A key performance indicator tells you how a study performed. A key risk indicator tells you where it is likely to go wrong. Sponsors need both, but most governance dashboards are full of KPIs and nearly empty of KRIs. That is why problems so often arrive at the governance meeting fully formed.
What is the difference between a KPI and a KRI?
A KPI measures output against a target: sites activated against plan, participants enrolled, queries resolved within an agreed window. A KRI measures a condition that predicts future trouble: a rising screen failure rate at a site, a growing lag between visit and data entry, an accumulating list of open action items, turnover on the CRO team.
KPIs are lagging by nature. KRIs are leading. ICH E6(R3), adopted in January 2025, emphasizes proactive, risk-proportionate oversight, and that pushes sponsors toward leading indicators.
Why do dashboards drift toward KPIs?
KPIs are easy to count, often defined in the contract and make the CRO's performance legible. KRIs require judgment about what predicts trouble for this particular study. They often draw on data from several systems. And they can feel uncomfortable, because a KRI that flags risk before a miss can read like an accusation.
How do you choose KRIs that matter?
Start from your study's critical risks. For each one, ask: what would we see first if this were starting to go wrong?
Enrollment: screen failure trend, prescreening-to-consent conversion, time from site activation to first participant.
Data quality: data entry lag, query rate per page, protocol deviations per participant.
Site performance: monitoring visits against plan, age of outstanding site action items.
Vendor stability: turnover on the account, missed handoffs between functions.
Safety reporting: time from site awareness to serious adverse event report.
Keep the list short. Eight to twelve indicators is plenty for most studies. Give each one a threshold, an owner and a pre-agreed response.
How should KRIs be used in governance?
A KRI is only useful if it produces a conversation. Put KRIs at the top of the governance agenda, not the end. Review trends rather than snapshots. When an indicator crosses its threshold, the default output is a decision: investigate, mitigate, escalate or accept with a documented rationale. That record also becomes strong evidence of sponsor oversight at inspection.
What are the most common KRI mistakes?
Too many indicators. Thresholds so loose they never trigger. Indicators nobody owns. And measuring only the CRO while ignoring sponsor-side indicators, such as the time it takes the sponsor to approve documents or make decisions. The sponsor is part of the system, and sometimes the bottleneck.
Where should a sponsor start?
Pick three critical risks, define one KRI for each and review them monthly for a quarter. Refine what proves useful and drop what doesn't. Mature KRI practice grows from use, not from a long list written at kickoff and never revisited.
Have more questions about oversight metrics? Browse ABRA's frequently asked questions.
