Project Lincoln
Rethinking, how we get to know our clients in a fintech product.
Two years ago, we went through a project to make on-boarding personal information more human and less robotic. We really tried to mimic the experience of sitting at a desk with a Tax Pro as they learn about you. We took that and implemented a new flow we called the “simple flow” for our new users. Our general principles where:
- one thought per page
- not obsess over number of pages
- make the pages easy to digest with more friendly content to decrease the cognitive load
- make everything more conversational
We launched and saw good results, we decided to keep it the same for the next year with some minor updates to make way for larger changes elsewhere. You know how it goes, small team lots or priorities. We quickly noticed after launch that returning users who went through the simple flow the year before, are now not very happy with having to go through the old flow. Basically a classic, “who moved my cheese.”
This is how “Project Lincoln” started. Our goal? Increase the experience for our returning user and make it so that year over year, you are going through the same flow. We also focused on fixing the minor flaws we saw in simple flow and make updates to our UI to make our product more friendly across all devices.
Discovery Phase
We started by going through an exercise to see what the larger scope could be. We put a post it on the wall for each screen we went through, in the new and returning user flow. From there we combined them to try and create as many similarities as possible.
After finding quite a few similarities it brought up questions as to why we even had two different flows in the first place? It turns out we never really asked those questions due to looking at everything in such a focused and limited scope. It was just…how things had been. Just by doing that we are able to streamline and clean up the flow.
Bringing the band back
After we found out the general guidelines for what we wanted to build, we pulled together a cross-functional team to review this. We included people from the following groups:
- Dev
- Product Owner
- Content Strategist
- QA
- Tax Analyst
- Research
From there we walked through the idea and sketched wireframes together as a group. It was obviously pretty rough, but it got our ideas together for what we as a group thought.
Prototype, test, repeat
A rough prototype to demo the idea of what we are thinking was created utilizing InVision. This clickable wireframe wasn’t pretty, but it got the job done. Once we felt like it was in a tax accurate and user friendly, we took it to people around the company that don’t do their taxes online. We had them walk through it and give us their feedback, after about 20 sessions total we had some great feedback and a great direction to evolve before we bring in a real user focus group.
Make it real!
Now we have a prototype that we think is user approved, tax accurate and is possible for dev to build. We knew we wanted to work on making small tweaks to our UI to really improve the experience. If you have read my other article about tables, this was a part of that project.
We bit off a ton, from buttons, to form fields, to a couple new controls. We also did an entire facelift to the product as part of this. The main reason we did this was to increase our font size, make our controls easier to interact with and fix a lot of problems with our UI we have seen year over year. Our product had started to look dated, just because we have had so many higher priorities in the past.
We launched, now what?
Since tax season only happens once a year, we are still waiting on results from people actually filing their taxes. We obviously have done a lot of testing, but there is a difference between testing and your actually tax situation.
It’s never over in the world of product design, is it?
Now let’s look back, what could we have done better? Here’s a snippet:
- This was a large undertaking. As I mention in my other article, if we bit off smaller chunks we could really ensure that it was all implemented perfectly.
- Our dev and design team are always getting better at working together, but right now we do our QA off of if they match the functional specs. With that, we tend to have a lot of design bugs.
- A lot of time was wasted with things we didn’t know would be involved in the scope, it’s hard to fully understand all the different personas. We could create more inclusive personas that go through a few different scenarios to more easily understand the main personalities.
We just started our retro process on how we did this past year. Based off our learnings from this project, we are starting a new one to grow upon what we did this year. Below is an example of the timeline we are hoping to hit, right now I am in the process of creating those personas.
As we get results and learn we will continue to iterate.
James Jensvold is a Chicago based UX designer.
