All-source analysis

All-source analysis is the practice of fusing information gathered from several distinct intelligence disciplines, for example open-source intelligence, imagery-based geospatial intelligence, human reporting, and signals collection, into a single, coherent assessment, rather than analyzing each stream in isolation. The goal is to corroborate weak signals from one source with stronger evidence from another and to resolve contradictions between sources before a judgment is issued.

This differs from single-source analysis, which stays within one discipline's data and is faster to produce but more vulnerable to that source's specific blind spots or manipulation. An all-source analyst instead builds a narrative that has to be consistent across independently collected evidence, weighting each source by its known reliability and recency.

All-source analysis underlies most finished intelligence products used for situational awareness and decision support, and is the standard approach behind pattern-of-life analysis assessments that draw on both imagery and behavioral data. The common pitfall is confirmation bias: once an analyst forms an early hypothesis, there is a temptation to weight corroborating sources more heavily and explain away contradicting ones, rather than letting anomaly detection or contrary evidence genuinely update the assessment.