As explained in the article Add or update sales data to SKU Science, you can obtain new sales forecast at each cycle if you regularly upload your sales data.
This will allow you to track Key Performance Indicators (KPIs) in order to improve your forecasts cycle after cycle, which should be your end goal.
Note that this feature is only accessible if you have a "Core", "Pro" or "Enterprise" subscription plan.
You can display KPIs at the lowest level or at the aggregate level.
If you access SKU details, from the navigator, for example, you can display KPIs by clicking on the Key Performance Indicators tab.
To display KPIs at different aggregation levels, click KPIs in the left menu. You can display KPI details by clicking on the details icon in the table.
In both cases, you can switch between charts and table by clicking the buttons:
Choosing the right lag
A dropdown menu allows selecting the lag for which you want to track KPIs.
Lag is the number of periods between forecast generation and forecast target period.
The lag you want to select is usually equal to your lead time. The lead time being the delay after which the article is available, your forecast needs to be as accurate as possible in regards to this particular delay.
For example, if you need 3 months between the manufacturing of an item and the time at which the item is available for the customers, you need to anticipate the demand 3 months before the actual sales. Therefore you should care to create an accurate forecast in June for September. Said differently, you need to monitor your forecast accuracy at lag 3 and try to improve it. Other lags are less interesting for you since there is a low value to improve your forecasts at Lag 1 or Lag 2.
Understanding KPIs and forecast value-added
Here is an example of a KPI table.
There's a first column showing average KPI values. Each column then shows KPI for a given cycle, according to the selected lag. For example, if you're displaying lag 3, the April column KPIs are computed by comparing the forecast you generated in January (for April) to the actual sales in April.
There are 4 different KPIs:
Bias: bias is the difference between forecast and demand (f-d), in units.
Bias (%): ratio between bias and demand, in percent (100x(f-d)/d).
Error : forecast error is the ratio between bias and demand, in absolute value, in percent. Note that the average value is the MAE (Mean Absolute Error).
Accuracy: accuracy is (100 - error), in percent.
For each KPI, there are 3 lines:
SKU Science: KPIs for SKU Science generated forecasts.
User: KPIs after user adjustments.
Value Added: improvement (or downgrade) from user adjustments. A negative value-added is an improvement for bias and errors. A positive value-added is an improvement for accuracy.
Cell colors are showing how KPI values compare with user-defined thresholds:
Red: error or bias is higher than the high threshold.
Orange: error or bias is between high and low threshold.
Green: error or bias is lower than the low threshold.
You can change those thresholds in Settings / Rules / KPI Thresholds.
The charts can show Error, Bias, or Accuracy according to the dropdown option you choose.