Bryant Peng
Published in

Bryant Peng


Accelerating legal discovery with AI

Everlaw is a cloud-based ediscovery platform. When I joined, the company had been around for a while and was starting to expand rapidly. With the recent switch to a monthly release cycle and design requests pouring in from other departments, our two-person design team had more than enough to do.

What’s ediscovery?

Before a trial, both sides are required to exchange information. This process is called discovery, and involves turning over documents related to the case. When the documents are digital, it’s called ediscovery.

There’s a lot to go through; these days, our digital trail dwarfs our paper one. Legal teams use ediscovery software like Everlaw to produce documents for the opposing counsel, as well as review produced documents for evidence.


I was part of a team comprised of the CEO (our de facto product manager), one other designer, a release manager, and about ten engineers. As 1 of 2 designers, I worked on everything from product features to conference flyers.



A new set of priorities.

I’d never designed for enterprise before, and it had its own unique challenges. The relationship with users is fundamentally different: people who use Everlaw really use it, up to 8 hours a day.

As a designer, this meant striking a balance between simplicity and efficiency. Adding steps to a flow can make it more user-friendly, but repeated hundreds of times in a day, it becomes grating.

There’s also a financial cost: a typical case involves millions of documents, and document review isn’t cheap. In one of the Apple v. Samsung cases, Samsung paid $13 million in ediscovery costs for 11 million documents. Saving a few seconds here and there adds up.

A limited toolkit.

Design is problem-solving within constraints, and the legal industry is not without. We couldn’t observe the product in the wild because Everlaw handles sensitive information. Analytics were limited for privacy, and rarely used in decision-making. Feature impact was measured through customer feedback, not metrics. I had to be resourceful with a limited toolkit.

A lack of processes.

Many of the processes I was used to didn’t exist, so I created them. In addition to formalizing processes within the design team (guidelines for designing features, updating the icon library, etc.), I collaborated with customer-facing teams to implement user research methods.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store