Conversion in Payments fell after a redesign. How would you investigate

Problem Statement Description

Product context: Stripe is financial infrastructure for internet businesses; its products include payments, Checkout, Billing, Connect, Radar, Issuing, Terminal, and tax tools.

Stripe has recently redesigned part of its Payments experience, and after launch the team observes a drop in payment conversion. The impacted experience may involve merchants using Stripe-hosted or embedded payment flows, developer-integrated payment forms, or checkout experiences where end customers complete a transaction. Your task is to investigate the decline as a product manager leading a root-cause analysis.

Assume this is a production issue with business impact: merchants care about completed payments, end customers expect a fast and trustworthy checkout, and Stripe must preserve reliability, compliance, fraud controls, and developer confidence. The redesign may have changed UI, payment method presentation, authentication flows, error handling, latency, or integration behavior, but you should not assume the cause upfront.

Frame how you would diagnose the anomaly, validate whether the conversion drop is real, identify where in the payment funnel it is happening, and determine what evidence would support or disprove likely hypotheses. Consider both user-facing and system-facing explanations, including instrumentation changes, traffic mix, merchant segments, device/browser differences, regional payment methods, fraud or authentication behavior, and rollout timing.

The experience should consider:

- How to define “conversion” precisely, including numerator, denominator, funnel step, time window, and eligible payment attempts

- Whether the observed decline is statistically and operationally meaningful versus noise, seasonality, traffic mix, or measurement changes

- Segmentation by merchant type, geography, payment method, device, browser, integration type, currency, new versus returning customers, and rollout cohort

- Checks for instrumentation, logging, event taxonomy, attribution, and dashboard changes introduced during or after the redesign

- Funnel-stage diagnosis across page load, payment method selection, form completion, confirmation, authentication, authorization, failure states, and retry behavior

- Hypotheses spanning UX friction, latency, errors, declined payments, 3DS/SCA behavior, fraud rules, API changes, and developer integration compatibility

- Immediate mitigation options, escalation paths, communication to affected internal teams or merchants, and criteria for rollback or partial rollback

- Longer-term prevention through monitoring, experiment design, guardrails, alerting, release gates, and post-launch review

Your goal is to describe a structured investigation plan that helps Stripe quickly determine whether the redesign caused the conversion decline, where the issue is concentrated, what evidence is needed to isolate the root cause, and how the team should protect merchant revenue and payment reliability while the investigation is underway.