Skip to main content

COMPEL Glossary / data-drift

Data Drift

Data drift occurs when the statistical properties of the input data a deployed model receives change compared to the data it was trained on.

What this means in practice

Customer behavior shifts, market conditions evolve, seasonal patterns change, and regulatory requirements update -- all causing the data environment to diverge from what the model learned. Data drift is a primary cause of model performance degradation in production. Organizations need automated monitoring systems that track input data distributions and alert when significant shifts occur. In the COMPEL maturity model, data drift detection capability is a key differentiator between Level 2 (manual monitoring) and Level 3 (automated detection with defined thresholds and response procedures) in the MLOps domain.

Why it matters

Data drift is the silent killer of production AI systems. As customer behaviors shift, market conditions evolve, and business contexts change, models trained on historical data become progressively less accurate. Organizations that fail to detect drift discover performance problems only after business outcomes have already deteriorated, resulting in lost revenue, poor decisions, and eroded trust in AI capabilities.

How COMPEL uses it

In the COMPEL maturity model, data drift detection capability is a key differentiator between Level 2 (manual, ad-hoc monitoring) and Level 3 (automated detection with defined thresholds and response procedures) in the MLOps domain (Domain 7). Drift monitoring is designed during Model, implemented during Produce, and its effectiveness is measured during Evaluate as part of operational health metrics.

Related Terms

Other glossary terms mentioned in this entry's definition and context.