Usage Insights
Track how your team uses feature flags across projects, environments, and SDKs. Usage Insights gives you the data you need to optimize flag adoption, identify underused or overused flags, and plan capacity.
Overview
Usage Insights provides a high-level view of flag evaluation activity across your organization. Access it from the sidebar under Insights. The dashboard aggregates data across all projects and environments to give you a complete picture of flag usage.
Key Insights
Most Evaluated Flags
Flags with the highest evaluation volume over the selected time period. These are your critical-path flags — changes to them have the largest blast radius.
- Top 10 flags by request volume
- Compare with the previous period to see trending flags
- Use to prioritize monitoring and alerting on high-traffic flags
- High-volume flags that are stale indicate urgent cleanup needs
Least Evaluated Flags
Flags with the lowest evaluation volume. Some may be for edge-case features, but many are candidates for archival — especially if they've been in this state for weeks.
- Bottom 10 flags by request volume
- Flags with zero evaluations in the last 30 days are flagged for review
- Cross-reference with Flag Health to identify cleanup candidates
- Low-usage flags in production may indicate failed feature rollouts
Evaluation Trends Over Time
Historical evaluation volume trends help you understand how flag usage grows with your application. Spot anomalies, seasonal patterns, and growth trajectories.
- Daily, weekly, and monthly aggregation views
- Compare current period vs previous period for growth metrics
- Overlay deployment events to correlate usage spikes with releases
- Export trend data for capacity planning and stakeholder reporting
Team Usage Breakdown
See which teams and projects are the heaviest users of feature flags. Understand adoption patterns across your organization.
- Break down by project to see per-team flag usage
- Compare evaluation volume across teams to identify power users
- Track new project onboarding over time
- Identify teams that may benefit from additional training or support
SDK Version Adoption
Track which SDKs and versions your applications are using. Identify outdated SDKs that should be upgraded to receive performance improvements, bug fixes, and new features.
- Distribution of SDK types: Go, Node.js, Python, Java, .NET, Ruby, React, Vue
- Version breakdown for each SDK type
- Highlight deprecated SDK versions that are no longer supported
- Track adoption of new SDK releases over time
Team Adoption Dashboard
The team adoption section helps you understand how feature flags are being used across your engineering organization:
- Active projects — Number of projects with at least one evaluation in the last 30 days.
- Flags per project — Distribution of flag counts across projects. Projects with very high flag counts may benefit from cleanup.
- Environments used — Adoption of dev, staging, and production environments across projects. Teams using only production may be missing out on pre-release testing.
- Flag types used — Distribution of boolean vs string/number/JSON flags. Most teams start with booleans and expand to structured types as they mature.
Data Retention & Privacy
Usage Insights data is aggregated and anonymized. Individual user keys are never stored in the insights pipeline — only aggregate counts are retained.
- Granular data (per-flag, per-environment) retained for 90 days
- Aggregated data (daily rollups) retained for 2 years
- SDK version data retained indefinitely for adoption tracking
- Data is stored in your FeatureSignals database — nothing leaves your infrastructure
Using Insights Effectively
Monthly flag review
Schedule a monthly review of Usage Insights alongside Flag Health. Identify flags that are no longer needed, teams that need support, and SDKs that need upgrading.
Capacity planning
Use evaluation trend data to forecast growth and plan infrastructure capacity. A 20% month-over-month growth in evaluations means you should plan for ~9x growth annually.
SDK upgrade campaigns
When you release a new SDK version with important fixes, use the SDK adoption chart to track the upgrade rollout and follow up with teams still on old versions.
Next Steps
- Evaluation Metrics— deep dive into latency, cache, and error metrics
- Flag Health— identify and remediate stale and zombie flags
- SDK Overview— explore all available SDKs and their latest versions