Senior UX Designer
•
Lytx
•
4 months
•
2023
Lytx Video Overlay
Designed a dynamic video overlay to simplify customer coaching conversations, improving user trust and reducing contention rate to near zero.
ROLE
Senior UX Designer
Company
Lytx
Duration
4 months
Year
2023
The Challenge
Fleet managers using Lytx video safety hardware needed to coach drivers on following distance violations where they followed a vehicle at an unsafe distance — the raw video footage lacked context. Without an objective visual reference, coaching conversations frequently turned contentious. Drivers disputed interpretations, managers lost confidence in the tool, and the product's core safety value was undermined.
The Solution
An augmented video overlay that helps coaches and drivers visualize following distance behaviors, providing clarity to support productive coaching conversations. Downloadable video with the overlay displayed extends this capability further.
"I would say on the following distance piece, you guys are leaps and, not a little bit, you are light years ahead of the others [your competitors]."
— Rick, Safety Manager, customer interviewed on usage of Following Distance Overlay
Outcomes
↑ Satisfaction
Customer Response
100%
Customer Enablement
↓ Contested Events
From Drivers when Coached
The process
1
Discovery & Research - Summary
Conducted field interviews with fleet safety managers across multiple customer accounts.
Mapped the anatomy of a coaching conversation from event trigger to driver acknowledgment.
Found that disputes arose not from the event itself, but from the ambiguity of raw footage — a gap between what the machine measured and what a human could see.
Users
All Lytx customers who were coaches and safety managers.
Needs
In-video context or a method of transparency that shows why the camera AI detected following distance behavior.
Problems
Frustrations within their organization regarding drivers not believing the video was accurate when reporting following distance behavior. This led to unproductive coaching conversations where the driver and coach couldn't agree on the based facts.
Findings
Request for a video overlay type of feature was a top requested item from customers when discussing coaching pain points, even unprompted.
Customers wanted an contextual video feature to display by default so there would be no mistake regarding what the camera was detected.
Transparency builds trust.
2
Definition - Summary
The core problem: the video alone was ambiguous. Drivers could reasonably dispute a coach's interpretation without an objective visual anchor. The overlay needed to make the machine vision system's measurement legible and indisputable to a human audience, without overwhelming the video or requiring technical literacy to interpret.
Opportunity
Given the machine vision cababilities already built into the AI detection model, we could leverage this data to provide better transparency to customers and facilitate coaching conversations in which driver contest the evidence.
Personas
Coach - Meets with drivers when able to find them before shifts, and mentors them on how to improve driving behaviors.
Wants/Needs
Ability to prove the captured event video is correct with direct evidence rather than relying on alternate methods like using a stopwatch.
Ability to download the video with the overlay still displayed.
Understanding of context for when overlay will display versus not display.
Pain Points
It's difficult to get drivers to listen to constructive feedback when they are focused on refuting the video.
Lack of in-video context like a road marker makes it difficult to prove the following distance timing when using a stopwatch.
3
Design - Summary
Explored overlay treatments ranging from box outlines around vehicles to road overlays and birds eye view. Tested 3 concepts with safety managers over the course of a month. The winning design used a color overlay most visible with different road colors (concrete, asphalt, dirt, night vision), overlaid directly on the video feed with real-time distance values and reasonable refresh rates at a per-second cadence.
Ideation
Experiemented with multiple color combinations over various road types to find best visibility averaged across all road conditions.
Designed road overlay display along with detection boxes around vehicles, ultimately opting for minimal video overlay elements to provide simple data transparency.
Specified following distance timing refresh rate to 1 second for improving readability and understanding.
Incorporated method for downloading video with an overlay without disrupting existing functions.
Designed overlay methods for other road behaviors, but scaled down to Following Distance initial launch.
Prototyping
Built prototype simulating overlay on top of video for early customer feedback.
Tested multiple concepts for overlay control while designing forward-thinking options for future video experience improvements.
4
Delivery - Summary
Shipped to production. Monitored coaching session outcomes and user sentiment via customer interviews and follow-up surveys. Contention rate dropped to near zero within the first month. The overlay became a widely requested feature among Lytx customers who didn't already have it enabled, and none toggled off the feature once provided.
Result
Customers reported full satisfaction and trust in the accuracy presented to them, several reporting a reduction of driver-contested following distance events to zero.
Lessons Learned
Simple and quick prototypes work well at gathering early feedback during the design phase and can help validate design choices without halting design work.
