Choosing the Right KPI for your test

Updated by Kelsey Kearns

Selecting the right KPI (key performance indicator) is one of the most important decisions when designing a test. In this guide, we walk through how to choose a KPI that aligns with your business goals, media responsiveness, and measurement capabilities.

📌 What Makes a Good KPI?

  1. Responsive to Media in the Short-Term
    • For short purchase cycles: use conversions or revenue
    • For long purchase cycles: use upper-funnel signals like leads, site visits, or add-to-cart events
      (Long tests can be expensive, complex, and risk data pollution.)
  2. Measurable at the Geo Level
    • The KPI must be trackable across geographies—either using Rockerbox data or third-party sources.
  3. Tied to Business Outcomes
    • If using a non-purchase KPI (like leads), ensure you understand and trust the lead-to-conversion rate, and that it remains stable over time and across media channels.

🔄 Can I Measure More Than One Metric?

Yes. You can:

  • Choose a lead metric (e.g. site visit or sign-up)
  • Measure test results against multiple outcomes, including downstream conversions

This provides added flexibility and ensures you're capturing both immediate and longer-term media impact.

Interested in learning more? Reach out to your CSM for more detail!

How did we do?