the problem.
One of the problems that Alexa had as mentioned earlier, was the lack of a kid-friendly mode. Now, they have created FreeTime integration with which they can set time limits to Alexa, disable voice shopping, change the personality et cetera. They adapted to a plethora of worst-case scenarios by tracking issues and providing budget & time to solve them.

2nd takeaway: “In conversational design we not only need to think about what we say, but also how we say it.”

In real-time conversations, we use a combination of body language, tone and what a person says to deduct the real meaning of the conversation. In the current graphical interfaces, however, a lot of these elements are lost along the way. With conversational design such as chatbots and voice interaction however, we can’t risk it to lose these cues.
We need to be able to conversate using these technologies and not only think about what the program needs to say, but also how it should be said.
And this is why we have to take into account the following two elements in conversation design: voice and tone.
But what’s the difference?

The voice represents the character of the brand or bot. It’s the guideline for all conversations that are held and should be consistent throughout every interaction. You can compare it with the characters of your parents and how they always act. Maybe they are always caring, sweet and thoughtful in each of their interactions.
Tone, on the other hand, is the expression of that voice that changes with time and the people we interact with. Think about yourself and how different your communication is between you and your parents or you and a job interviewer. That’s tone. You are still the same person with the same voice, how you communicate, depends on the situation and people you communicate with.

If you want to reap the full benefit of conversational interactions, voice and tone are the building blocks to do it.

3rd takeaway: “Revival of the old Wizard of Oz technique”

For all designers out there: do you remember the Wizard of Oz testing approach? And if so, when was the last time you used it? The answer will probably be: “I used it a long, long time ago”. Well here’s some news for you: there’s a revival of the all known wizard of Oz Technique for testing voice interaction. For the people that don’t know the Wizard of Oz technique, it’s a technique in which a user thinks that he is interacting with the end system. In reality, there’s a person “behind the curtain”. He or she pushes all the buttons to mimic the interaction that you would have had if the system was in place.

Why use the Wizard of Oz technique you might ask? Well, there are a number of reasons:

• You don’t need that many technical skills to create a prototype to test.
• It’s really easy to create a prototype.
• You can iterate very quickly.

While using this technique, however, you need to make sure that you don’t kiss and tell. Make sure that you don’t use words in your user tasks that may implicate the keywords you want the user to use.

4th takeaway: “Emotions are the key to brands breaking away from the pack”

What is the last interaction with a brand that you remember? And how did you feel at the moment? You probably felt happy or enraged. Well, these questions show you that the last reaction that you remember probably was connected to a strong emotion. And this was exactly the theme of Liraz Margalit’s talk.
She showed us that your memory is influenced by emotions. If an experience is emotional enough, it will end up in our long term memory.

How can we brands make use of these emotions to improve the brand experience? The key to this are the 5 behavior patterns that she discovered:

• Disoriented: no interest / insufficient motivation to keep exploring page/ content
• Lack of interest: lost sense of direction, position or relationship with page
• Exploring: zero-in on options by investigating options page has to offer
• Mindful: Pay attention to content on page deeply engaged
• Focused: Focused, know what they want; pay less attention to the page; no emotions are involved

These patterns reflect the emotional state in which the user is and can be leveraged to improve the brand experience.

5th and last takeaway: “Think about the end.”