User Testing Briefing 

Welcome and thanks for stopping by. I’m <EXPERIMENT LEADER NAME> and this is <OBSERVER NAME> and we’d like to have your help to evaluate our proposed user interface. It’s for a system designed to help MIT dorm desk workers with their commons tasks. Today you will be a desk worker and we’ll guide you through a couple of tasks and scenarios that desk workers commonly face. Before we start, we’d like to remind you that your participation in this evaluation is completely voluntary and you may stop at any time. Any problems you encounter are not your fault, they are solely the interface’s, and we need your help to find them. I’ll am the experiment leader and <OBSERVER NAME> is the observer. This means I’ll prompt you tasks and try to answer your questions the best I can and <OBSERVER NAME> will be watching and taking notes for me. We’d like you to think out loud as much as possible to help us get an idea of your thought processes as you explore and use our interface.

Scenario Tasks

Task 1

Log in to deskdash using the following login information and check to see if there are any pressing matters since the last time you worked.

username: username

password: password

Task 2

The UPS delivery man just delivered a package with the following address fields:

From:

Amazon Inc.

100 Amazon Way

Seattle, WA 12345

To:

Ben Bitdiddle

Room 123

MIT Simmons Hall

229 Vassar Street

Cambridge, MA 02139

The dorm you are working at sorts packages according to the first letter of the recipient's last name as follows:

A-H : bin1

I-P : bin 2

Q-Z : bin 3

Register this package according to the dorm's sorting method.

Task 3

Alyssa P Hacker would like to borrow the dorm's vacuum cleaner. Check it out for her.

Task 4

Ben Bitdiddle arrives at desk and asks for his package and the movie The Lion King. Check out his package and The Lion King.

Task 4.5 (only used during the second iteration of testing)

The bathroom in the lobby is supposed to be painted on Monday but the painters just called and said they can't make it until Tuesday. Make sure the other deskworkers know about this change.

Task 4.55 (only used during the second iteration of testing)

Earlier you registered a package for Alyssa P Hacker but it was actually for Alyssa P Smacker. Fix this.

Task 5

Logout of deskdash.

User Testing Round 1 Prototype Photos

These figures reflect the design of deskdash during the initial round of user testing:

Fig 1. Log in screen

Fig 2. Home screen

Fig 3. Packages screen

Fig 4. Items screen

Fig. 5 Notes screen

 
Fig. 5 History screen

User Testing Round 1 Observations

  • Two testers said they wished the system knew which bin packages were supposed to go in and autopopulated the field as soon as they typed the recipient's name. This is a usability issue.
  • All of the testers said they weren’t fully comfortable using the sidebar at first because they were new to the system---they said they preferred to return to the homepage where every button had a caption, but that they would use the sidebar once they got used to the system. This is a learnability issue but it's not a big issue.
  • Every tester tried to add the vacuum cleaner to the system instead of making a note. This is a usability issue.
    • Two testers tried to click at the bottom of the spreadsheet to add a new row
  • Two testers said they wished they could edit a pre-existing package or item. This is a usability issue.
  • Two testers typed room numbers of package recipients into the location field instead of bin numbers. This is a learnability issue.
  • One tester disliked that the package check out button and the item check out button did different things and suggested standardizing interactions with the interface
  • Testers said they would not have checked the notes upon login unless the received a notification saying there were new notes

User Testing Round 2 Prototype Photos

These figures reflect the design of deskdash during the second round of user testing:

Fig 1. Log in screen

Fig 2a. First home screen

Fig 2b. Home screen

Fig 3. Packages screen

Fig 4. Items screen

Fig 5. Searching for "Harry Potter and the Deathly"

Fig. 6 Checking out an item

Fig. 7 Notes Screen

User Testing Round 2 Observations

  • One tester said the following: "Why is the button to pick up a package marked 'check out'? It's not like they're going to check it back in. It should say 'pick up.'"
  • Testers didn't like the pop up modal dialog upon login that told them they had new notes. They wanted a nonintrusive notification like on facebook.
  • One tester suggested that we notify desk workers of new calendar events in addition to new notes. This is a safety issue because there are some important events that desk workers need to know about.
  • During task 4.55 two testers checked out the incorrectly addressed package and then registered a new package with the correct name. Afterward, testers suggested an edit button on every row. This is a usability/learnability problem.
  • Testers liked the sidebar icons and the big buttons with pictures on them because they were easy to find and their functionality was obvious. This is a learnability pro!
  • One tester wondered why the search bar was so prominent and big.
  • Testers liked the specificity of the search bars on each page and were glad there wasn't an omni search bar on every page
  • Two testers asked for undo capabilities like in gmail (an undo button that appears for 5 seconds after an action). This is a safety issue.
  • One tester wasn't sure if he should remove notes after he read them or if he should do something about the notes. This poses a safety problem because if users delete the notes, it will delete them on every desk worker's account.
  • Testers weren't sure what the sticky note in the bottom righthand corner would do so they didn't click it but, when asked, they correctly guessed what it did (brought up a dialog to create a new note). This feature was learnable but users didn't use the system long enough to learn it.

Prototype Iteration

  • We changed our design to support full CRUD (create, read, update, delete)
  • We changed the system to auto-populate fields when able
  • We tried to standardize buttons across as many screens of the user interface as possible although the functionality of the item check out button is different than it is on the other screens
  • we added a notification upon login to alert the desk worker that there are new notes
  • No labels

1 Comment

  1. Prototype: Photos: Were good and clear. Numerous photos were shown covering two rounds of user testing -- good! They covered all the interesting states of the system, and conveyed the message clearly.

    Iteration Changes: Were done well based on the first round of observations (i.e., the new notes notification now is the shown the moment they log in).

    Briefing & scenario tasks: Briefing: Good! Covered the motivation and goals of the system in a concise way, and also took steps to ensure that the user was oriented, voluntarily present and was in control at all times.

    Tasks: Majority of the tasks were high-level, and short
    User testing observations: No. of Users: Didn't explicit showcase that there were 3 users per itereation.

    Observations: Usability problems that were identified were clearly identified and presented. Problems identified were indeed usability problems. Notes taken were detailed.
    Wiki presentation: Good, clear wiki.

    Overall: Good job! This project is really shaping up well.