In February 2026, Amazon introduced a change to its API that impacts how analytics platforms, including Peel Insights, are able to track customer behavior over time. If you rely on Amazon data inside Peel, you may notice some inconsistencies in certain metrics such as repeat customer rate, lifetime value (LTV), and cohort analysis. This post explains what changed, why it matters, and what we’re doing about it.
What Changed?
Amazon recently removed access to a key identifier that analytics tools previously used to track customers across multiple orders. Historically, Amazon’s API included an anonymized email field that allowed platforms like Peel to connect orders to the same customer while still protecting personally identifiable information.
With the removal of this field, there is currently no consistent identifier available through the API that allows third‑party analytics tools to reliably determine whether two orders belong to the same customer.
This change was made without prior notice and has been confirmed by multiple developers and analytics platforms across the Amazon ecosystem.
What This Means for Your Data
Because repeat customer data relied on that anonymized identifier, certain metrics may now appear incomplete or inconsistent when analyzing Amazon sales data.
The primary metrics affected include:
- Repeat customer rate
- Customer lifetime value (LTV)
- Cohort analysis
- Retention metrics
It’s important to note that revenue, order counts, and other order-level metrics remain accurate. The change specifically impacts analytics that require identifying a customer across multiple orders.
Important Context
This limitation is not unique to Peel. Any analytics platform that processes Amazon order data through the official API is affected by the same restriction. The underlying data simply no longer contains a consistent customer identifier that can be used for repeat purchase analysis.
What We’re Doing
Our team is actively evaluating potential workarounds and longer-term approaches to improve how Amazon data is modeled within Peel. While there is no confirmed replacement for the removed identifier today, we are continuing to explore alternative signals and approaches that may help restore some of the affected metrics.
As we learn more and test potential solutions, we’ll share updates with customers and document any improvements we’re able to implement.
Questions?
If Amazon represents a significant portion of your revenue or you’re seeing unexpected shifts in your analytics, our team is happy to help review your data and discuss how these changes may affect your reporting.
Please reach out to the Peel team if you’d like help interpreting your Amazon metrics or understanding the impact on your dashboards.

.png)

