We use Mechanical Turk a lot when adding a new feature. Before releasing it we want to know if the UI clearly defines this new feature.
Turk is great overall and usually 7 out 10 Turk testers are able to figure through our UI the new feature. Though I wonder should I worry about the 3 (various tests) in which our UI failed them?
Would you not release an update until your UI passed the 10 out 10 test?
Turk is great overall and usually 7 out 10 Turk testers are able to figure through our UI the new feature. Though I wonder should I worry about the 3 (various tests) in which our UI failed them?
Would you not release an update until your UI passed the 10 out 10 test?