Metrics for Cross-Sell Success

To measure the success of a cross-sell feature where Product A leads to Product B, I would look beyond total revenue and focus on a balanced scorecard. My primary North Star metric would be the Cross-Sell Conversion Rate normalized by session depth, ensuring we aren't just capturing low-hanging fruit but driving genuine interest. However, revenue is a lagging indicator; the leading indicators are crucial for health. First, I would track the Net Promoter Score (NPS) delta specifically for users exposed to the cross-sell versus those who were not. If the NPS drops, we are likely annoying customers, which violates the customer obsession principle. Second, I would monitor the 30-day churn rate for the original buyer. If buying Product A increases their likelihood of leaving the platform shortly after seeing Product B, the feature is toxic. Third, I'd analyze the return rate of Product B compared to organic sales. High returns indicate the recommendation logic is misaligned with user needs. Finally, I would measure the incremental lift in Average Order Value (AOV) while controlling for discount usage. At Amazon, we must ensure that every cross-sell adds real value to the customer journey, not just a line item on a P&L statement. By tracking these combined metrics, we can iterate on the algorithm to maximize long-term CLV rather than short-term transaction volume.