Project Description
Patriot Link
Introduction
User Problem
Three User Groups
– Code of Support Staff
– Other Non-Profits & Social Workers
– Veterans
While the app worked great for the Code of Support Staff it was too complex for a veteran user to navigate easily. Based on exit rates and bounce rates we could tell that veterans were not able to find the services they needed within a reasonable period of time. According to our research, they regularly ended up calling Code of Support directly which was a scalable solution.
Business Problem
Code of Support is a non-profit that would like to monetize it’s service so they are not reliant on donations. Current attempts at doing this are attempting to charge other non profits for the ability to search the system which has consolidated and verified all potential veteran services into one location. Other non-profits are only incentivized to recommend and provide the serices they offer which they are already fully aware of. Put simply, these non-profits do not need help finding their own services and have no interest in finding others similar to what Code of Support does.
Solution
The solution was a bug free, simpler, mobile optimized user experience. Per our initial research findings we also suggested re-designing and building out search. However, due to the nature of our contract and the funding for the project our intial scope was limited and deliverables were set. The new app would have an updated visual design giving it a more modern look and feel.
Expected Results & Success Metrics
The new Patriot Link would provide an overall simpler and more enjoyable experience for all users and would then lead to more people using it and getting the services they need. The data did not support an increase in the number of non-profit admin users. We made the Code of Support team aware of this but there was little we could do to change the current priorities and focus.
1. Lower bounce and exit rates.
2. Longer sessions times.
3. Increase in veteran users.
4. Increase in mobile users.
5. Increase in non profit users.
6. Attract the attention of the VA
Logistics
Scope
The 6 month long engagement started out with strong focus on fixing bugs, serious usability issues and then transitioning to overall design improvement and mobile optimization. Four months into the project we were able to act on some of our inital feedback and make improvements to search and a other parts of the application that impacted veterans. To jump start this process we ran a design sprint with the clients, did several interviews and then tested our prototype with real veterans that had used the app before. We were then able to iterate on our design but also had to consider thatour dev team only had so many hours left they could allocate towards search and had to make several tradeoffs.
My Role
Design time on this project was split between myself as the Lead UX Designer and a UX Designer on my team.
Main areas of responsibility:
– UX analysis (flows and persona)
– Design (app, website)
– Lo-fi prototyping (basic flow and interactions)
– Consultant in bi-weekly board meetings.
– Account Manager
– Sales point of contact for initial meetings
– Design Rep in Patriot Link Executive Meetings
Team Make Up
| Team Role | Hours / Sprint | Total Hours |
|---|---|---|
| Product Manager | 40 | Concluded |
| UX Designer | (40 hours per sprint *4 sprints) 20 afterwards | Concluded |
| Front End Dev | 80 | Concluded |
| Back End Dev | 80 | Concluded |
| QA | 20 | Concluded |
Solution
Mobile Home Page
New home pages for mobile and desktop were key to starting the experience off right. We wanted to make it clear the point of the application was search. We also wanted it to be clear how to do it. In addition, we needed veterans to know that others had used this program to their own benefit. You can see this on both the mobile and desktop mockups.
Desktop Home
Improved Mobile Search
Mobile Search pages were key to what we were improving upon and testing with our users as the data indicated that the largest group of user would be mobile veterans. Here is some of what we focused on:
- Letting users choose between just typing in their query and selecting from broad categories to narrow down their search.
- Showing immediate results based on limited information based on location.
- Letting the user narrow down search results via filters.
- Enabling deeper dives and more specific results via secondary sub-category filters.
- Breaking down search result detail pages into easily identified categories so users can identify if the program fits with their needs.
Improved Desktop Search
We made sure the new and improved search was translated to desktop. The focus and goals were the same. Most of the desktop users were the Patriot Link Admin team. Even though they played a major role in the initial design of the original search, and had significant subject matter expertise, they were all excited about the changes and found it quicker and easier than the previous design. The biggest difference between the desktop and mobile is that the extra screen real estate allows you to do more on each individual screen.
Simpler Mobile Results & Detail Views
Improved Desktop Admin Pages
The goal of recreating these pages was to make the data the application was collecting significantly easier to view and turn into actionable intelligence. With that being the goal we took the current pages and recreated the information architecture so there would be significantly less information on each page, but hopefully it would be much clearer to the user.
We also let the user choose what they wanted to see via different fields and how they wanted to see it by filters, sorting, dates and an option to switch from a table to a chart.
Code of Support Admin Search Metrics Bar Graph
Code of Support Admin Overview
Code of Support Admin Search Metrics
Outcomes
- 500 users to over 13,000 veteran users
- Longer session times and lower exit and bounce rates
- More veterans finding the services they need
- Fewer phone calls into code of support from Patriot Link users
- 70% of new users are mobile
- Partnerships with the VA and other organizations
Discovery & Strategy
THE LONG GAME
- Stakeholder Interviews
- UX Audit of current application.
- UX analysis.
- Assessment of need for hot-fixes based on functionality & usability.
- Assessing Levels of Effort with Engineers
- Our team had responded to an RFP. The project was being run like many government projects and many of the priorities and objectives were defined with very little room for interpretation.
It took time to build trust and respect, but being invited to be a part of executive meetings for PatriotLink and voice user concerns was extremely helpful to our team and the project. It eventually lead to our design sprint and some of the biggest changes to the application.
Definition & Research
Target Demographics Information
Veterans were are the primary target market for this application. More specifically veterans that were in need of services provided by a host non profits looking to help. Unfortunately there is a massive amount of red tape and requirements veterans have to go through in order to get the help they deserve.
The third target demographic is code of Support itself. They use patriot link as an internal tool to find exactly what veterans need within their immediate area
as quickly as possible. That being said they are professionals who train with the tool and use it 8 hours a day and had a lot of input into how it was built. This did not help the primary end user and the people who need the most help, veterans.
Next is non-profits. Code of Support sets out to help any and all veterans find the services they need no no matter what. They thought that other non-profits
would want to do the same and would be willing to pay them in order to be able better help veterans. Unfortunately this assumption was incorrect and while
other non-profit workers were willing to refer veterans who did not need their specific services to Code of Support or the VA, they were not willing to spend their
time search for those services on another platform, or willing to pay to utilize that platform
Short Term
Aside from mobile optimization and an overall improvement to the user experience, the short term focus was in monetization of the platform as it currently was.
American Express had assisted PatriotLink in Sales and Growth Strategy with segments focusing on:
- Veteran Specific Organizations
- Social Workers & Case Management
- Large Non- Profits
Their research indicated that these groups were already in need of and would be willing to pay for the services PatriotLink offered.
This conflicted with our research.
Long Term
A large hurdle for the application was the focus on non-profits as opposed to veterans.
Once the focus changed, we were able to do a design sprint that specifically focused on improving search for veterans. When we brought in veterans for interviews, they all echoed some similar ideas.
- Veterans that need help will likely not get it right away.
- This can lead to the problem getting worse.
- When they need help, they need it now, yesterday or last week.
- Many veterans do not understand how to navigate the complex red tape around getting help. An example of this is knowing what percent disabled you are.
- The simpler the better.
- The faster you can get them to an action the better, even if it does not work out.
Why does it matter?
How veterans are helped and what their pain points are was critical in us being able to provide them the services they need and enabling them to find them through search quickly and easily.
23% of veterans have some type of disability.
8.2 % of veterans are at risk of homelessness due to poverty and 1.4% of veterans are homeless.
Analysis & Ideas
Moving Forward!
After gathering data from the stakeholders and from our research. We were able to identify the following opportunities.
- New Information Architecture improved Navigation for mobile users & desktop users.
- The new information architecture would also improve navigation for PatriotLink Admin, veterans, and non-profits which were still a focus initially.
- To create a new IA we utilized information gathered from our interviews to assess what each specific user group was using and how often.
- We Also took into account feedback provided from a Systems Usability Scale which gave the app a 42. A passing grade was a 68.
This helped us simplify the application even though were enabling users to do more.
Red Routes
Basic End User (Veteran)
Code of Support / PatriotLink Team Member
Premium Account (Non Profit / Social Worker)
Premium Account Admin (Non Profit / Social Worker)
Design & Iteration
Previous Design Example
- The initial focus was on mobile optimization.
- The application did not work on mobile devices at all.
- Funding, deliverables and features were locked in place.
- We had flexibility in how these were implemented.
For Example: There was a focus on monetization and a plan to implement an enterprise mode and a personal subscription plan. Our research indicated these were not things non profit users were willing to pay for, but were locked in place.
- Personal Subscribers were simply given unlimited searches.
- Enterprise accounts were given unlimited searches per each user and limited access to data Patriot Link already pulled from searches to try and help veterans.
An example of what the application looked like when we started.
Iteration #1 Veteran Users
These screens show the basic end users activity. Research indicated that many users found results but did not reach out to those services.
There was also an extremely strong business case for getting data about organizations that provided these services.
Search Criteria Breakdown and Initial Results Details
Stakeholder felt very strongly about users registering to use the service.
The data showed that the majority of veterans that needed help would need multiple services. If we created a profile for them with the information they would have to enter to search, subsequent searches would be much easier.
We initially wanted to remove the location details at the top of search results and only show results based on proximity. Stakeholders made the case that many service members traveled and would need to search for services not based in their current location.
The goal was to minimize information within our initial search results. Below you can see the detail views for the Programs which show up as results.
Detailed Program Search Results Breakdown
Iteration #1 Non Profit / Social Worker
The biggest difference for the Non Profit Worker or Social Worker is that they will have multiple cases. Patriot Link will not store any personally identifiable information including names so these are broken down into case ID’s where the user can store information about that specific case ID so they can reference it again.
Case Review & Case Profile Details
Research Data indicated that most users who need help will need multiple services over a period of time. Enabling case workers to store specific information about the case so they do not have to type it in every single search would save tons of time and gather valuable insights about the services veterans needed that would not be tied to specific names.
Basic User Desktop (Veteran)
One of the difficulties Patriot Link had i improving their program was that people would come and do a search and Patriot Link had no way to tell if end users were actually reaching out to the services they searched for and getting the help they needed. We made it easy for veterans to say whether or not they contacted the program found via their search results. Depending on their answer we tried to capture as much information as possible regarding the situation and their experience. Part of the idea behind this was that it could potentially be monetized with larger organizations.
Implementation & Collaboration
ENGINEERING IMPACTS ON DESIGN
The biggest impact engineering had on design was that number of overall hours with the roadmap. As with many projects, the list of wants is much bigger than the
budget available. This meant moving forward with minimal simple designs that could accomplish a desired outcome with absolutely no waste.
Another challenge with this project was that the engineers were not native English speakers so communication was not the easiest at times. This meant that the design team had to go the extra mile in order to communicate our ideas to the engineering team and in turn provide rough levels of effort to the PM.
With our limited hours our design team had to design iterate fast and clearly communicate concepts at times with out the use of language.
Evaluation & Observation
USER DATA (PRE DESIGN SPRINT)
We did not have access to all of the user data coming into the site. What we could see was:
- Number of users
- Locations
- Device Type
- Search Categories
- Military Branch
- Service Connected Disability
WINS
- We were able to see an increase in users after our initial changes and mobile optimization.
- A large portion of these users were mobile users as well.
- Many users also started using the app more than once.
DESIGN SPRINT PROTOTYPE QUOTES
“I have neck pain I want to go to physical health. I am not going to search under my medical benefits. I don’t care about those.”
“Location and Proximity would drive me above other factors. I have kids and my time is valuable.”
“I would always search first by what I was doing, then location then other filters.”
“Showing testimonials is a great idea. My partner needed help and would not go get it until his friend told him about the help he got and how it worked. Lots of people are not sure if it is worth their time.”
“I would not look for WIC under benefits. When I think benefits, I think military benefits, claims, compensation, not food vouches or WIC. I would think some of these services could fit under multiple categories.”
“I like how it says that it is vetted veteran services up at the top. It makes it clear what this is.”
TAKEAWAY
Based on our feedback from user testing and interviews. Re-organizing the categories and filters that determined how search worked would result in a vast improvement to the application for Veteran Users. We collaborated with the PatriotLink Team take these from something created at high levels within military organizations and make it easy for your average veteran to understand. Unfortunately, an engineering assessment revealed we would not have the time to do this so we optimized our designs without this change.
Conclusion
Short term execution for Long term benefit
Some of the data points that made this project interesting.
- Locked in deliverables at the start with limited flexibility.
- Data from our user research did not match up with previous companies the client had worked with.
Having a conversation about where the product is going with stakeholders can be tough. Especially if you think that is might be going in the wrong direction. As and Agency some people might just say that it is better to do what you are told and get paid as opposed to stirring the pot. This is not a mindset I agree with.
We were very fortunate that our team was able to do great work and earn the trust of the Patriot Link team. This in turn led to us being able to focus on making some changes that had a real impact on veterans looking for help and also helped Code of Support forge valuable partnerships.