Redefining the data entry process through an AI-learning partnership with Microsoft.
XDC is a platform built on top of a machine AI capable of learning how to fill out invoices and billing receipts from human input. It outputs simple analytics to ensure quality control, and simplifies the role of data entry employees on what can be a complex process.
View PrototypeLead designer & researcher
At AvidXchange, manual data entry of client invoices was costing the company nearly $2 million annually through outsourced labor. To reduce costs and accelerate the invoice intake process, we partnered with Microsoft to implement a cutting-edge Optical Character Recognition (OCR) solution.
"Our turnaround time for invoices is 2 weeks. That's just too long."
The current system contained many flaws in the user experience. Flaws that would trip up the AI and the users alike. Speaking with several indexers I identified several problem areas.
Create a data entry system using Microsoft's OCR software to reduce overhead times on time to customers and reduce complexity for employees.
We had the opportunity to address some of the problems with the current platform and reduce the friction in the data entry process. The AI on the backend could help resolve some of the complexity issues as it was trained. This would free our internal indexers to do what they did best: auditing.
From the prior investigative research and journey mapping session I was able to sit down and create some of the first designs for presentation. They were a starting point for discussing what we would like the future of the product to be.
A/B testing determined that Option B was easier for our indexing team to understand. The fields were closer to what they were accustomed to seeing in the old tool. Option B also simplified the color coding. Labels were added to the outlines of the corresponding data to aid our coworkers with vision issues in understanding what they were looking at. The biggest challenge was coming up with 24 distinct and accessible colors for each field that could happen.
B. It's what I'd expect to see. Easier to read.
The designs were refined with feedback from the meeting. This design would be the one to go into user testing.
Method
Unmoderated User Testing through UserZoom
Target Audience
6 Indexing Employees of AvidXchange
I conducted this test to ensure indexers found it easy to use and didn't have to refer to knowledge articles on the subject. This data was then correlated and presented back to the stakeholders, product management, and development leads. The feedback resulted in more design refinements and a finished product ready for hand off to development.
I implemented changes to address user testing feedback. In parallel, I was overseeing the rollout of an updated design system for the company, with XDC identified as a key candidate for the transition.
We did a soft launch with less complex customer invoices. As a metric for comparison we had both the old system and XDC process the same invoices.
Turnaround time for invoices to process and show down to 1 week from 2.
Rating went up from 16% to 90$
The company was able to cancel 1 contract from the 3 outsourcing companies, saving on costs.
During this project I learned the valuable lesson of how to juggle the need for accessible features without obstructing the use for everyday uses with the labelling system. There was also value to be added during the ideation phase to present visuals to the stakeholders for new project endeavors where there is no system to show as a reference point.
This was the first time where I had to work with an AI as one of the primary user personas. It was an interesting experience that I would like to revisit in the future. In short, the AI has needs too!
Want to hear more about my experiences and how I can help your products?
Let's chat