Aunt Bertha, Inc
2016 - 2018

Industry: Tech startup
Product: Healthcare & social services platform

COMPANY

Aunt Bertha, Inc is a Certified B Corp (aka a Social Impact company). The mission is “to make human service program information more accessible to those in need in order to help more people reach self-sufficiency. Aunt Bertha picks up where Uncle Sam leaves off by making it easy to find food, health, housing and education programs based on need. By organizing the world’s human service program information, people can easily find out which programs they qualify for in a matter of seconds. Aunt Bertha also helps human service organizations administer programs better by offering easy-to-use web-based case management software.”

MY ROLE

I was first hired by Aunt Bertha for a design research contract. After delivering insights, problem statements, and possible solutions for my research, Aunt Bertha realized that they had plenty of work to keep me busy as a full-time in-house UX Designer. During my two years at Aunt Bertha, that initial research was a constant thread for me in empathizing deeply with our users and always considering what affects any new features, tools, or overhauls might have on them.

LEARNINGS

There was so much to take away from my time at Aunt Bertha, but the major things I carried with me when I left were

  • Social Impact companies face a lot of the same challenges in resource allegation & availability as non-profits

  • Mission-driven companies rely on their employees’ commitment to their values and, as such, empathy for the user is vital to the morale of employees

  • Research isn’t always in-person and there are a lot of ways to learn about our users that are incredibly valid

  • Just because the research shows something should be done doesn’t always mean that it’s the priority for management.  Learning how to get buy-in from the stakeholders is essential for UX work at small companies (and, likely, everywhere).

  • When I was a Project Manager, I loved working in small companies because I was empowered to jump in and fill many roles, fixing inefficiencies & creating process solutions; however, that does not seem to be the case with UX and I sincerely value organizations who have the resources & tools to be able to do the job efficiently.

METHODOLOGY

Fresh out of design school where I’d spent a lot of time researching with individuals experiencing homelessness, I was eager to utilize as many of my newly-acquired design methodologies to the fullest to dig deep into the problems that users faced with our site.

Inspiration

Aunt Bertha’s Mission Statement -- To connect all people in need and the programs that serve them (with dignity & ease).

Research & Synthesis

RESEARCH

During design school, I’d spent my first quarter doing research with people experiencing homelessness and the various services with which they engaged.  The Insights & Design Implications from that project served as the foundation of the struggles I knew many of our end-users faced.

Insight 1: Services are created to fit the lifestyle of the service providers

Insight 2: The internet connects people experiencing homelessness with jobs that fit their lifestyle.

Insight 3: The requirement of physical objects keeps people experiencing homelessness...homeless (photo IDs, birth certificates, social security cards, a phone bill).


I also did a lot of new research to better understand Aunt Bertha’s unique users:

  • Heuristic evaluation of all features & tools of Aunt Bertha site

  • Contextual inquiry with 

  • People in need

  • Employees & volunteers at agencies that offer services (hospitals, healthcare offices, food pantries, churches, charities, etc.)

  • Internal team members using admin tools

  • Analysis of site’s search outcomes (i.e.; how many zero results & why)

  • Touchpoints, systems, and language from client-purchase through acquisition internally & externally

  • History of the Aunt Bertha taxonomy, future, and best-practices in healthcare standards

SYNTHESIS

Since the design team at Aunt Bertha was new, I helped create the environment for us to collaborate, purchasing foamcore, post-its, butcher paper, whiteboards, and all the basic War Room accomodations, so we could better start externalizing information.  

Some tools I used to synthesize were:

  • Posting contextual inquiry/interview transcriptions on the wall

  • Affinity diagramming to uncover patterns

  • User flows & customer journeys of the site, tools, and features

  • Storyboarding

  • User personas of both external & internal users

  • Participatory design with clients and end-users

Of course, any company will face numerous UX challenges over two years, but here are some of the highlights that spanned my time at Aunt Bertha for which I made the most impact:

Insight 1:
Problems with Search (the foundation of the product) did not work intuitively for most users in many cases, so users lost trust in the system before they even had a chance to begin.

  • Zero results for things like misspellings, program searches without using exact name matches, and synonyms for tags that actually exist in the Aunt Bertha taxonomy

  • Taxonomy hierarchy is designed for exploration, not research, which means that users can only see top-level tags at first encounter with site, but results page displays secondary, tertiary, and quaternary tags which don’t always relate to specific tag searched.

  • Map covers above-the-fold frame, confusing users as to what purpose the site serves.

Insight 2:
The Aunt Bertha taxonomy had no official process for new tag acquisition, so the squeakiest wheel got the grease and sometimes in ways that made it harder for users to find what they were looking for.

Insight 3:
Inconsistent language used across teams related to product terminology & offerings

Insight 4:
There is no such thing as a unique Seeker persona. Anyone can be a Seeker at anytime.

Generating Ideas & Prototyping

Brainstorming & collaborating with colleagues, whiteboarding, and sketching out ideas independently or together is always a lot of fun. I always enjoyed our design review sessions where we’d select what to keep, remove, or merge, then refine ideas with wireframes, and on to digital mockups.

Sometimes, we’d have time to prototype our idea with users, but because of the fast-paced environment, many ideas went straight to production after simple user-testing internally or with each other and would be improved later.

Insight 1 Suggested Solutions:
Problems with Search (the foundation of the product) did not work intuitively for most users in many cases, so users lost trust in the system before they even had a chance to begin.

In order for Search to be the most successful, we need designs that will

  • Display typeahead tags or programs in the search bar upon user’s initial interaction

  • Display a misspelling prompt 

  • Display results for partial matches

  • Help users understand search’s category navigation by unveiling all taxonomy

  • Give users buy-in to product right away, by relocating map to side of page

Insight 2 Suggested Solutions:
The Aunt Bertha taxonomy had no official process for new tag acquisition, so the squeakiest wheel got the grease and sometimes in ways that made it harder for users to find what they were looking for.

In order for the taxonomy to be fully understood, we must have stakeholders meet up regularly to understand the history of Aunt Bertha taxonomy, what challenges each team faces with new tag creation, and best-practices for the taxonomy’s future.

Insight 3 Suggested Solutions:
Inconsistent language used across teams related to product terminology & offerings

Create a special service design project to analyze cross-team & customer flows, identifying fail-points and solutions to tackle the language inconsistency challenge & other inefficiencies.

Insight 4 Suggestion Solutions:
There is no such thing as a unique Seeker persona. Anyone can be a Seeker at anytime.

Design team should create a company-wide workshop to build user empathy.

Implementation

[add something here] 

Insight 1 Implementation:
Problems with Search (the foundation of the product) did not work intuitively for most users in many cases, so users lost trust in the system before they even had a chance to begin.

Since the company was also juggling client requirements, new features, and bug fixes, in order for Search to be the most successful, we needed to implement quick implementations, first:

  • Display typeahead tags or programs in the search bar upon user’s initial interaction

  • Relocate map from taking up top of page to only a small icon on the side of the page

Insight 2 Implementation:
The Aunt Bertha taxonomy had no official process for new tag acquisition, so the squeakiest wheel got the grease and sometimes in ways that made it harder for users to find what they were looking for.

I created a committee comprised of stakeholders from every team in the company, so they could share the impact of our taxonomy on their unique roles focused on understanding the taxonomy, creating a shared perspective, and developing processes, called the Tagging Committee.

Insight 3 Implementation:
Inconsistent language used across teams related to product terminology & offerings

Embark upon 6 month service design project analyzing cross-team & customer flows, identifying fail-points and solutions.  Initial simple solution to create shared docs instead of separate docs containing duplicate information across teams.

Insight 4 Implementation:
There is no such thing as a unique Seeker persona. Anyone can be a Seeker at anytime.

Design team create a company-wide workshop to build user empathy.

Validation & Iteration

[add something here]


Insight 1 Validation & Iteration:
Problems with Search (the foundation of the product) did not work intuitively for most users in many cases, so users lost trust in the system before they even had a chance to begin.

Quick implementation of typeahead in the Search bar did not significantly decrease zero results activity.  An issue with irrelevant results eventually replaced the Zero Results issue and created more difficulty with tracking that issue, but over time, improvements to search were updated again & again.

Some examples:

  • Best vs Possible match feature

  • Updates to logging when only Possible matches are found

  • Updates to hierarchy of program fields being matched

  • Updates to location mapping

  • After full taxonomy unveiled in category navbar, search for one of our major tags increased by 341%!

Insight 2 Validation & Iteration:
The Aunt Bertha taxonomy had no official process for new tag acquisition, so the squeakiest wheel got the grease and sometimes in ways that made it harder for users to find what they were looking for.

In order for the Tagging Committee to tackle understanding the hundreds of tags in the Aunt Bertha taxonomy, I put all of the tags on a wall for everyone to see & truly grapple the extent of the taxonomy, physically. Since the wall was in a centrally-located area, every single person in the company got to see the enormity of the task, at-hand, which created empathy for our data team as well as increased awareness of the problem & response time for solutions.

By externalizing these hundreds of tags, we were able to easily see issues with duplication and perceived hierarchy. For example, the tertiary tag “books” was under both “education” and “goods”, implying that the “books” tag had some relationship to the category it was under, but when you click on “books”, all types of books are displayed — a challenge for most user’s concept of the product.

I led the Tagging Committee for nearly all of the 2 years I spent at Aunt Bertha. The team served to springboard a lot of collaboration across teams and create empathy for each other & our users. In total, 23 new tags were created during my time there, with 9 tags on the horizon after a taxonomy evaluation of how our Mental Health category paralleled to health industry standard taxonomy.

Insight 3 validation & iteration:
Inconsistent language used across teams related to product terminology & offerings

Initial simple solution to create shared docs instead of separate docs proved useful, while iterating on tools requiring development hours. Suggested improvements to process, but a clear fail point with the admin platform urged an entire overhaul of the client tools, which I spent mocking up, testing, iterating upon, and creating high-fidelity designs for until I left the team.



Insight 4 validation & iteration:
There is no such thing as a unique Seeker persona. Anyone can be a Seeker at anytime.

Many team members reached out to talk about their experience with the user empathy exercise. Teams started to collaborate more and hold meetings together to try to solve shared problems together.

RECOMMENDATIONS

Quotes from colleagues

Misty, you think so big, but never let the small details slip.
- UX Designer

You always bring so much thought & compassion to everything you do, every step of the way
- QA Team Member

You are constantly making things better wherever you put your attention.
- Anonymous colleague

You are the client whisperer when it comes to tricky configuration.
- Customer Experience team member

The Sales to Service form is a work of art. It’s like a zen garden. Thank you for making that.
- Customer Experience team member