Amazon is extending its tracking of customer purchase habits beyond the limits of its own e-commerce site.
The e-tail giant is introducing the Amazon Shopper Panel, an opt-in, invitation-only program. Participating customers can earn monthly rewards by sharing receipts from purchases made outside of Amazon.com and by completing short surveys.
Participants who upload 10 eligible receipts per month by using the Amazon Shopper Panel app to take pictures of paper receipts or by forwarding email receipts will earn $10 toward either their Amazon balance or a charitable donation. Customers can earn additional rewards each month for every survey they complete.
Amazon is making the panel available to a limited number of its U.S. shoppers. Customers who receive an invitation to join the panel can download the Amazon Shopper Panel app for Android or iOS devices. Interested customers who did not receive an invitation can download the app to join the waitlist and will be notified via email when space becomes available.
According to Amazon, it only receives information that panel members explicitly choose to share, such as product or retailer names and prices on uploaded receipts or survey responses. Amazon says it will delete any information that may be sensitive to panelists’ privacy, such as prescription information from drugstore receipts. The e-tailer will also receive information about customer use of the app, such as usage logs and device type. Individual customer information will not be shared with third parties.
Leveraging the data it collects from panelists, Amazon says it will help brands offer better products and make ads more relevant on its site. These efforts may include measuring the relationship between ads and product purchases at an aggregate level, building models about which groups of customers are likely to be interested in certain products, obtaining feedback on new or existing products, and improving product selection on Amazon.com and affiliate stores such as Whole Foods Market.
The Amazon Shopper Panel is designed to get faster and more accurate at processing receipts over time. It relies in part on supervised machine learning, a practice where humans review a small sample of submissions to help train artificial intelligence (AI) systems to correctly process receipt contents and eligibility.