Global Customer Satisfaction
Senior UX Researcher at Amazon
The Problem
Amazon's Subscribe & Save business had no dedicated customer satisfaction measurement system. Teams were making decisions about acquisition, engagement, and retention without a consistent read on how subscribers actually felt about the experience such as what was working, what was driving frustration, and where the biggest opportunities sat.
I identified the gap and built the case for filling it. Without a measurement system, the business couldn't reliably tell whether it was moving in the right direction on the things customers cared most about, or whether feature investments were addressing real pain points versus assumed ones.
The Approach
I designed and launched the program as the sole architect, partnering closely with product managers across acquisition, engagement, and retention to pressure-test the framework against the decisions it would need to inform.
Methodology. I landed on a survey-based CSAT approach paired with open-text response analysis. The quantitative side gave us a benchmark we could track consistently over time; the qualitative side surfaced the why behind the scores — the language customers actually used to describe their frustrations and wins. That combination is what made the program useful for feature prioritization, not just executive reporting.
Design priorities. From the start, I built the program to be comparable across markets, even though we were starting in the US. That meant making early decisions about question structure, scaling, and sampling that would hold up under translation and localization later. It was harder upfront but saved significant rework when we expanded.
The Expansion
The program scaled in stages:
2021–2022: US only; established the baseline, refined the methodology, proved the value internally.
2023: Added Brazil.
2024–2025: Expanded to Australia, Canada, France, Mexico, and India; bringing the total to seven markets across four languages.
Two things stood out as the program went global. First, internal willingness to participate grew significantly once teams saw what the US and Brazil data had unlocked, early results created their own demand. Second, despite expectations that customer needs would diverge sharply by region, we found far more similarities across markets than differences. That insight itself reshaped how the business thought about global feature investment.
The Impact
The program now directly informs how Subscribe & Save prioritizes feature investment. Three concrete examples:
Bulk subscription management. Findings made the case for moving away from product-by-product management toward bulk edit tools, letting customers skip deliveries, change frequencies, and adjust dates across multiple subscriptions at once. This was a meaningful shift in how the team thought about the management experience.
Backup products and smarter substitutions. Customer frustration data pointed directly at out-of-stock and substitution experiences as a churn risk, which informed investment in backup product selection and improved substitution recommendations.
Cross-team influence beyond Subscribe & Save. Findings gave the team concrete feedback to share with Search, Detail Page, and Your Orders partners, driving improvements to make subscription eligibility clearer in search results, prices and savings easier for customers to comprehend, and surfacing billing confusion in Your Orders, where multi-item deliveries are listed line-by-line making it difficult for customers to reconcile against their budgets.
The program is now a standard part of how the business operates. Year-over-year tracking is built into how teams measure progress, and I present the findings annually at the Subscribe & Save Global Summit at the request of leadership and peers, making the program a regular input into both annual and longer-range strategic planning.
What I Took Away
Building this program reinforced a few things I bring to every role:
Insights only matter if they change decisions. I designed for stakeholder action from day one, not just measurement.
Cross-functional credibility compounds. Early partnership with PMs meant the program had champions before it had results.
Build for where you're going. Designing for multi-market comparability before we had multiple markets is what made the global expansion feasible.