Steve KrugUPA DC UserFocus

Everything you know is wrong

Ive been very fortunate

Ten years ago I got the chance to write a book Never expected to earn more than the advance Real motive: raise my consulting rate

2001 Steve Krug

2001 Steve Krug

As it turned out, though

2001 Steve Krug

2001 Steve Krug

And I got to teach workshops

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

Perfect for me

No chance to procrastinate No homework at the end of the day

2001 Steve Krug

Nine years later

My wife had finally forgotten the first one And Id become more convinced than ever that

testing was invaluable

2001 Steve Krug

Things I believed

Usability testing is the single best way to improve the effectiveness of what youre building Every site or app should undergo frequent

usability testing during (and after) development Most people cant afford to hire someone to do

it Even if they could, there arent enough

professionals to go around Ergo: They should be doing it themselves

2001 Steve Krug

Things I believed

Usability testing can retain most of its value even if done simply With a little training, almost anyone can do

valuable usability testing Its not rocket surgery So off I went

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

Again my intention was simple

Write a do-it-yourself usability testing book that anyone could use Show how simple I think testing should be Persuade people to start testing

2001 Steve Krug

Unintended consequences

By the time I was done, I ended up believing that some things I always thought about testing are wrong OK, not everything Thats what I want to talk

about today Hope youll argue with me

2001 Steve Krug

Im just curious

Show of hands:Read Dont Make Me Think?Read Rocket Surgery?Heard of Rocket Surgery before today?

2001 Steve Krug

1. Test on a fixed schedule

Dont schedule testing based on when things will be ready to test Schedule monthly testing

2001 Steve Krug

2001 Steve Krug

A morning a month, thats all we ask.

2001 Steve Krug

2. Test three people per round

Forget eight is enough arguments Even four is too many Its always easy to find more problems

Harder to find the resources to fix themHaving a long list you dont have time to fix is

dispiriting and distracting

Three users will generate more than enough serious problems per round

2001 Steve Krug

3. No more big honkin reports

Written report has little value and takes a lot fowork and time Debrief the same day over lunch At the end of lunch, youre done for the month Reinforces the importance of attending the

sessions Publish a one-page email with a bullet list of

tasks tested, problems youre going to fix

2001 Steve Krug

4. No one starts testing early enough

Professionals all agree that starting early is very valuable Almost no one does it

We know there are problems. Why waste their time and ours?We always have a better version in our head People need to be bullied into starting testing

earlier Suggest testing competitors

2001 Steve Krug

2001 Steve Krug

Start earlier than you think makes sense.

Incorrect thinking

2001 Steve Krug

Correct thinking

2001 Steve Krug

5. Attendance is the top priority

The conversion experience is what youre shooting for Live events are more compelling A lot goes on in the observation room If you have any money, spend it on great

snacks Scheduling a monthly UT day helps

People can plan on attending You can pick a slow time of month Testing onsite helps

2001 Steve Krug

2001 Steve Krug

Make it a spectator sport.

2001 Steve Krug

2001 Steve Krug 2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

2001 Steve Krug

My ideal home page, as told by

2001 Steve Krug

My ideal home page, as told by

6. Fix the worst first!

The most serious problems we discover often dont get fixed

2001 Steve Krug

2001 Steve Krug

Focus ruthlessly on a small number of the most important problems.

2001 Steve Krug

7. Tweak, dont redesign

2001 Steve Krug

2001 Steve Krug

When fixing problems, always do the least you can do.

2001 Steve Krug

Your motto should be

Whats the smallest change we can make that we think might solve the observed problem? Resist the impulse to defer serious problems to

a redesign that may never happen

2001 Steve Krug

2001 Steve Krug

8. Picture-in-picture must die

PIP isnt just unnecessary; its a bad thingMore technology to deal with Larger files Its not necessary. If you have good audio, tone of

voice gives just as much information

But the real problem: Its a distraction

2001 Steve Krug

Shameless self-promotion

Next week in DCMe (Weds.)

Do-It-Yourself Usability TestingGinny Reddish (Thurs.)

Writing Web Content that Works Lou Rosenfeld (Fri.)

Site Search Analytics

If you can read this, $100 off each day

2001 Steve Krug

2001 Steve Krug

Thanks for all the fish

Send any questions, feedback, gripes to skrug@sensible.com

Questions?

2001 Steve Krug

2010 Steve Krug

Slide Number 1Ive been very fortunateSlide Number 3As it turned out, thoughSlide Number 5And I got to teach workshopsSlide Number 7Slide Number 8Slide Number 9Slide Number 10Perfect for meNine years laterThings I believedThings I believedSlide Number 15Slide Number 16Slide Number 17Slide Number 18Slide Number 19Slide Number 20Slide Number 21Slide Number 22Again my intention was simpleUnintended consequencesIm just curious1. Test on a fixed scheduleSlide Number 27Slide Number 282. Test three people per round3. No more big honkin reports4. No one starts testing early enoughSlide Number 32Incorrect thinkingCorrect thinking5. Attendance is the top prioritySlide Number 36Slide Number 37Slide Number 38Slide Number 39Slide Number 40Slide Number 41Slide Number 42Slide Number 43Slide Number 44Slide Number 45Slide Number 46Slide Number 47Slide Number 48My ideal home page, as told byMy ideal home page, as told by6. Fix the worst first!Slide Number 52Slide Number 537. Tweak, dont redesignSlide Number 55Your motto should beSlide Number 57Slide Number 588. Picture-in-picture must dieShameless self-promotionThanks for all the fishQuestions?Slide Number 63