Recall
The proportion of correctly identified positive cases out of all actual positive cases.
Recall measures how many of the actually relevant cases were found – critical when false negative costs are high like disease detection or fraud prevention.
Explanation
Recall, also known as sensitivity or true positive rate, is a metric used to evaluate the quality of classification models. It measures the proportion of correctly identified positive cases (True Positives) out of all actual positive cases in the data (True Positives + False Negatives). A high recall value indicates that the model has correctly identified a large proportion of actual positive cases and few have been missed. Recall is particularly relevant in use cases where missing positive cases entails high costs or risks, such as in security threat detection, identifying rare diseases, or uncovering critical errors in production lines, where false negatives must be avoided.
Marketing Relevance
For marketing and business strategies, recall is important to ensure that no relevant target groups or critical events are overlooked. When identifying high-value customers, potential churners, or critical product defects, high recall is crucial. It helps to optimally seize opportunities and proactively minimize risks by ensuring that as many relevant cases as possible are identified and appropriate measures can be initiated, thereby maximizing the effectiveness of prevention and retention strategies.
Example
A financial institution uses an AI model to detect potential money laundering transactions. High recall is absolutely necessary here. The model must identify as many suspicious transactions as possible (True Positives) to avoid overlooking any potentially illegal activities (False Negatives), even if this means some harmless transactions also need closer scrutiny.
Common Pitfalls
Very high recall can often come with lower precision, meaning many false positives are accepted. This can lead to unnecessary effort or false alarms. A model that classifies almost everything as positive has high recall but may not be practical. The optimal balance between recall and precision strongly depends on specific business requirements.
Origin & History
Recall comes from signal detection theory (1950s) and is closely related to sensitivity in medicine. Standard in information retrieval since the 1960s.
Comparisons & Differences
Recall vs. Precision
Recall asks "How many actual positives were found?"; Precision asks "How many positive predictions were correct?"
Recall vs. Recall@k
Recall@k restricts consideration to top-k results – specific to ranking and information retrieval.
Marketing Use Cases
Analytics teams use Recall to consolidate first-party data and build a single source of truth for reporting.
Data science teams apply Recall for predictive modelling, churn forecasting and attribution.
BI and reporting teams wire Recall into dashboards to give stakeholders current, defensible insights.
CRM and lifecycle teams use Recall to keep segments fresh in real time and fire marketing automation with precision.
Privacy and compliance leads anchor Recall in consent management, data minimisation and GDPR audits.
Finance and controlling teams use Recall to validate marketing investment with MMM and incrementality tests.
Frequently Asked Questions
What is Recall?
The proportion of correctly identified positive cases out of all actual positive cases. In the context of Data & Analytics, Recall describes an established approach increasingly used in production by AI-marketing teams to lift efficiency and quality in a measurable way.
Why does Recall matter for marketing teams in 2026?
For marketing and business strategies, recall is important to ensure that no relevant target groups or critical events are overlooked. When identifying high-value customers, potential churners, or critical product defects, high recall is crucial. Companies that introduce Recall in a structured way typically report 20–40% efficiency gains within the first 6 months.
How do I introduce Recall in my company?
A pragmatic rollout of Recall starts with a clearly scoped pilot use case, sharp KPIs (e.g. time, cost or conversion impact), a cross-functional team across marketing, data and IT, and a governance baseline aligned with EU AI Act and GDPR. After 6–8 weeks, scale to additional use cases.
What are the risks and pitfalls of Recall?
Common pitfalls of Recall include vague target outcomes, weak data quality, low team adoption, and bringing privacy and compliance in too late. A structured readiness check, clear ownership and a realistic roadmap materially reduce these risks.
Related Services
Go deeper: Measurement & attribution · Model comparison 2026