Background
The eDiscovery software Relativity allows users to review a wide array of documents. One particular document type, a transcript, is the transcribed text of a recorded sworn, out-of-court oral testimony; a deposition.
The Transcripts application gives users the ability to upload, review, code, annotate, and watch transcripts within Relativity to take action and to better prepare for trial.Identifying the Opportunity
Ideas in the organization come from user submitted ideas to an online portal, speaking with customers, user and market research, and other forms of discovery. There were a number of feature requests from our transcripts application users for video clipping specifically. Once the feature was identified, the leaders of the product triad met to brainstorm a list of possible features.
To better understand the problem, I identified our key users as those who submitted feature requests and heavily use the product currently. I reached out to those key users to get more insight with goals to:
• Discover the main use cases
• Prioritize a list of features for MVP
• Learn what users use today
Using SurveyMonkey, I sent out a survey to identify those pain points. The two biggest pain points mentioned were inconsistencies in file formats and moving/organizing files. Certain formats only work in certain tools, and moving clips between tools is difficult. Other respondents mentioned network and machine speeds as issues, along with the need for a high-end video card or stable network connection. With large files, encoding issues arise when exporting clips, and those clips' audio or video sync can be off. I also provided a list of potential features the participants could rank as most to least important.
User's Goals
To better understand what our users' are looking to accomplish, I also set up a few calls for more in-depth interviews. During these conversations, I found that attorneys and case administrators want to use video clips for witness and trial preparation, as well as for use in trial presentation.
Through these conversations, I learned that users ultimately want to insert these clips into a presentation tool like PowerPoint, or to show at trial for impeachment.
At the time, there were a number of tools available to get that job done. After doing some competitive analysis, I was able to understand the type of user interfaces our users were accustomed to in terms of creating video clips with other transcript and deposition software.
Competitor 1
Competitor 2
Competitor 3
With this information, I was able to narrow my focus on an experience to create a video clip with the ability to acutely tweak start and end times, based on a portion of the transcript.
Getting to Work
Working within my triad of UX, PM, and Engineering, we collectively decided on the technical and business requirements for the project. Given the use cases we discovered, the team decided to focus on four key workflows:
Using feedback from the team, inspiration from competitors, and the constraints of our agreed upon goals, I set out white-boarding the solution for further review from internal stakeholders. Based on conversations with users, it was imperative that the video clipping function be streamlined with the rest of the Transcripts interface.
Working with an offshore 3rd party developer, much of the collaboration and communication happened virtually. I did take advantage of their occasional office visits to host brainstorming and white-boarding sessions.
Early Iterations
Once I was happy with a general direction, I started to create higher fidelity prototypes to both show and test the intended interactions. I experimented with a range of sliding panels, modals, and other user interface elements to allow a user to create a video clip.
Initial designs presented a modal on top of the transcript to create video clips. Users pointed out shortcomings in this design with respect to window sizing and the inability to navigate the transcript once launched.
A strong design emerged through rounds of iterations and internal feedback. From there, I set up a usability test plan to determine if users could complete the tasks assigned.
These usability sessions were performed remotely, providing participants with an interactive prototype and a list of objectives to complete. I tested multiple interactions across a few designs to ensure usability. The findings from these usability sessions lead us to determine which features would be included in the initial release, using the following criteria:
• Working Well
• Keep & Tweak
• New v1 Requirements
• Remove
• Future Possibilities
Sprint Planning
Once the triad agreed upon a design, I worked with my engineering counterparts and the product's PM to break up the video clipping story into smaller pieces, to be more consumable by developers. This ensured that, while the feature is being built, I am still able to iterate on other pieces of the design in parallel.
Using a variety of tools, from collaborative design software like Figma & Axure, to story tracking in JIRA, handoff was smooth between design and engineering.
During design and development, I made sure to follow accessibility guidelines to ensure the feature was available for both users who can control a mouse and keyboard, as well as those who rely on other types of assistive technologies, like screen-readers and voice recognition software.
Looking Ahead
Since the feature was released, the team has been able to add a number of enhancements. From our research in discovery for this product, we built a healthy backlog of future possibilities that didn't make it into the initial release. Some of these enhancements have included a wider support for file types, merging video clips, and broader exporting capabilities.
Additional features we are exploring are the ability to redact portions of a video, create clips based on designations, automating clip creation, and more ways to integrate video clips throughout the software. Much of the information we learned from the early days of discovery is still very valuable in improving the product today.