KLOA iOS Keyboard Usability Study
Evaluating a novel third-party iOS keyboard concept through onboarding, usability testing, interaction mapping, and design recommendations.
KLOA was a novel third-party iOS keyboard concept designed to reduce typing errors through larger input regions and autocorrect support. The interface looked and worked differently from standard iOS keyboards. Instead of tapping individual letter keys, users typed through combinations of taps, double-taps, and swipes in a lower touch target area while using the alphabet bar as a visual guide.
Our four-person team conducted a usability evaluation for the client. The goal was to understand whether users could learn the interaction model, whether they found the keyboard useful, and whether the novelty of the concept created too much friction for adoption.
I served as system manager and client liaison/coordinator for the project. My contributions included:
- Coordinating with the client and helping clarify study goals
- Creating an interaction map of the keyboard's modes, gestures, and input behavior
- Leading visual design and editing for the usability study kit
- Helping create the onboarding materials used during testing
- Overseeing the usability testing and recording setup
- Managing data logging during sessions
- Editing and compositing participant video footage
- Editing the final presentation and report
- Creating and editing graphics used in the study materials, report, and presentation
Our team collaborated in real time using shared documents, video calls, and group editing tools to plan the study, revise materials, analyze findings, and prepare client deliverables.
The biggest challenge was that the product was unfamiliar by design. The keyboard looked simple, but its interaction model was not obvious to new users. We needed to understand whether people could learn it, whether onboarding could help, and whether the potential benefit outweighed the stress of adoption.
- Could users understand how the keyboard worked?
- Would users want to keep using it once they understood it?
That distinction mattered because an interface can become learnable over time but still fail if the early experience is too frustrating or if the payoff is not clear.
Before testing, I created an interaction map that described the app's modes, gestures, and input behavior in both visual and written form. This helped our team understand the full product rather than focusing only on the most obvious interactions. The map covered single taps, double taps, swipes, alphabet input, number input, and the way users moved between keyboard modes.
We first tried informal tests with friends and passersby to see how people reacted without preparation. In all ten cases, people became frustrated and handed the phone back within about five minutes. That early result showed that if we wanted meaningful feedback on the full concept, we needed to help users get past the first moment of confusion. We decided to create onboarding materials before the main study.
We built a usability study kit that included a recruitment ad, screener, consent form, pre-test questionnaire, testing script, task list, post-test questionnaire, and data collection materials. We also created a web-based onboarding site. The screener focused on mobile device familiarity and openness to third-party keyboards.
We used QuickTime on a Mac to record the iOS device screen and used streaming tools to share the session view for remote note-taking. We also experimented with webcam and overhead camera views to capture hand movement and participant behavior. After each session, I edited multiple video feeds into combined clips that the team could review together.
During the main study, we used a script to keep sessions consistent. Participants completed tasks, then filled out a post-test questionnaire. We summarized findings in a presentation and report that included embedded video clips so the client could see specific examples of confusion, progress, and interaction breakdowns.
Participants did not understand that they needed to use the numbered touch target regions to input letters. The alphabet bar looked meaningful, but users did not immediately understand its role as a visual guide rather than the direct typing surface.
The interface appeared simple, but the required interaction model created high cognitive load. Users had to learn a new mental model before they could evaluate whether the keyboard might be useful.
By the end of the study, all participants were more competent. Five of the six participants could correctly describe how to complete the full range of interactions after using the onboarding materials and completing tasks.
As participants moved through more demanding tasks, errors increased quickly. This suggested that learning the concept was possible, but using it fluently created ongoing cognitive and performance challenges.
Participants reported low confidence and high stress while using the keyboard. The early experience was consistently more frustrating than useful.
"Stress level? It's definitely rising. I have no clue why I would switch to this at this point."
Findings Report Delivered to Client
Users needed a guided tutorial to prevent early loss of interest. The concept was too unfamiliar for most participants to discover through exploration alone.
The incentive to learn the keyboard needed to outweigh the time, stress, and effort required to adopt it. The product needed to make its benefit more visible early, whether through speed, error reduction, accessibility, or a more satisfying interaction.
The interface needed stronger cues to explain the relationship between the alphabet bar and the touch target regions. Users needed clearer feedback about what their actions were doing and why.
The keyboard concept might be more promising in contexts where standard keyboards are difficult or unavailable. Possible directions included larger screens, alternative interfaces, gaming, VR, hardware-based input, limited-dexterity use cases, gloves, underwater contexts, military scenarios, or astronaut use.
We delivered a usability study presentation, written report, study materials, interaction map, findings, recommendations, and edited video examples.
The client received clear evidence that the concept was learnable with onboarding, but that initial confusion, cognitive load, and stress were serious barriers to adoption. Our recommendations helped frame the next design questions: how to teach the model, how to reduce friction, and whether the concept might be better suited to specialized contexts rather than everyday iPhone typing.
This project reinforced that a simple-looking interface can still require a lot of learning. Novel input models need careful onboarding, clear affordances, and a strong reason for users to keep going through the difficult first moments.
It also showed how much the testing context matters. Our early informal tests captured real first-reaction frustration, while the more structured study gave participants enough support to evaluate the product more fully. Both were useful, but they answered different questions.
On the technical side, the recording setup was more complicated than it needed to be. If I ran this study again, I would use a simpler multi-camera workflow composed live in one view. I would also record more of the pre-test and post-test discussion, since participants often spoke useful comments aloud while filling out questionnaires.