DCX Final: Prototyping and Internal Testing & usability testing

why prototype?

shows the journey between pages

reveal what the wireframes hide

cheap to fix

match how users actually experience the site

common mistake with redesigning

opening figma and adjusting spacing, pick colors, align to pixel grids

(digitize should take 3-5 mins, if longer that's redesign)

design mode vs prototype mode

design: select shapes, editing text, adjusting layouts

prototype: adding clickability, connection handles, and arrows

when user clicks connection for transition, three options exist

instant

dissolve

slide

instant

no animation, just jumps to the next page, default and recommended choice for testing (use this one)

dissolve

a soft fade between pages

slide

the next page slides in from a direction

3 roles for internal testing protocol

navigator (click through prototype and narrates thoughts aloud)

observer (watches silently and takes notes)

roles switch (5 mins per prototype, then trade

reframe

if navigator gets stuck, that is success because stuck in a prototype costs nothing, which is different from live site

usability testing

observing real users as they attempt tasks on your design

- emphasis on what they DO

analytics tools tell you...

WHERE users struggle, tell you that 40% visitors leave after page, but not WHY

what is one of the most important findings in usability research

5 usability tests reveal 80% or more of major usability issues

think-aloud protocol

user narrates

facilitator observes

probing questions

user narrates

As the participant interacts with your design, they say out loud what they see, what they expect, and what confuses them. This is not natural behavior. You set it up explicitly

the facilitator observes

You watch the narration stream without interrupting, helping, or reacting to their choices. Your job is to see the gap between what you intended when you designed the page and what the user actually experiences. That gap is the finding

probing questions

When the user pauses or does something unexpected, you prompt with open questions. Every probe is a question, not a suggestion

silence is data

user stops talking and stares at the screen, that silence means they are confused or lost

3 criteria for good test tasks

realistic scenario

no navigational hints

clear success criteria

realistic scenario

Frame the task as a situation the user would actually encounter. Not "find the events page" but "you heard about an event from a friend and want to learn more."

no navigational hints

Never mention button names, page labels, or where to click. If your task says "click the Events button," you have told the user exactly where to go. You learn nothing about whether your navigation is discoverable

clear success criteria

You need to know when the user has completed the task or given up. "Explore the site" has no success criteria. "Find the time and location of a specific event" has a clear endpoint

four rules for facilitator discipline

stay neutral

probe, don't lead

don't help

don't defend