Here we go folks! It’s testbash Germany and we’re starting the day with Cassandra. She’s come onto the stage like a superstar (she says buffoonery but I say awesome).
The topic is personas and UX, on the topic of the seven dwarfs. Cassandra is starting out by introducing herself as a tester, UX enthusiast, blogger, speaker, intersectional feminist (yeah!) and Ministry of Test fangirl. Also, there’s gonna be swearing. I’m fine with that.
We’re starting off with some definitions – personas are fictional characters to represent user types. They’re used by marketers typically. Traditional personas contain information about personal backgrounds and other things (sorry, that slide went by fast).
One important thing is that personas are not the same as roles. That’s something that gets blurred frequently. Roles are more about permissions. Personas are describing users on a personal level – how technically proficient they are, whether they use a screen reader. They help us to figure out how a user might respond to a feature.
I’m laughing inside (oh now outside as well) because the first persona example is a typical white male : Cassandra says “cos obviously”. It’s got information in it about how much he earns, his family, likes and so on. But it doesn’t help us to really know him. Cassandra translates this to him being e.g. near retirement, having no dependents, having disposable income. That might be more useful to us.
Using her own persona, she’s showing the example again.
Unfortunately, Cassandra says, traditional personas used by marketeers aren’t that useful for testers. There’s a lot of interpretation involved, they focus on traditional demographics and they assume that a persona is fixed. Basically, we still have to make up a lot of details. And – they stop being representative because they become too generalised.
Cassandra proposes having a different core. In the centre of the traditional personas are the demographics. When Cassandra used to work in support, they had personality as the core of their personas. What kind of communication style / level of detail do they prefer? Some people are bubbly, want to build rapport, don’t need so much detail (oh goddess it’s me). Other people need more detail, want to know what’s going on at every stage and might be less interested in building rapport.
For testing, Cassandra suggests that we put mental state at the core. Then we can talk about behaviours they might have. We can talk about their situation (I’m translating this to context in my mind) – how are they in the mental state they are in? From that we can talk about their priorities and the impact they want to have.
Cassandra is going to show this using the analogy of the seven dwarves (with appropriate disclaimers about the name Dopey). These personas focus on the user’s mental states, and assume / appreciate that users are complex and can’t be represented by a single persona.
Starting with Happy – which Cassandra also calls “the temporary one” – she’S content, open minded and forgiving, and has no expectations. But the mood or mental state could change at any time. Her situation might be that she’s playing games on holidays and her priorities might be a nice design / fun to use. There’s low impact for her in this mental state.
Doc is the picky one. She’s suspicious and harder to please, oriented to details. She prefers specificity over ambiguity. Her situation might be that she’s learned about privacy and is worried about it. She prioritises accurate information. The personal impact might be mistrust (and therefore using another service). If she can’t find any provider, she might stick with something that she’s not actually happy with, despite specifics being very important to her. That’s not a good place to be in for a user.
Sneezy is up next. Cassandra sees herself here – distracted, short attention span, doesn’t like long workflows. Is most likely to say “sorry, did you ask me something?”. In a situation where she’s planning a trip, she’s got a lot of tabs open. She needs quick simple workflows and enough time to book things while she’s drifting between topics. In terms of impact – she’s likely to end up with an incomplete booking or booking the wrong thing. If the session times out – she’s going to be annoyed (I feel this in my soul….).
Dopey is the confused one. He takes comfort in familiar patterns, seeks information and guides. Nevertheless, he’s prone to error and might make mistakes. If Dopey is using internal systems at a new company, he’s going to prioritise clear instructions and warnings. The potential impact is that he might take a wrong action, a customer might complain and his job might be at risk. We get quickly from a bad user interface to someone being worried about their job.
This is a pattern through the industry. Products that aren’t customer facing often lack UX. But people with jobs and priorities and stakes are using these. A recent example was the Hawaiian missile mistake. She’s showing the UI – and it’s awful! No wonder mistakes were made!
On to sleepy. She’s exhausted. On the edge, stressed and has a high cognitive load. Sleepy just doesn’t have time for this crap right now. If Sleepy is in a situation where she’s travelling across time zones, she needs to keep track of meetings etc. A priority for her is self explanatory features. If they’re not, the impact could be that time zones aren’t clear and flights get missed (this is an actual stress factor for me!!). It’s also happened to me that appointments are saved in the wrong time zone. Just like Sleepy.
What I really love is how Cassandra is joining “small software errors” to actual real stress factors for real humans using these things. I approve!
Bashful is the anxious one. Shy, reserved and nervous. Willingness to seek help can vary. Bashful might have just moved to another country and might have to use lots of websites in another language. What will help here is localisation and images to help understand text. If that doesn’t happen, then Bashful’s attempts to research tax law (yurk, I’ve been there) might backfire and they could end up overpaying or underpaying. Another great point – if text is saved as an image and Bashful can’t copy it to put it through google translate, then she might give up. If that’s for social events, then Bashful doesn’T integrate.
The final dwarf is Grumpy. Device shaking, angry and impatient. He says “why is everyone so incompetent?”. Grumpy’s situation might be that he’s having an awful day. His priority is reliability. If that isn’t given and he loses data, or if things take too long, then he’s going to waste time.
(Looking at all of these dwarves, I’m really realising how much my mental state on any given day (plus my base values / beliefs etc) really affects how I not only interact with humans, but also with software. This is worth reflecting on – for our own sanity and for our users. Personal story – I realised that bad internet connections on trains make me angry because I feel like I’m losing productivity. Maybe I need to relax more and accept I’m not just my productivity? . Anyway. Back from the tangent).
Another suggestion is to consider which personas you don’t want using your app (hackers, bad users etc.)
To wrap up, Cassandra is looking at some objections to those models.
- “We need to focus on users that make us money”
- “Those are just edge cases”
Her answers are that a traditional persona could embody every single dwarf persona at a different point. Also, people are complex and shouldn’t be reduced to a single persona. In terms of edge cases – all Cassandra’s examples are real from her life. They’re not ideal situations, but they are common.
Cassandra is finishing by suggesting that we break away from traditional demographics, to base personas on real people and to perform tours as each persona. Think about how your product might bring out specific mental states (personas) in users. Also – take this and adapt it! Final call from Cassandra is to diversify your personas. Use different body types, genders, non-binary people, …
Awesome talk! I need to process this and reflect for myself too!