At national level: CPF45 and PRODFRA results can be used for geographical comparisons, provided that the share of estimates is not too large depending on the level of detail analysed.
At the European level, the possibility to compare PRODCOM results between countries can only be determined by Eurostat.

This survey is not intended to be used to produce long series. The results are difficult to use for comparison over time because of the changes in nomenclature that have occurred since the survey was created in 2008 :
- changes in the PRODCOM list are possible every year, with many changes during the revisions of the Combined Nomenclature that occur every 5 years ;
- change from CPA rev2 to CPA rev2.1 in the 2016 survey.

At national level, the results are always disseminated in pairs : results for year N and year N-1 in nomenclature N. The comparison over 2 years, down to the finest level, is therefore guaranteed. However, no further backcasting is done.
Temporal comparisons can then easily be made by going back to 2016 for the CPF45 level, and over only 2 consecutive years for the PRODFRA level. For any other comparison, one should be very careful about nomenclature changes.

Internal coherence of data is ensured.