Standardization of practice in classification of variants in cancer susceptibility genes has improved substantially since UK adoption of the 2015 American College of Medical Genetics (ACMG) framework. However, considerable inconsistency between laboratories persists, largely around how evidence items are weighted and combined. We present consensus recommendations from Cancer Variant Interpretation Group UK (CanVIG-UK), our national subspecialty multidisciplinary network, on which evidence types are legitimate to combine and how to deal with conflicting evidence. We also include a simple ‘exponent-scoring system’ enabling flexible consistent evidence item combination, based on prior exposition by Tavtigian et al (2018) of the ACMG variant framework as Bayesian in structure. (By Dr. Alice Garrett, https://jmg.bmj.com/content/early/2020/11/17/jmedgenet-2020-107248 )