This is maybe one of the most common questions we receive.
Sadly, there is no one-size-fits-all answer here, but there are some things you can do, and some pitfalls to avoid:
- Do not use "industry data"
- Do not use "gut feel"
- Do not use "this is what we did last year"
Benchmarks
Instead, compare to a benchmark forecast.
As the above topic specifies, the only real thing you can do to see if your forecasts are doing well is to see if they are beating a simple benchmark. If your manual or auto forecasts are beating a baseline forecast in accuracy by 10% or more (compare the Benchmark wMAPE with the Stat and Op Fcst wMAPE on the Forecast Error screen, or read the FVA columns), then you can be confident that you are doing a decent job on forecast error for your products.