A learning KPI (Key Performance Indicator) is a metric that has been selected and defined to represent the success of a learning program or initiative against a specific business or educational objective. Unlike raw operational metrics — completion rate, watch-time, quiz score — a KPI is always tied to a goal: "reduce onboarding time by 20 %" or "increase post-training assessment pass rate to 85 % within one quarter". The discipline of choosing the right KPIs is itself a significant part of the work: organisations often default to activity metrics (courses launched, hours of learning) that are easy to pull from an LMS but measure effort rather than outcome. The Kirkpatrick model and its successors provide a framework for choosing KPIs at the right level — from learner reaction through knowledge gain, behaviour change, and business result — with each level requiring different data sources and measurement windows. In a video learning context, KPI data flows from multiple sources: completion and pass rates from the LMS or LRS, watch-time and interaction signals from xAPI (Experience API) statements, and downstream business metrics such as sales performance or error rate from operational systems that must be joined in an analytics pipeline. A common failure mode is reporting KPIs retrospectively in a quarterly slide deck rather than surfacing them in real time on a dashboard where a programme manager can act; the former produces insight, the latter enables intervention. Defining KPIs before a programme launches — and agreeing with stakeholders on what "success" means — is far more effective than reverse-engineering meaning from whatever data happens to be available after the fact.
