Crisp will automatically flag any data points that it considers to be outliers. Explaining historical outliers is a useful data cleansing activity that will improve your unique forecasting algorithms.
For products with many historical outliers, it can be difficult to know where to start when trying to clean up your data. To help with this, Crisp provides you with an impact score for each outlier. You can use this score to gauge the effect that each outlier is having on your forecasting algorithm.
Impact scores represent how far away the specific data point is from your average sales history. Outliers with scores closer to 1 are more impactful than scores closer to 0. Therefore, you can use the impact score to prioritize which outliers to focus on explaining first.