Forecast Scores & Summaries

On the Instances tab of the Forecast page, you can see an assessment of the quality of each forecast instance.

Confidence Score

The Confidence Score is generated on creating a forecast instance (currently limited to Auto Forecasts, but soon to include forecasts created using the wizard).  
It is an assessment of how closely correlated the selected history data is.

If your source data is tightly grouped, you can have high confidence in the forecast.
If the source data is widely spread, the averages may still give you a sensible forecast, but you should be more prepared for all eventualities as it is less predictable.  You may also want to check the weeks that have been used as source data more closely to make sure that they are suitable for forecasting.

Accuracy Score

The Accuracy Score is generated on Monday morning for all forecast instances in the previous week.
It is an assessment of the differences between the forecast & actual data.

AI Summary

For each available score, you can see a further description of confidence & accuracy, with suggestions on how you might improve further forecasts.

These are based an a range of statical measures, made at the same times that we created the overall scores.

You can access the summaries using the stars button or from the data menu