Why this matters
You test concepts, messaging, and creative before launch. But most pre-testing relies on self-reported ratings or post-view surveys. People say they liked something. They check a box. You move forward.
The problem: self-reports miss the moments that actually drive behavior. A viewer might rate an ad 7/10 overall but feel confused at second 12, annoyed at second 34, and genuinely surprised at second 58. Those micro-reactions predict whether they'll remember your message, share it, or tune out.
EmotionTrac captures facial expressions frame-by-frame using FACS (Facial Action Coding System) on opt-in panelists. You get a timeline of joy, surprise, confusion, disgust, anger, sadness, and neutral states for every second of your video. That data becomes your category benchmark.
The workflow
Step 1: Define your content set
Pick 5 to 15 videos in your category. Competitor ads, your past campaigns, category leaders, challenger brands. If you're testing a new product video, include the top 3 competitor product demos.
Keep videos under 3 minutes each. Longer content works, but attention drops and you'll see more noise in the data. Upload all files to the EmotionTrac platform in one batch.
Step 2: Set your panel criteria
Choose demographics that match your target audience. Age, gender, region, income, category usage. EmotionTrac recruits from opt-in pools and screens for webcam quality and lighting.
Aim for 100 to 300 respondents per video. Smaller samples (50) work for early-stage tests. Larger samples (500+) tighten confidence intervals if you're testing high-stakes creative before a national buy.
Step 3: Run the session
Panelists watch each video on their own device. The platform records their face via webcam while they view. No prompts, no interruptions. Just watch and react naturally.
FACS coding happens in real time. The system tracks Action Units (specific muscle movements) and maps them to emotions. You don't need to train anyone or score anything manually.
Try it: Schedule an EmotionTrac demo and see second-by-second emotion tracking in action. Or visit Insights for more information.
Step 4: Review the emotion timelines
You'll see a line graph for each emotion across the video's runtime. Joy spikes when your brand reveal hits. Confusion jumps when the voiceover contradicts the visual. Disgust ticks up during an awkward testimonial moment.
Compare your new creative to the benchmark set. Does your video hold attention longer? Does it generate more joy in the first 10 seconds? Does it avoid the confusion spike that tanked your competitor's last launch?
Filter by demographic segment. Your 25-34 audience might love the humor at 0:42, while 55+ viewers show neutral or negative reactions at the same timestamp.
Step 5: Build your category norms
Aggregate the benchmark videos into a single reference dataset. Calculate average emotion scores by second, by segment, by creative type (product demo vs. testimonial vs. explainer).
Now you have context. A joy score of 0.6 at the 15-second mark might sound abstract. But if category norms sit at 0.4, you know you're outperforming. If norms are 0.8, you're underperforming.
Update your benchmarks quarterly. Emotional norms shift as audiences see more content, as platform algorithms change, as cultural moments alter what feels fresh or tired.
What you get
A heatmap of emotional highs and lows for every second of your video. You can pinpoint exactly where viewers disengage, where they feel confused, where they react with joy or surprise.
Comparative data across your category. You'll know if your creative is on par, ahead, or behind the emotional benchmarks that predict recall and action.
Segment-level breakdowns. Test whether your message lands differently with men vs. women, younger vs. older viewers, urban vs. rural audiences. Adjust creative or targeting based on which segments respond.
Directional guidance before you spend media dollars. If your video underperforms benchmarks in the first 10 seconds, re-edit the open. If it generates confusion at a key product claim, rewrite the script. Test again until the emotion curve matches or beats category norms.
Common pitfalls
Testing too few videos. A single competitor ad isn't a benchmark. You need a range of content to understand what "normal" looks like in your category.
Ignoring the first 5 seconds. Emotion curves in the opening moments predict whether viewers will keep watching. If your benchmark set shows joy or surprise early and your video shows neutral or confusion, you'll lose attention before your message lands.
Treating all emotions equally. Joy and surprise tend to correlate with positive outcomes (recall, sharing, purchase intent). Confusion and disgust correlate with drop-off. Sadness can work in certain contexts (charity appeals, healthcare messaging) but often signals disengagement in product marketing.
Skipping demographic filters. Aggregate data hides important differences. A video that performs well overall might bomb with your core target segment. Always check segment-level emotion curves before finalizing creative.
Assuming one test is enough. Emotional benchmarks change. What worked 6 months ago might feel stale now. Run follow-up tests on new creative and refresh your norms regularly.
Over-indexing on a single spike. One moment of high joy doesn't save a video if the rest of the timeline shows confusion or neutral affect. Look at the full curve, not just the peaks.
Forgetting to test your own past work. Your archive is part of the category. Include your previous campaigns in the benchmark set so you can see whether your new creative represents an improvement or a step backward.