Revenue from Advertising is flat despite user growth. Diagnose the root causes

Problem Statement Description

Product context: Amazon is a commerce, logistics, media, devices, and cloud company; its products include Marketplace, Prime, Prime Video, Alexa devices, ads, fulfillment, and AWS.

You are a PM investigating an Amazon Advertising anomaly: advertising revenue has remained flat even though the relevant user base has grown. The revenue line may include sponsored placements, display/video inventory, or delivery-adjacent advertising surfaces where advertiser demand is expected to scale with increased traffic, engagement, or order volume.

Your task is to diagnose the root causes, not to jump directly to fixes. You should frame the anomaly clearly, define what “user growth” and “flat revenue” mean, and determine whether the issue is a real business decline, a mix shift, a marketplace dynamics problem, or a measurement/instrumentation artifact.

Consider the end-to-end workflow: users generate ad opportunities, Amazon decides eligible inventory, advertisers bid or allocate budgets, ads are served, users engage or convert, and revenue is recorded. The investigation should account for both sides of the marketplace—advertiser demand and available ad supply—while staying grounded in Amazon’s commerce, logistics, and customer-experience constraints.

The experience should consider:

- The exact revenue metric: total ad revenue, revenue per user, revenue per impression, CPM/CPC revenue, attributed sales, or segment-specific revenue.

- Whether user growth is occurring in monetizable cohorts, geographies, devices, delivery-related contexts, or lower-value traffic segments.

- Funnel decomposition across ad opportunities, eligible impressions, fill rate, auction density, bids, CTR, CVR, CPC/CPM, budget exhaustion, and advertiser retention.

- Instrumentation and data-quality checks, including event logging, attribution windows, reporting delays, bot/fraud filtering, and recent taxonomy changes.

- Segmentation by advertiser type, category, placement, campaign objective, marketplace, user cohort, traffic source, and fulfillment or delivery context.

- Potential internal changes such as ranking algorithms, ad load policies, inventory allocation, pricing rules, experimentation, budget controls, or customer-experience guardrails.

- External factors such as advertiser budget shifts, competitive pressure from retail media platforms, seasonality, macro demand, category mix, or regulatory/privacy changes.

- Evidence needed to validate or reject hypotheses, along with short-term mitigation, longer-term prevention, and monitoring plans.

The goal is to present a structured RCA plan that narrows the problem from a broad revenue plateau into testable hypotheses, identifies the highest-likelihood drivers, and defines what data, stakeholders, and decision points are needed before recommending corrective action.