mobile Usability

#iOSDevUK: Maxim Cramer – Users Don’t Bite

Credit: Wikipedia
Credit: Wikipedia

First thing learned when starting to make apps. Was making an app about beauty products. Went looking for people who would purchase beauty products (Oxford Street!) started in Starbucks.

Asked for 5 minutes.

People were really nice, if it wasn’t for them wouldn’t have discovered how hard it was for people to get to the product page.

People daunted, confused, takes a lot of time.

If release it like that you’ll get app store reviews of 1* “your app is shit”, this doesn’t help you figure out what is wrong.

If you take the time to sit down can uncover issues and learn a lot in the process. Doesn’t have to have a 2-way mirror, can just sit down and learn the application and see what they do.

Analytics will tell you what, but not why. We want apps to be intuitive. How? Listen to the people who use your apps every day.

Why don’t we test?

  • We forget.
  • We don’t always know how.
  • We don’t see the benefit.
  • We’re scared.

Getting involved makes you a better developer.

Running automated tests isn’t enough, need to get your hands dirty.

WHY do it?

  1. Save dev time, throw away less code.

Money transfer site.

  • 9 months dev
  • Tested day before launch.
  • Ranked lowest in comparison to competitors.
  • CEO pulled the plug. Not launching tomorrow.
  • If people couldn’t use it, would just get a bad rep.

Build a better product from day 1.

More code you write the better you are at it. Same for this. Devs get better at tackling problems, the more you watch users struggle.

  • We can help. We know what’s possible.
  • Can see what people are trying to do, build it, make it easier for them.
  • Data driven decision in your team.
  • Know how it works, but can be surprised when put it into someones hands.
  • Tap and hold (iOS), suggested for edit a note. People disagreed, but thought it would work. Put it into user testing and everyone managed just fine. People knew, because it’s how you delete apps. Could have turned out the other way, but at least now it was based on some kind of data.
  • No-one starts out with all the answers.

SwiftKey been doing this for the past few months. Had no time to lose, tested every single week.

  • Helps you understand changes.

Spent ages working on something, want to know why. If you see someone struggle, makes a lot more sense.

  • Helps connect to the people you build for.

When stuck behind your laptop don’t see scared face. Someone who doesn’t get what they should do.

E.g. “the case of the missing text” – can’t let it go when seen someone experience that. Can’t ship with a bug like that. Fix it.

How Does This Fit In?

  • Do I have time for that?
  • Depends on how you work. UX team.
  • Watch one session a week (if do weekly testing).
  • Brings people and experiences into discussions.
  • Put it into people’s hands.


Need people.

  • Target audience.
  • 5 people.
  • 80% usability issues.
  • Aim for 6-8.


  • Tasks
  • Prep
  • Observations
  • Prompts
  • e.g. Could you delete a notebook?
  • Have to have the right content
  • Follow up from previous tasks.


  • Do they ever swipe?
  • Can they press buttons easily
  • How do they respond to a prompt?


  • How would you think you could do that?
  • How do you do this normally/elsewher?


  • What do you think happened to the content?
  • Did you feel you could do this by accident?
  • Why did you do what you just did?
  • What did the action remind you of?


  • How easy was that?
  • What did you like?
  • How could it be better?


  • Welcome them
  • “We’re testing the product, not you”
  • “Please, talk out loud”
  • “Nothing will offend us”

Interaction During the Session

  • Don’t help
  • Don’t encourage
  • Use your prompts
  • Keep the conversation going.

People say devs shouldn’t do user testing. Because we made it, so much easier for someone independent rather than involved. Subconsciously will smile when they get it right, be annoyed when they get it wrong. Users want you to be happy, will try and please you.

(Solves this by being enthusiastic all the time)

Closing Qs

  • Repetition of good and bad
  • “how much would you pay?”
  • likelihood of recommendation (1-10) (NPS)
  • “What would we need to get to that 10?”


  • Focus on observation
  • Be positive
  • Be conversational


What do do with results? Fix them. Any major issues will surface

There’s nothing like watching someone in love with what you made.