OBJECTIVES: With the recognition that adherence is commonly imperfect, even with once daily regimen, ‘forgiveness’ is becoming acknowledged as an important characteristic in predicting real-world effectiveness. Forgiveness is typically reported as a group average value, ignoring person-to-person variability. Here we illustrate an alternative way of estimating and presenting information on drug forgiveness which allows the estimation of the combined effect of imperfect adherence and variable forgiveness METHODS: Adherence data from 4783 patients with hypertension were used in these simulations. A projected dosing history over 365 days was obtained from each of these individual adherence profiles.  Longitudinal clinic systolic blood pressure (SBP) collected from 4879 patients after drug withdrawal were analysed using non linear mixed effect models, allowing the estimation of the between-patient variability in the loss of effect (the ‘offset’).  Projected SBP profiles were obtained by combining each dosing history with 100 offset curves randomly sampled from the variance-covariance matrix estimated in the previous step.  Re-samplings were performed in the resulting dataset using two adherence distributions described in the literature to estimate the proportion of time patients who achieved the maximal effect on SBP over a treatment period of 365 days.  RESULTS: In the first population, 90% of the patients achieved the maximal effect during 95% of the time, whereas, in the population presenting a lower adherence, about 80 % maintained the maximal effect during 95% of the time.    CONCLUSIONS: The impact imperfect adherence on response is mitigated by drug forgiveness. This methodology reveals the person-to-person variability in clinical effectiveness which is hidden when forgiveness is considered as a group-average quantity, and shows that the agent studied provides a reliable clinical effectiveness despite imperfect adherence.  Ultimately, this approach could be used to compare drugs with different levels of forgiveness and/or regimens in the presence of  non-adherence.