Internal validations to the rescue
Our internal validation sessions were a great way to jumpstart our user testing phase and position us for better external validation efforts
Our design and validation phases are in full swing this week! From the story mapping session from last week, we identified key features to build into the MVP before beginning our coworking pilots later this summer. We designed the key features out and built a rudimentary system architecture to began validation efforts. The testing is built to help us surface issues around usability, identify improvements to flows, and develop a deeper understanding of user’s mental models.
This week specifically, our internal validation efforts included heuristics evaluations, card sorting, & think alouds with MHCI faculty and students. By conducting tests with other members of the HCI community early in the development process, we can identify and resolve critical issues in design using local resources.
— Heuristic Evaluations
We built our heuristics evaluations to leverage our cohort of UX experts to quickly identify bugs in early prototypes. We ran five sessions, in which we gave our participants a clickthrough medium fidelity prototypes and asked them to complete a task around question authoring. As they went through the process we asked them to identify usability issues with the interface. Having fresh eyes on our designs gave us insights about inconsistent terminology, unclear interactions, and inconsistent mental models.
— Think-Alouds
Think aloud sessions played an important role for us this sprint, we used them to test concepts around question authoring. We built three different question authoring flows and wanted to see which one people preferred and completed the task efficienlty.
In these think aloud sessions we had participants verbalize their actions and thoughts as they went through the task of creating a new question. The three concepts we tested include: (1) an e-commerce inspired step by step flow, (2) a single page view of all the steps on one screen, and (3) a scrollable input panel with a static output panel.
We found people preferred the single page view of all the steps, they liked that all the steps was condensed onto one page. The e-commerce step click through felt very long, and participants felt it made the process too cumbersome and intimidating. Additionally, we found particpants wanted the system to handle many of the inputs we were asking from them — such as question type or frequency, they wanted the system to figure that out from the question they inputted.
It will be an interesting challenge for our team moving forward to design a system that can reconcile the differences between the system requirements and user’s expectations.
— Card Sorting
Finally, we ran a card sorting exercise to understand how participants would categorize and create a hierarchies between the high-level topics and the actions. The card sorting sessions helped us develop a better information architecture that would inform how we communicate relationships between questions, cameras, and answers, and the navigation of the system.
In these sessions we had cards with high level topics (like camera, settings, questions, and answers) and then associated actions and low level information (like camera configuring, ask new question, look at data insights, permissioning, company information, billing, etc.). We ask particpants to orgainze these cards in way that made sense to them and vocalize their thought process and rationale.
We gained insights around:
(1) Setting up cameras is heavily location based and we need to include ways for users to create groups of cameras based on floors and location specific to rooms.
(2) Create a seperate entry point of creating questions from the dashboard or integration into the main navigation bar.
Pilot Status Updates
This week the team also got a chance to visit BeautyShoppe’s Butler St. and Downtown coworking spaces to scope out camera placement. The Butler St. location is a smaller establishment that is partially staffed. It would benefit with a camera system that offers data about utilization and overall status — but the caveat is it doesn’t have any open desk space that requires heavy monitoring. The Downtown coworking space is still under construction, and isn’t set to be operational until mid July. Given that, we are turning our efforts to beginning a pilot in the East Liberty office. This location is a great option for multiple reasons: it is a larger space, has multiple floors, contains large lounge and open desk spaces, and importantly is operated by a single community manager. We are excited to get Zensors running at this BeautyShoppe location because we really feel like it would be a great chance to test the MVP we are building and offer valuable insights to the BeautyShoppe.
Coming Up
Next week we are going to begin our development phase and external validation! The external validation efforts are going to include user testing of coworking community managers and the development team is going to build out the question authoring flow.
