❐ Just Enough Research

• by Erika Hall

User Research - Specifics

• ~ in user testing, the unpredictability of real life is desirable
• ~ you can't ask people what they want, because they want to be liked, so they'll just give you a 'likeable' answer
• ~ start user interviews with a genuine thanks
• ~ watch for people blaming themselves for problems in user tests
• ~ if people involved in the design do user testing, small hints and leading questions quickly creep in
• ~ leaving the neutral option in questionnaires is important to find out what people don't care about

User Research - Philosophy and Approach

• also, UX mindset in general:

• ~ accept no substitute for actually observing your (potential) users

• assumptions about your users may lead to (needless) discrimination
• ~ people either lie, or lack the self-knowledge to be truthful

• ~ in user research, talking to the wrong people is the worst thing — you learn nothing

• ~ be ready to joyfully accept that people don't need what you thought they need

• • (this also works standalone, not just in funny overlap with the above)
    • ~ the faster you're proven wrong, the less time you will spend being wrong
• ~ find gaps between your offering and user needs
• ~fight the notion that doing user research is weakness — it's scary, but smart
• ~ in user research, prefer messy real life over the lab
• ~ seeing your theories about how people use your design crash against reality is good fun
• ~ you want to know a small number of representative users really well
• ~ usability is the absolute minimum, ~ usability is basic manners

Own, somewhat tangential further thinking

• Missed stakeholder analysis completely with Bode
• The more I read books relating to big coorps, the less I want to ever go back
• what kind of fundamental user research is useful to do for ed games?
• Habits as a target of user research

Misc Tips

• ~ get people to buy into your project by 'consulting' them on their opinion
• ~ Split Testing is susceptible to local maxima
• memorize Nielsen’s ten heuristics?
• ~ a good manager's skill is making everything run smoother

• ~Groupthink as symptom of wanting to appear in agreement

• ~ collaboration needs shared objectives, not just time together

• ~doing research together as genuine team-building activity

• ~ user-test your competitor's product

• ~ get good at recruiting, it makes research so much better
• ~ if you do user research in the field, you get to see the weird, unconscious habits, like clunky workarounds
• ~ to write a good survey, explicitly write down what do you want to find out, why, and which questions is best for that
• ~ consider 'invalidating' your design
• ~ force people above you to watch people struggle with a design to convince them change is needed

Pitfalls and Problems

• ~ bad survey questions completely prevent you from getting any usable data (while interviews are often somewhat useful)
• ~ unless your survey is excellent, it's bad, biased, meaningless data
• ~ surveys are too easy

• ~ the neutral answer in a questionaire is hard to interpret clearly

• ~problem of Agile is that it's only about how, not what to build

• ~ 'validating the design' frames research as 'prove yourself right'

• ~ 'validation' mostly means box-checking (in practice)

Concepts, Methods

• ~ 'intutive' means 'matches mental model of user'

• Net Promoter Score

• Affinity map
  • ~ use affinity diagrams to understand user's mental model
• 4 methods of evaluating UI according to Nielsen
• ~ use Third-Party studies to understand consumer and user trends
• sled