Client
Region
Tech stack
Two in 3 eaters don't know what they're having for dinner at 5 pm. Worse, people spend 20 minutes per day scrolling what to eat. Since emotions drive food choices, what if Coke turned this decision fatigue with a playful value-add?
Order by mood
Through facial recognition technology, we built a web-based AR filter that analyzed users' moods to recommend a meal to match. Pairing with Uber Eats and Wolt, the result was orderable through app. Not only was the experience seamless between Coke's and Uber's infrastructure, 8th Wall's facial landmark technology ensured users' privacy by storing no face data. The app was first launched in North America then localized to EU markets through partnerships regional food service apps like Wolt. The result: a fun, shareable activation that made Coke part of the decision, not an afterthought.
How it worked
Facial expression recognition
Computer vision detected facial landmarks to identify mood in real-time. The experience never stored face data; emotion detection happened in-session only.
Personalized recommendations
Once mood was detected, the app suggested a meal based on their mood result.
Localized partnerships
Users could redeem a $5 coupon for any Coke-carrying restaurant that matched the result food type. Our EU launches in Greece and Romania partnered wtih Wolt and Bolt.
Project learnings
Whenever considering a web AR application, you run the risk of user drop-off from a paid media or organic post into the experience itself. Off-platform solutions from native Snapchat, TikTok or Instagram filters may risk users drop offing. Thankfully, 8th Wall, a third-party AR experience platform, provides an intuitive experience for users who were routed through a robust paid media plan: Meta, Snap, and Uber's own ecosystem: In-app banners, Post-checkout, Journey ads, email push. Conversions of redeemed coupons climbed as we moved down the funnel. We beat bechmarks by 3x for our media placements!