Amazon Quest
A gamified campaign that encouraged customers to complete tasks to boost user engagement and purchases.
UX Lead — Sebastien Chery
PM — Wooda Goshen-Seo
Background
Our Japanese customers did not frequently purchase digital content. Despite engaging with our Kindle platform, they were often passive, browsing content without converting to active buyers. Research indicated a lack of structured incentives and engaging UI contributed to low purchasing behavior. In response, my team created Amazon Quest, a campaign that sent customers on a purchase quest. This campaign encouraged customers to buy certain books and receive a sum of Amazon points after every purchase. The points could then be used to purchase additional products to counteract low purchasing behavior.
Problem
I designed the campaign's homepage to integrate a progress bar and rewards system. This allowed customers to track their journey from visiting a title's detail page to purchasing and receiving Amazon Points. The homepage further prioritized clear and accessible goals, providing step-by-step guidance to help users earn rewards effortlessly. The countdown timer added urgency, emphasizing the time-sensitive nature of the rewards.
The PM and I agreed that the usual entry points for campaign homepages wouldn’t effectively reach our target cohorts. Early customer research revealed that low purchasing behavior stemmed from weak incentives and an unengaging UI. To evaluate the new campaign's efficacy, we aimed to reach users who already showed this low purchasing behavior as a target cohort. This required buy-in from teams outside our org to place entry points in other areas of Amazon where the campaign would be more visible to these users.
Situation
We proposed an ingress on Amazon's global TYP (Thank You Page). This touchpoint had strict technical constraints as it was managed by a modular framework. The challenge was integrating the ingress on a module without disrupting the TYP's performance or conflicting with existing promotions.
Task
As the UX Design Lead, I was tasked with:
1. Designing an ingress card that adhered to the global TYP's modular framework while achieving campaign visibility and engagement goals.
2. Resolving the global team's performance concerns, including load-time optimization and global content prioritization, and securing approval for a card slot on the global TYP.
3. Demonstrating to the Japanese-based team how the card's design aligned with the campaign’s creative and functional mechanics.
Action
Collaboration with Engineers:
I scheduled meetings with engineers to understand the TYP's asynchronous loading behavior and modular design.
I scheduled meetings with engineers to understand the TYP's asynchronous loading behavior and modular design.
Iterative Design Process:
After gathering generative information, two design options were proposed for the ingress. A stylistic approach and a minimalistic design focused on performance and content prioritization. Stakeholder feedback pushed priority toward the minimalistic design as P0, leaving the stylistic as P1.
After gathering generative information, two design options were proposed for the ingress. A stylistic approach and a minimalistic design focused on performance and content prioritization. Stakeholder feedback pushed priority toward the minimalistic design as P0, leaving the stylistic as P1.
Adjustments, such as reducing card height, ensured global promotions remained visible above the fold to balance the campaign and overall business goals. For the sake of brevity, these were the final options that emerged from that collaborative process with the engineers:
Results
1. Successfully integrated the ingress card without impacting global "Thank You" page performance.
2. Gained the global team's full approval during negotiations, securing a dedicated card slot for future Japanese campaigns that would streamline future collaborations.
3. Amazon Japan's Quest delivered impressive statistical results, showcasing the power of gamification to drive digital purchases with our chosen cohorts, where 29,000 customers enrolled in the campaign in March and April, highlighting strong participation. The monthly average of paid units per customer within this group increased significantly. Cohort 1, where customers had demonstrated low purchasing behavior, saw a remarkable +2279% jump, reaching 6.9 units per customer, while Cohort 2, serving as control, experienced a +662% jump, reaching 8.39 units per customer.
These results demonstrate that the quest mechanics effectively turned cohorts from browsing to buying, driving customer activity and sustained engagement.
Pitfalls
In digression, to keep customers motivated, we added "Add to Wishlist" as a free, intermediate step to break up the journey to purchasing into smaller, manageable goals. However, data showed that 43% of customers in Cohort 1 and 33% in Cohort 2 who completed a purchase skipped the Wishlist step. This indicated that the "Add to Wishlist" feature may not have been as intuitive or noticeable as intended.
As a next move, removing this wishlist step or refining it could improve progression through the quest and better support conversion.
