Source diversity
Source diversity is coverage across meaningfully different candidate-discovery and evidence surfaces rather than superficial use of multiple tools that largely expose the same underlying population.
Last reviewed: 2026-09-06
Why it matters in recruiting and sourcing
A sourcing stack can contain five vendors and still be one source in practice if they return the same profiles. Meaningful diversity adds new evidence or new people: owned ATS records, open technical work, research, registries, donor maps, niche communities, and licensed people data can each contribute differently.
Example
A technical search gets 70% duplicate overlap between two people databases but GitHub and ATS rediscovery each add evidence-fit candidates neither database returned. The latter lanes may have far higher marginal value despite lower raw volume.
Common failure modes
- Counting vendors instead of unique contribution
- Assuming database size equals incremental yield
- Ignoring identity resolution before measuring overlap
- Ignoring role-family differences
- Optimizing for raw discoveries instead of retained evidence-fit candidates