Qlik Discovery Agent^™

Overview

Discovery Agent is Qlik's AI-powered anomaly detection and monitoring capability built into Qlik Cloud. It continuously scans your Qlik Cloud applications for meaningful changes in your data and delivers prioritized, AI-generated insights directly to users — without requiring manual monitoring, predefined rules, or static thresholds.

Discovery Agent is part of Qlik's broader Agentic AI framework, where assistants answer questions and agents monitor and act. While Qlik Answers responds to questions users ask, Discovery Agent proactively surfaces what users need to know, even when they haven't thought to ask.

Value and Business Impact

Discovery Agent enables organizations to move from reactive reporting to proactive, data-driven action. It enables:

Early detection — Emerging issues are caught before they escalate into costly problems. Teams are informed at the moment something meaningful changes, not after the fact.

Reduced manual work — Repetitive monitoring tasks are eliminated, freeing analyst time for higher-value work. Discovery Agent handles continuous surveillance across all monitored applications automatically.

Business clarity — Rather than presenting every data point equally, Discovery Agent surfaces the changes that truly matter, ranked by significance and presented in plain language.

Faster decision-making — Teams across the organization can act on insights immediately, regardless of their familiarity with the underlying Qlik applications or data models.

Operational responsiveness — Automated, proactive alerting means organizations can react faster to changing conditions, protecting performance and uncovering new opportunities sooner.

What Discovery Agent Does

Discovery Agent continuously monitors your Qlik Cloud applications in the background. It automatically detects meaningful changes in your data, evaluates their significance, and delivers clear, prioritized insights to users in plain language — with no configuration of thresholds or alert conditions required. These insights are provided to users though their own personal Feed in the analytics hub.

How Discovery Agent Prioritizes What Matters

A core challenge in any monitoring system is distinguishing signal from noise. Discovery Agent addresses this through three interconnected mechanisms:

Dynamic baselines - Rather than relying on static thresholds, Discovery Agent uses forecasting to establish an expected range for each metric. Deviations from this forecast trigger evaluation, and there are no rules to manually maintain as the business changes.

Intelligent thresholding - Discovery Agent evaluates trends, spikes, and shifts against the dynamic baseline and filters out changes that fall within normal variation. Only anomalies that are genuinely significant are surfaced.

Adaptive sensitivity - Discovery Agent automatically adjusts its sensitivity based on the volume of insights it is generating. If too many insights are being delivered, sensitivity is reduced to focus on the most significant changes. If too few are being generated, sensitivity increases to ensure meaningful signals are not missed. This continuous self-tuning keeps the insight feed balanced and actionable without manual intervention.

Where Discovery Agent Fits in Qlik Cloud

Discovery Agent is part of Qlik's Agentic AI framework within Qlik Cloud Analytics. Within this framework, assistants — such as Qlik Answers — respond to questions that users ask. Agents, by contrast, monitor conditions and act autonomously on behalf of users.

Discovery Agent occupies the monitoring role within this framework: it watches your data continuously, determines what is significant, and brings that information to users proactively. This complements Qlik Answers, which remains the appropriate tool when users want to ask specific questions of their data. Together, these capabilities represent a shift from a model where analytics requires active user engagement to one where the platform brings relevant insight to users as conditions change.

Security architecture

Access Control

Discovery Agent fully respects application level access controls. The applications a user with see insights from are restricted to those applications they have permissions to see though the hub. Support for row-level section access is planned for a future release.

Cross-Region Inference

Discovery Agent requires cross-region inference to function. This means that when Discovery Agent processes AI requests, data may be temporarily sent to a different AWS region for inference processing. This is consistent with how other AI services in Qlik Cloud operate and is subject to the same security controls described in the Qlik Answers security architecture documentation. All processing occurs within the AWS private network, no data is persisted in the inference region, and customer data is never used to train AI models. Customers must opt in to cross-region data processing to enable Discovery Agent. This setting is disabled by default for Qlik Cloud tenants.

Discovery agent follows the same process as is used for access to structured data with Qlik Answers. This process is described in detail in the section Qlik Answers - Security Architecture and Controls . For the latest information on inference processing regions by product feature and home region, see Enabling cross-region data processing on Qlik Help.