Why this matters
Surveillance footage sits at the center of most liability disputes. A slip-and-fall at a grocery store, a fender-bender in a parking lot, an altercation outside a bar. You have the video. The other side has the video. Everyone agrees on what happened, yet settlement talks stall because each party reads the same 90 seconds completely differently.
The problem isn't the facts. The problem is how a jury will feel watching those facts unfold. You can rehearse your narrative in the conference room a dozen times, but you won't know if it lands until you're already in mediation or trial.
EmotionTrac gives you that answer before you walk into the room. The platform recruits opt-in panelists who match your venue demographics, records their facial expressions frame-by-frame using the Facial Action Coding System, and maps second-by-second emotional responses to your video. You see exactly where attention drops, where confusion spikes, where anger or sympathy peaks.
The workflow
Step 1: Define your audience and upload the video
Start by identifying the demographics that matter. If you're heading to mediation in Cook County on a premises liability case, you want panelists who mirror that jury pool: age range, gender split, income brackets, geographic location. EmotionTrac recruits from its opt-in panel to match those specs.
Upload your surveillance footage. Most clips run 60 to 180 seconds. You can test the raw feed or a version with your voiceover narrative layered on top. Both approaches work. Testing the raw footage first tells you what the video communicates on its own. Adding your narrative shows whether your framing helps or hurts.
Step 2: Panelists watch and react
Each panelist watches your video once while their webcam captures facial micro-expressions. The system codes every frame for seven core emotions: joy, surprise, fear, disgust, anger, contempt, sadness. It also tracks engagement (are they paying attention?) and confusion (do they understand what they're seeing?).
No surveys. No post-viewing interviews. Just real-time biological responses that panelists can't fake or edit after the fact.
Step 3: Review the emotion timeline
You get a second-by-second graph showing how each emotion fluctuates across the video. Anger might spike at the 0:42 mark when the plaintiff steps into frame. Confusion might jump at 1:15 when the camera angle switches. Engagement might crater at 1:50 when nothing visibly happens for ten seconds.
These peaks and valleys tell you what your audience feels before they consciously process why they feel it. A defense attorney in a slip-and-fall case might discover that panelists feel sympathy for the plaintiff at the exact moment she grabs the handrail, not when she falls. That's the frame you need to address in your narrative.
Step 4: Identify problem moments
Look for three red flags. First, confusion spikes that last more than three seconds. If panelists don't understand what they're watching, they'll fill the gap with their own assumptions, and those assumptions rarely favor your client.
Second, engagement drops below 40%. When attention wanders, panelists miss critical details. That ten-second stretch where your client checks both directions before pulling out of the parking space? If engagement is low, they won't register it.
Third, unexpected emotion peaks. If anger spikes during a moment you thought would generate sympathy, your narrative has a problem. If contempt appears when you're establishing credibility, you need to rethink your framing.
Step 5: Test revisions
Once you've identified the trouble spots, adjust your narrative and test again. Maybe you add a voiceover at the 1:15 confusion spike explaining the camera angle. Maybe you trim the dead air at 1:50 to keep engagement high. Maybe you reframe the 0:42 moment with context about store policy.
Run the revised version through EmotionTrac with a new panel. Compare the emotion timelines. If confusion drops and engagement climbs, you've fixed the problem. If not, keep iterating.
Try it: Schedule an EmotionTrac demo and see second-by-second emotion tracking in action. Or visit Legal for more information.
What you get
The platform delivers three outputs. First, the emotion timeline graph showing second-by-second fluctuations for all seven emotions plus engagement and confusion. You can toggle individual emotions on and off to isolate specific patterns.
Second, heatmaps overlaid on the video itself. Red zones indicate high emotional intensity, blue zones indicate low engagement. Scrub through the timeline and watch the heatmap shift in real time.
Third, aggregate data across your panelist group. You see average responses, standard deviations, and demographic breakdowns. If male panelists aged 45-60 react differently than female panelists aged 25-40, you'll know.
All three outputs export to PDF or CSV for sharing with co-counsel, clients, or mediators. The data integrates into your existing case files without requiring new software.
Common pitfalls
The biggest mistake is testing too late. If you wait until the week before mediation, you won't have time to revise and retest. Start testing as soon as you have usable footage, even if your narrative isn't finalized. Early data shapes your strategy instead of validating decisions you've already committed to.
Second pitfall: testing with the wrong panel. If your case will be heard in rural Georgia but you test with urban California panelists, the data won't predict your actual audience's reactions. Demographic matching matters. EmotionTrac's panel covers all 50 states and can drill down to county-level targeting, but you have to specify those parameters upfront.
Third mistake: ignoring small emotion shifts. A 15% uptick in contempt might not look dramatic on the graph, but contempt is a high-stakes emotion in litigation. Even modest increases can poison settlement negotiations or sway a jury. Pay attention to directional trends, not just absolute peaks.
Fourth issue: over-testing. Running 12 iterations of the same 90-second clip yields diminishing returns. After three or four rounds, you're chasing noise instead of signal. Test your initial version, make targeted fixes, test again, refine once more if needed, then move forward.
Finally, don't mistake emotional data for legal strategy. EmotionTrac tells you how people feel watching your video. It doesn't tell you whether those feelings will translate into a favorable verdict or settlement. A video that generates high sympathy for your client still loses if the law isn't on your side. Use the emotion data to inform your narrative, not replace your legal analysis.