Usage research reports self-published by vendors are generally classified in academia as grey literature, and suffer from issues such as selection bias, survivorship bias, metric selection bias, and absence of causal inference.