It is common practice for animal shelters and rescue organisations to conduct behaviour assessments on the dogs in their care. The information obtained is used to identify dogs suitable for re-homing and also assist with matching dogs to appropriate homes. Although the aim of these assessments is to provide a snapshot of the behavioural characteristics of individual dogs, research to determine whether such tools are effective and accurate in predicting the behaviour of dogs post adoption is currently lacking. In the present study, we investigated the predictive validity of a standardised protocol called the Behavioural Assessment for Re-homing K9's (B.A.R.K.). We used the B.A.R.K. protocol to assess 74 dogs housed in an animal shelter, prior to their adoption. All dogs were at least 1 year of age (mean = 2.86 years, SD = 2.02). New owners of these dogs took part in a post adoption survey 2–8 months (mean = 4) after they adopted their dog. The survey aimed to establish the degree to which the behavioural tendencies identified by the B.A.R.K. protocol carried through to the adoptive home. The predictive validity of the B.A.R.K. protocol was relatively poor. Multiple regression analyses revealed that fear, measured by the B.A.R.K. protocol, significantly predicted ‘fearful/inappropriate toileting’ behaviours post adoption (β = 0.36, P < 0.05), as did anxiety (β = −0.31, P < 0.05). However problem behaviours more generally and aggression post adoption were not predicted by the B.A.R.K. protocol. Almost 25% of adopters reported that their new dog had ‘growled, snapped at, or attempted to bite a person’ and nearly 75% indicated that their dog exhibited behaviour they would change if they could. Despite this, just over half (56.8%) of the new owners said they were very satisfied with the behaviour of their newly adopted dog and 71.2% said their adopted dog had met their expectations. Our results suggest that additional research is urgently needed to evaluate the predictive validity of in-field behaviour assessments and whether a more holistic, or alternative, approach to assessing shelter dog behaviour, such as longer-term foster care programmes, is required to safeguard the welfare of dogs in the shelter system and the community at large.