How it works
ROI can be expressed two ways:
- As a multiple: incremental revenue ÷ total cost. A result of 2.5x means the assistant brings in 2.5 times what it costs.
- As a percentage: (incremental revenue − total cost) ÷ total cost × 100. The same case gives a 150% ROI.
Illustrative example: an assistant that brings in $200,000 of incremental revenue over a year for a total cost of $80,000 has a 2.5x ROI.
Three things to settle before you calculate:
- Total cost includes the subscription, setup and the time your teams spend on the catalog and content.
- Calculating on margin rather than revenue is stricter, and closer to what a CFO looks at.
- The assistant only runs on sessions that accepted cookies. Your estimate should start from that traffic, not total traffic.
Why it matters in e-commerce
An AI Shopping Assistant is an investment you have to defend to finance, at launch and at every renewal. A cleanly measured ROI lets you compare the assistant with other levers, such as paid acquisition or a page redesign, and decide where the next budget goes.
Further reading: Exploring AI Shopping Assistant Capabilities: Transforming the E‑commerce Experience
Real-life example: Samsonite
Samsonite measured its AI Shopping Assistant with an A/B test. The result: a 2.2x incremental ROI. The incremental revenue the assistant brought in was 2.2 times its total cost.
Common confusion
ROI is often calculated, wrongly, on all the revenue from shoppers who talked to the assistant. Some of those shoppers would have bought anyway. That number is AI-influenced revenue, useful for tracking how the assistant is used. Only incremental revenue measures what it adds.