Yesterday I attended day two of the Google UX Masterclass. Here are my notes on things that stuck out for me.
- Always over-recruit a little. 1 in 5 participants is often not quite right: they don’t match the persona you want to test.
- Rapid Iteration Testing can be useful: tweaking the prototype a little after each test.
- Different age groups can have very different expectations. (This seems obvious, but can be easy to forget about.)
Launch and beyond
- “Err towards the cupcake.” Go for a smaller, simpler, thing first. (Related to Adriaan’s excellent talk from the recent UX SA conference.)
- There was some crossover on topics between UX and Product Management, especially for Product Discovery.
- User surveys can be useful. Provide a feedback link. Make it quick and easy for user to provide feedback, and say thank you when they do. (It’s worth bearing in mind what Phil Barrett from Flow said in his keynote at the UX SA conference, though. You’ll probably only get feedback at the far ends of the spectrum: very engaged users who are either very happy or very angry.)
From day one:
- Update our existing Personas (and Customer Journey Map). Print them out nice and big and stick them up on the wall.
- Set up regular, small, usability tests. (Although this is in my To Do list already)
From day two:
- Review the first touch experience and make notes on where it can be improved, especially the registration bit.
- Start tracking stats on usage: see what bits of the app people are using, and what bits they’re not.
- Review the text of the app, especially CTAs and section names.
Yesterday I attended day one of the Google UX Masterclass. Here are some of the things that stood out for me.
The UX process
The introductory presentation covered UX and UCD.
- Personas can help drive the focus to a User Centered Design process. We did an exercise where the same brief gave very different results when designed for different personas.
- You can use UX principles to validate your assumptions, and to bring you back to simplicity. It can help you establish a clear story for your product.
- User Centered Design is an ongoing, repeating, process. Your customer journey map is an important artefact: refer back to it frequently to help keep you on course.
- Paper prototyping is great for idea generation. (Since there are no interactions, though, it can suffer in usability testing because it relies on what the users say they would do rather than what they actually do.)
Flow’s case studies
Philip Langley and Chris Metcalfe from Flow interactive gave a(n excellent) short talk discussing some projects they have worked on, and the UX processes and tools they’ve used.
- Search trumps navigation on mobile. (I’m interesting in seeing where this line is crossed: how about tablets? Where do users draw the line, and how is that line moving?)
- When doing an affinity sorting exercise, it can very useful to bring users back into your thinking. Rather than just labelling groups with a word, add “People want” or “People struggled with” to the front. Then add three “because” statements to each group. You can refer back to these later to check a proposed solution or idea.
- “Above the fold” still matters for feature phones and BlackBerry devices because scrolling on them is hard.
- Chris and Phil both stressed the importance of a cross-functional team that worked in close collaboration throughout the process of designing, building, and shaping a product. (One of the reasons I was sad to leave Flow was that we were starting to get this right.)
Day two at the conference was a little quieter, but the crowd was just as energised. Here are my notes from the second day of talks: a few things that caught my attention. For each session below, I’ve tried to find one SMART To Do item.
The Elephant and the Dassie
presented by Kerry-Anne Gilowey (Slides; Twitter: kerry_anne)
- It’s important to break down the silos between roles in our teams and organisations. The edges of our roles and responsibilities are already blurry. This includes the client, who should be a collaborator in the process.
- Lenses > Roles. Even on problems that aren’t specifically part of your role, your lens can bring new and different insights to it: shifts the prominence of things in your view. Don’t be afraid to use tools from other roles.
- The focus of our work is changing from the organisational structure to the customers’s cross-channel journey.
- Kerry-Anne promoted openness and sharing. She encouraged attending, sharing, and speaking at meet ups. It doesn’t have to be a huge thing: a six person get together over dinner can work really well.
Typography and improving user experience
presented by Justin Slack (Slides; Twitter: urbanrenewal)
- We should care about typography because so much of the web is about reading.
- Optimising typography optimises many things: readability, accessibility, usability, overall graphic balance.
- Typography should be involved as early in the design process as possible: settle on a typeface as early as possible; test type in the browser as early as possible; organise your design with readability in mind.
- Review the typography of sites I work on, with an eye for readability. Update the typographic scale.
Product Management, Cognitive Bias and You
presented by Carlo Kruger (Slides, Twitter: ironicbuddha
- Carlo talked about the human brain, and System 1 (fast, instinctive and emotional) and System 2 (slow, deliberative, logical) thinking from Daniel Kahneman’s book Thinking, Fast and Slow.
- He talked us through The Ikea Effect, the Sunk Cost Fallacy (also know as Irrational escalation), and the Bandwagon Effect, and (with very wry humour) talked us through how they had applied to a project at Unboxed. The case study really helped to bring into focus how easy it is to fall victim to these biases.
- UX practices that bring measuring and the scientific method in to our work can help overcome these. Usability testing can help combat the Ikea Effect. A/B testing gives you metric you can use to get around emotion-based arguments. Cheap prototyping can help avoid the Sunk Cost Fallacy. However, there are cognitive biases mess with these practices too (like the Hawthorne effect, Fundamental Attribution Error, and Confirmation Bias)!
- Despite the fact they our brains have adapted to let us tell each other where the good fruit is, we can fight back by measuring things, evaluating things, and practising mindfulness.
- Review some recent (and upcoming) decisions the team has made, looking out for evidence of cognitive bias, and discuss with the team.
Applying design principles to software development
Presented by Adriaan Fenwick (Slides, Twitter: adriaanfenwick)
- Your users are the most important aspect of building a product. Personas let you figure out who your users are. We’re good at remembering abut users’ needs, but we shouldn’t forget about their wants too.
- Adriaan also talked about breaking down silos and working in cross-functional teams.
- The User Review part of a process must contain real world users, not just someone from your team. Take the team along to usability tests: it helps create empathy for the users. Take some time to evangelise UX and educate the team around you.
- Set up one usability testing session and get some of the engineers to come and watch. This seems like a good first step to the “Set up regular, small, usability tests” To Do from my notes from day one.
I spotted an article the other day (Getting the most out of conferences and events) that caught my attention because I was about to head off to UX South Africa. The bit that I particularly liked was ending each bit of session notes with a To Do item. I like the idea of having and doing something tangible from each session, rather than the more hand-wavy “I got smarter.” Here are some of my thoughts and notes from today’s talks.
UX in SA: any second now
Presented by Phil Barrett of Flow
- UX and software takes time, and it’s an ongoing process. Phil summed this up very nicely by saying: “Digital products are soap operas, not movies.”
- It’s important to build prototypes. Usability testing using them and iterating on them can lead to breakthroughs. Breakthroughs can also come from asking “stupid” questions.
- Phil referred to the market several time during his, talking about how exposure to real users is very useful. “Making stuff is the new strategy.”
- He talked about changing requirements into assumptions and hypotheses: doing this provides you with a metric for proving it worked.
- Look at existing user stories and see about changing them to be more like assumptions and hypotheses.
The ROI of UX Design
Presented by Werner Janse van Rensburg
- Measuring can justify spending and prove the results of UX work. You can measure effectiveness, efficiency, satisfaction.
- The cost of making changes to a product ramps up dramatically as you progress through the Software Development Life Cycle. I think that this is mitigated a little if you’re using an agile process.
- Set up measuring of various stats (for my work, that will be Front-end performance and accessibility metrics). Do this first, before making any changes to the system.
How can UX work with agile teams?
presented by Samantha Laing and Karen Greaves of Growing Agile
- Sam & Karen also talked about the market, and the value of testing with real users in real life rather than just in the lab. Get feedback from the market, then rework your product.
- Do ongoing, short usability tests. Spending more time with real users is good for everyone on the team: developers, designers, and everyone in between.
- “Get as close to your user as you can.” If you can’t reach the user, but you can talk to someone in customer support: do that.
- They also reminded us that in Agile only the the final working product counts as “done.” They talked a lot about making slices of pieces of a product.
- Set up regular, small, usability tests.