Signalmap is built on a commitment to methodological transparency. Every Signal Score, prediction, and hiring trend in the platform is derived from a defined, consistent data collection and analysis process. This page explains exactly how Signalmap works — from data collection through to the Signal Score algorithm — so that users can evaluate the intelligence they receive with full context.
Data Collection: Sources, Frequency, and De-duplication
Signalmap collects job posting data from a combination of exchange-operated career pages, major job boards, and professional network listings. Data is refreshed weekly for all tracked exchanges, with higher-frequency updates for exchanges that show significant Signal Score movement. De-duplication is applied at the role-level using a combination of job title normalization, location matching, and posting date comparison to prevent the same role appearing multiple times in the dataset.
| Data Source Type | Coverage | Refresh Frequency |
|---|---|---|
| Exchange career pages (direct) | All 14 tracked exchanges | Weekly |
| Major job boards | All 14 tracked exchanges | Weekly |
| Professional networks | 10 of 14 exchanges | Weekly |
| Specialist crypto job platforms | 12 of 14 exchanges | Weekly |
Department Classification and Signal Score Algorithm
Every role in the Signalmap dataset is classified into one of 12 department categories using a combination of rule-based title parsing and manual review for ambiguous cases. The department categories — Engineering, Product, Compliance/Legal, Finance, Marketing/Growth, Operations, Sales/Business Development, People/HR, Research, Security, Data/Analytics, and Other — form the basis for department mix scoring in the Signal Score algorithm.
The Signal Score algorithm weights four normalized components: hiring volume (40%), month-over-month velocity (30%), department mix score (20%), and seniority index (10%). Each component is calculated relative to the full dataset for that week, ensuring scores are comparable across exchanges of different sizes. The algorithm is updated when new historical validation data becomes available, and any methodology changes are published with full explanation.
Signalmap's track record of predictions derived from this methodology is publicly maintained. Users can evaluate past predictions against outcomes directly through the platform. This transparency is a core principle — the value of hiring intelligence depends entirely on users being able to trust the methodology behind it. Explore the full dataset at Signalmap CEX Intelligence and see predictions derived from this data at Signalmap Predictions.