Hallway usability test: how many user interfaces do you really work for?

Question

When running usability tests in the hallway, do most of you fully or nearly fully function in your applications? Or do you just make sure the links or the thread sequence is correct? Or do you just draw on paper and go with it?

I would like to test the prototype at an early stage and try to find a good balance. But at the same time, I am worried that some non-functional parts may not actually produce representative results.

Thanks.

Answer 1

Usability tests, a corridor or something else, you only need the features you need to test. In most usability tests, you should turn to specific design questions to answer and design your prototype to the extent that it can answer these questions. For example, if you need to check if users understand your sorting data for a table, all you need is a paper image of the table showing the sorting indication (with blurred table contents) and ask them how the table is sorted. If you need to check IA , you only need a few web pages that are blank, with the exception of the title, which are linked through menu navigation.

You only need pages that are relevant to the tasks you give your users. If you are just checking IA, you only need pages on the normative path. If you are also testing error recovery, you need regulatory pages along with complete navigation controls. If you also check for error detection, you also need content on the pages.

You can also simulate functionality when it's easier to do. For example, when testing, if users can figure out how to get the desired sort order, when the user clicks on the idle control to sort the table, you can say, “Okay, this will do it for you”, and you take the mouse and select the bookmark that displays table in a new sort order.

In the walking test, if users violate the limit of loyalty, you can simply say: "I have not done this part yet. Let's go back to A and continue from there." Of course, you should notice that the user made the wrong turn in the task that you intended for them. I had no problems with users complaining about idle functions when I tell them about my incomplete prototype and only tested the interface for the x, y and z functions at the moment.

For low-precision prototypes, I often call them “mock-ups” or “drawings” for users, rather than “prototypes” to indicate low functionality. You can put obvious placeholders for missing content (for example, "Blah, blah, blah ...", "TODO: Product Image Here"). If the user comments on something outside the fidelity envelope (for example, “This symbol should be red to stand out more”), just mark it and say that the topic is under development (for example, “Thank you. We did not start working on colors but just tried to figure out how to organize the site right now. ").

Usability testing with limited revision prototypes is really necessary so that an iterative project is practicable for most projects. Otherwise, you work too much, developing things that need to be redone.

Answer 2

A few things to remember:

• Test early and often.
• The purpose of usability testing is to find problems with the user interface, not the Q / A of your code.

Therefore, if users can see parts of your user interface, you are interested in testing and interacting with them in a realistic way (for example, on buttons and links), you can collect useful data. If some links are dead ends, it’s normal if there is some way for users to recover and continue. In principle, with prototypes, the “right” path should work, but it’s normal if there are no wrong paths (as long as there is a reasonably quick way to get back to the correct path). Even static storyboards (broken user interface drawings) can provide you with some information if you ask the right questions, for example: "What would you do on this screen if you want to see your cart?" )

Answer 3

I would suggest a couple of rounds of usability testing. First on paper, perhaps later on the screen, usually throughout the life cycle of the application (use the Agile approach).

There is a good argument for paper prototypes. When users see the screen, even limited functionality, they may hesitate to suggest changes, as they look “done.”

Make no mistake, it’s not trivial to have it all on paper, but on where I will start. Probably start with a section or two applications. And make sure someone with good people skills and / or explanatory skills needs to go through the user. Ask the second person to take a note. Try asking open-ended questions, etc.

Answer 4

For the test in the corridor, I would check with NONE from the implemented functionality.

Test the design done on the board or on paper. You will be surprised how much you learn in these minimal layouts. And they are very inexpensive to do!

Functional prototypes for follow-up. If you give usability to a functional interface, they are much less likely to ask if you have implemented the right set of functions in the first place.

Answer 5

I would make the user interface functional, so that the user can really play with it, it will be much better than a static image. People can tell you if they feel comfortable in the user interface.

Answer 6

I would make sure everything in the user interface works, or at least you got into a clear, unambiguous message indicating that the function has not yet been implemented .

Showing prototypes to customers with a disclaimer of how function X is not working yet is usually ignored. They will try the prototype, click on X and reply indignantly: "Function X does not work! It really needs to work in the final version! Why does it not work?". The customer is confused and dissatisfied with the product, and he disappoints himself because he overshadows the positive feedback. In addition, you told them that it did not work, why they can not use their imagination to imagine how it will work in the final version?

Make it work, whether it's a rough version, dummy data, or even a simple message that says: "Now the results will be sorted in alphabetical order."