UInclude

Role:
(Team of 3)
• UX Research Lead
• Client point of contact
• Design of High Fidelity iterations

Timeline + Deliverables:
• 3 week sprint
• Identify personas
• Add new features based on research to V2 of UInclude platform
• Iterations based upon usability testing

Key Methods:
• User Interviews
• Beta testing
• Journey Mapping
• LowFi Wireframes
• Usability Testing
• HiFi Prototype

Problem:
UInclude's augmented writing tool scans job descriptions for gender-biased and racially exclusive language that subconsciously deter qualified candidates from applying. The company has recently wrapped up a pilot program of V1 of the product, and have noticed that users are opting out of their monthly subscription after the first free month is over. UInclude asked our team to both collect feedback from the recent Beta testers, as well as to conduct User Interviews with HR professionals to identify changes and/or additions that would benefit V2 of the tool.

Solution:
Key takeaways from our research, such as the desire for a collaborative platform, interest in using as an educational tool, and resistance to too much information presented in one screen, informed the addition of three new features that we presented back to the UInclude team for V2 of the product.

Overview

Version One of the UInclude platform scans input text for both racially and gender biased language, and suggests edits by highlighting flagged words and providing alternatives that are more inclusive:

A snapshot of the existing UInclude tool. The tool shows a window to input text, which highlights biased words.
There is also a scorecard panel on the screen that provides an "inclusion score" and rating scales for the language input in the editor.

UInclude's current customer discovery data revealed:
‍
• The current color selections aren't clear or accessible to color blind users
• Users desire to learn more about why certain words are biased, so as to not make the same mistake in the future
• Most users of the platform are editing job descriptions with a team, and thus desire more opportunity for collaboration.

Research Objective:

To identify pain points and opportunities for improvement from past UInclude users, as well as to discover new personas based upon outreach and user interviews.

Research

Our team of three divided and conquered to analyze beta testing data, conduct competitive analysis, and to interview hiring managers, HR managers, and recruiters.
I focused on interviewing (9) individuals and gathered the following key insights:

A drawing of 5 sticky notes from user interviews. These sticky notes say:
"I use templates as a starting point when writing JDs."
"I need to be educated on WHY certain words are better than others."
"The simpler the tool, the better."
"I would like to use a tool like UInclude on other materials (i.e. resumes or marketing materials."

Key Takeaways:

Recruiters, hiring managers, and HR professionals are often collaborating with one another on writing job descriptions. They desire a simple tool that tracks changes, and also educates the team along the way.

Research Playback

Our team used a Miro Board to compile our research and to play back our findings to the UInclude team. Our research synthesis led us to identifying three distinct personas who would benefit from the UInclude tool:

A poster that shows Persona A: Jaime Jones' wants and needs:

I am a DEI leader on the HR team.
I spend a lot of her time editing job descriptions from hiring managers and team leads. The hand-off method is not always very efficient, and I wish there was a better way to collaborate with everyone involved in the hiring process.

I aim to:
-track changes made in job descriptions.
-appeal to a diverse candidate group.
-learn more about coded language.

I am frustrated by:
-the time it takes to communicate with hiring managers.
-the lack of standardization in my company.
-too many tools and windows open at the same time.

I enjoy:
-collaborating cross-functionally.
-crafting inviting job descriptions.
-an organized process.

Sometimes these things help:
-Google Docs and collaborative editing platforms.
-Educating hiring managers and teams on best practices in DEI.

A poster that shows Persona B: Kendall Kraft' wants and needs:

I am a hiring manager.
Writing job descriptions doesn't come naturally to me. I aim to address DEI in my language, but don't have the time or resources to do so effectively.
I aim to:
-expedite my JD writing process.
-write JDs that are aligned with industry standards and reasonable expectations.
-craft more inclusive JDs to attract a wider talent pool.

I am frustrated by:
-bare bones and antiquated JD templates.
-my lack of time to craft better JDs.
-not learning in the editing process with my HR team, so i make the same mistakes repeatedly.

I enjoy:
-finding talent and teammates who match our company culture
-focusing on my primary job...managing the team.
-the conversations that come out of an interview process.

Sometimes these things help:
-researching other companies' JDs on LinkedIn.
-Grammarly plug-ins to check my grammar.
-DEI training sessions at my company.

Future State Journey Maps

In order to paint a picture of how the above personas might benefit from the UInclude tool, I created two Future State Journey maps for both Jaime and Kendall:

A future-state journey map that illustrates the journey for Persona A: Jaime Jones. She works in HR, and is often tasked with editing and supplementing job descriptions before they are posted onto LinkedIn or company channels. She requires a tool that supports a collaborative process with cross-functional teams. This may include a revision tracker, the ability to tag teammates, and ultimately being able to save the document as a template for future use.

A future-state journey map that illustrates the journey for Persona B: Kendall Kraft. He is a hiring manager, and has a lot of tasks on his plate. He requires a streamlined process in writing job descriptions, and relies on templates as a starting point. It is also important that his process supports collaboration with the recruiting and HR teams.

Problem Statement:

Jaime needs a space to collaborate on job descriptions with cross functional teams so that they can share markups and edits in real time and track all of the changes made in the editing process.

Design

After discussing the above research and synthesis with the UInclude team, we gathered both our UX Design team and the UInclude stakeholders for a 60-minute Design Studio Workshop.
We prompted all participants to produce quick sketches of proposed solutions for our primary persona, Jaime:

We all pitched our ideas to one another, and ultimately decided to build out three new features for the UInclude tool:

The Solution:

We developed three new features to address Jaime's needs, as well as to support UInclude's growth in adding new bias categories:
‍
• Filter by bias category
• Revision tracker
• "Learn More" drop down window

Color + Accessibility

In addition to exploring new features for the UInclude tool, our client also requested that we take a look at their color coding palette and strategy.
‍
We also heard in user interviews that beta testers were having a tough time discerning which colors were flagging which categories. Some also asked how this tool may work for someone who is colorblind.

We used the Figma Color Blind plug-in to choose a set of (7) colors that was the friendliest to most color-blind groups.
Of course, these colors do not account for Achromatopsia (complete color blindness), but we do solve for this need with the addition of the Filter-By-Category Feature.

I also went back and checked all of the proposed colors against the WCAG 1.4.3 Color Contrast Guidelines. All proposed colors exceeded the most stringent (AAA) ratio.

A snapshot of three suggestions we made to the UInclude team for their application of color:
1. Use only one color per category rather than showing masculine, feminine, and superlative words as separate colors. Show Gender bias as one color only.

2. Remove the green underline for flagging inclusive words. This was confusing in beta testing, and only adds to the cognitive load on the screen.

3. Show the inclusion score wheel as black and white. Adding color here competes with the colors used in highlighting biased words...and thus adds more confusion.

We also suggested the above revisions to the UInclude color strategy:
‍
• Integrating the 7 new colors into the biased word highlights.
• Removing the green underline that indicates inclusive language (this was confusing according to many beta testers of the tool).
• Revising the Inclusion Score wheel to a black and white design so as not to conflict with the colors used in flagging words in the text editor.

Iteration

Myself and one other teammate conducted usability testing of the Mid-Fidelity prototype while our third teammate updated the High Fidelity wireframes. We collected feedback from 6 individuals and made the following changes:

Keep on scrolling to see our final draft...

High Fidelity Prototype

A number of the opportunities for improvement listed below were picked up in the high fidelity prototype below:

Client Testimonial:

"It was an absolute pleasure working with Amy, Annie, and Callie. They're each extremely personable, which made the entire project duration so enjoyable. I really appreciated how communicative they were; every step of the way, my team and I knew exactly what to expect because of their consistent and thorough updates. They also made sure to do their due diligence and understand the space and industry in which we operate very well, instead of assuming expertise. The end result of all of their hard work far exceeded our expectations. Everything they produced was done in such a professional manner. It was clear that they put a lot of thought, research, and time into delivering a prototype that incorporates the feedback and asks we've received from our users. I'm looking forward to working with each of them again in the future!"
- Toshe Ayo-Ariyo, Co-Founder of UInclude

Next Steps

While we didn't have time to prototype all of our ideas in the three week time frame, we still left UInclude with a few more ideas to consider for future iterations of the product:

1. Providing users with the choice to "ignore" a suggestion, or to "Add to library" a word or phrase that they wish to not be flagged in the future. This addition was in response to multiple users' feedback requesting the ability to ignore suggestions that they wish to bypass:

A snapshot of three pop up windows that appear when a user hovers over a highlighted word in the UInclude tool. The pop up windows shown here include a trash can and "add to folder" icon to allow users to ignore a suggestion made by UInclude or to add a word to their library so that the word is not flagged again in the future.

2. We took a first pass at how users might add comments to the text editor, to communicate with their teams and to support the collaborative nature of job description writing:

A snapshot of three zoomed-in frames in Figma. These frames demonstrate an idea for a commenting feature. The comment indicator is shown as a black dot, hovering over a word. When a user hovers over the dot, a speech bubble with the commenter's initials pops up. When clicked, the speech bubble reveals the comment and allows the reader to ignore or resolve the comment.