Understanding the Benchmark Dataset
How GigAnalytics collects, anonymizes, and uses aggregate earnings data to power rate comparisons — and how your privacy is protected.
How benchmark data is collected
✅ Strictly opt-in
Benchmark data is only collected from Pro users who explicitly enable "Contribute to benchmarks" in Settings → Privacy. Free users never contribute data. Disabling the toggle at any time removes your contribution from future aggregations within 24 hours.
When you opt in, GigAnalytics periodically sends aggregated (not raw) metrics from your income streams to our benchmark pipeline. We never send individual transactions, client names, or payment details.
What gets sent when you opt in
Anonymization methodology
Before your data enters the benchmark pool, it goes through a four-step anonymization pipeline:
- 1
K-anonymity bucketing
No bucket in the benchmark dataset contains fewer than 25 users. If your specific combination of platform + rate + region has fewer than 25 contributors, your data is suppressed from that segment until the pool grows.
- 2
Differential privacy noise
We add calibrated statistical noise to aggregate values (Laplace mechanism, ε = 0.5) so that no individual's presence or absence can be inferred from the published percentiles.
- 3
Delayed publishing
Contributions from the last 72 hours are held before entering the aggregate. This prevents near-real-time inference attacks.
- 4
User ID removal
Contributions are decoupled from user IDs before entering the benchmark store. The pipeline is append-only and logs are rotated after 30 days.
How benchmarks surface in GigAnalytics
Benchmark data is surfaced in three places in the Pro dashboard:
ROI dashboard → Rate comparison widget
Shows p25/median/p75 hourly rates for your platform category and region. Your rate is shown as a dot on the distribution.
Pricing experiment recommendations
When you run an A/B pricing test, the AI insight layer uses benchmark data to suggest whether your tested price is above or below market median.
Onboarding income goal wizard
The "How much should I charge?" guide uses benchmark hourly rates to give you a realistic starting point for your target income calculation.
Opting out
Go to Settings → Privacy → Benchmark contribution and toggle it off. Your data stops contributing within 24 hours. You can still view benchmark data in the dashboard — contributions and viewing are independent controls.