Which time periods to use when calculating forecast accuracy for 12 month's rolling forecast?

Question

I understand that the error in forecasting is the difference between the actual value and the forecasted value for time $t$. My confusion lies in which $t$ to consider for evaluating the forecast accuracy in a 12-month rolling forecast.

In my organization, demand planners generate 12-month rolling forecasts for supply chain planning. Currently, we measure forecast accuracy as what the prediction was two months prior to the evaluation period. As an example, the forecast accuracy for March 2021 is what was the prediction in Jan 2021 for March 2021 and what was actually sold in March 2021. Is this the correct method?

Answer 1

It usually depends on the lead time. So if you on average can make the necessary changes at that 2 month lag then it is a fine window to use. If you have other constraints which means that 2 months is not enough time to do anything because it takes 5 months to get the products from the supplier then you will want to look at that longer lag for your accuracy.