KIWI WORKPLACE ENGAGEMENT APP

UX Research + Design • May 2015

PROJECT SUMMARY

This project was born out of a three-week contract with Navicet, a strategic design consultancy based in Kirkland, WA. Navicet (DBA Squawk Metrics) makes a product called Kiwi Dials that facilitates the measurement of team health, with the aim of building awareness of employee engagement.

Employee engagement matters. According to a recent Gallup report, 70% of the U.S. workforce is disengaged, and 20% are actively trying to sabotage their employer's objectives. Gallup reported that organizations with high engagement "consistently outperform their competitors, and they consistently grow and thrive-- even in challenging economic times."

The folks at Navicet believe that the first step toward fixing engagement issues is to begin measuring them, and a good way to start is by baselining your team health using a product like Kiwi Dials. To date, Kiwi Dials have only existed in a physical format that team members can interact with in a co-located workspace. There are several complications with this approach-- it obviously doesn't work well for remote workers, or those who may be shy, introverted, or otherwise reticent to share their feelings in the workplace. Naturally, one isn't able to query the system for historical data, unless someone takes it upon themselves to log the kiwi's movements.

Our team of three (myself, Lindsay Henwood, and Liv Browning) set out to design and prototype the next evolution of Kiwi Dials based on comprehensive user research, and more importantly, empathy for those who feel less than 100% engaged at work. We've all been there. 

PROCESS

We laid out an aggressive project plan in order to accomplish the following list of activities, and deliver on our goal of creating a fun, effective, and real-time solution for tracking employee engagement.

TOOLS USED

Axure, Photoshop, Illustrator, Asana, Sketch, Visio, Keynote, KwikSurveys, Excel

MY ROLE

This was a very fast-paced project, but the three of us played more or less of an equal role throughout the research, planning, and testing phases of the project in the first two weeks. Given my background in math and analytics, I took on the responsibility of designing the reporting and detailed views of team metrics in the app. In the third week, when it came time to polish our deliverables, I took responsibility for the final prototyping and interaction design for the app. 

RESEARCH, PLANNING, DESIGN

The managers we interviewed noted that the indicators of impending trouble are often subtle: unexpected overtime, missing or lacking commitment, the overarching "feeling" in the room being off, conflicts, missed deadlines, and when one team member is overly domineering. One interviewee added:

"By the time we notice someone is unhappy, their performance is already suffering." 

Regarding communication in particular, it was interesting to note the divergence between the managers' view vs. that of the individual contributor. When asked about communication in the workplace, managers were more likely to reflect positively on their communication skills, regaling their participative leadership, open door policies, weekly touch-bases, one-on-ones, etc.

Individual contributors, on the other hand, used phrases that illustrated a distrust of management and/or company culture: fear of reprisal, retaliatory environment, afraid of ramifications, lacking recognition, and serious misgivings about workplace surveys that claim to be anonymous. In fact, 72% of the individuals we surveyed reported that they do not trust the "anonymity" of surveys that claim to be anonymous (N=42). 

Of course, it's imperative that management has insight into how and what their workers are feeling and thinking, but it's also important that individual contributors feel valued and safe at work. As mentioned above, team health is highly correlated to success, so it was clear that our solution must help leadership identify issues early on, but maintain focus on the end-user and their feelings of security. 

We got to work sketching, and tested with paper prototypes early in the second week of the project. We learned quite a bit with each test-- users liked the friendly voice, tone, and playfulness of the Kiwi, but weren't sure whether or not they'd use something like this at work, unless it was a mandate from management. This wasn't what we were going for, so we went back to the drawing board. 

At this time, we developed tenets for Kiwi, or principles to guide our design decisions, maintain focus on the end-user, and create a structure for which we believe Kiwi is best used.

The physical manifestation of the Kiwi Dial works best when all team members are engaged in deciding what to measure (e.g. fun, clarity of purpose, autonomy, impact), and Navicet recommends going through an affinity diagramming process to nail down these meaningful aspects of team health. In hopes of creating more engagement with the app, we tried to incorporate a brief card sort-like experience in our second prototype to mimic this exercise. Users found it too burdensome , and we ultimately opted to have a team moderator (likely the person who signs their team up to use Kiwi) decide which aspects of team health to measure.

We developed two personas: One is the more gregarious team lead (who sets the team up on Kiwi), and the other is an individual contributor on the same team. We created a user flow for both personas, but due to time constraints, were only able to fully prototype the individual contributor's workflow. 

THE SOLUTIOn

In total, we iterated seven times and administered 14 usability tests. The final version is a strong foundation for future research and exploration, though we feel that there are many more features that should be added for a full-release version. 

key findings and next steps

Anonymity vs. Control of Data

One of Kiwi’s distinguishing features from its competitors on the market is the commitment to anonymity: that a user can share their opinion without anyone knowing it was them. Complete anonymity is hard to promise (distinct user actions must be logged in order to provide the structured data we've mocked up in our designs), therefore we suggest that Kiwi markets this feature as giving users control over the visibility of their data. 

Providing users the ability to add comments to their check-ins was a feature that we implemented in one of our last iterations, and we recommend giving the user control over whether or not those comments are public (anyone can see them when running reports or looking at daily details), or private (like a memo or diary feature expressly for viewing by the user who wrote them). 

selecting aspects of team health

As mentioned above, the value of using Kiwi was not initially apparent to testers who were presented with pre-chosen Aspects of Team Health. Testers indicated a preference for democratic selection and felt there was more value in using Kiwi if they were involved in selecting the Aspects their team was measuring. It provides the user a greater sense of ownership, and we feel that it is a key feature that needs further exploration. 

THE CHECK-IN PROCESS

It may be appropriate to add a seventh tenet to our list above: Kiwi is not a survey. Check-in processes can teeter on the edge of feeling like surveys, which we want to avoid. Users dislike and distrust surveys in the workplace: they're not fun, often don't provide useful data, and unfortunately can be used to single out individuals despite their claims to be anonymous. 

Our very first prototype featured a check-in process, and our stakeholders challenged us to think differently about this aspect of our design. We came up with several alternatives, none of which tested well with users.  We also tested several iterations of “check-ins” before arriving at a Likert-type scale for users to select from. The use of kiwi images, coupled with color and simple phrases were to easy for users to understand, quick to use, and maintain a greater degree of accessibility for color-challenged users.

In our usability tests, users indicated a preference for checking in on Kiwi on a regular basis, either daily or several times a week, and would be more likely to use it if they were reminded to do so, either by their teammates, manager, or with a notification sent from the app. We recommend a reminder setting to be controlled either by the user or team moderator to set notification frequency.

for future consideration

Aside from the recommendations listed above, we suggest running a longitudinal study with a team that can test out a functional Kiwi prototype, to grok whether or not users remain engaged with Kiwi over a greater period of time, and how the design might need to be modified for this purpose.

UPDATE: KIWI DIALS HAS LAUNCHED!

After almost a year in development, Kiwi Dials launched in February 2016. I have been fortunate enough to keep in touch with the fine folks at Squawk Metrics, and I was invited to beta test the app and provide feedback before its launch to the public. 

Now that the app is live (for iOS, Android, and Windows Phone), I recently took a few key screen grabs of the current state of the app (version 3.4, as of June 2016) on my iPhone 6, and included a short critique of the live app, below.