Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
North Gate Boulevard du Roi Albert II 16, 1000 Bruxelles.
1.6. Contact email address
Confidential because of GDPR
1.7. Contact phone number
Confidential because of GDPR
1.8. Contact fax number
Confidential because of GDPR
2.1. Metadata last certified
27 August 2025
2.2. Metadata last posted
29 August 2025
2.3. Metadata last update
29 August 2025
3.1. Data description
The PRODCOM inquiry is conducted by Statistics Belgium. This is a yearly survey of the value and volume of products produced by industrial enterprises in Belgium and sold during the reference year. It is carried out in compliance with EU Council Regulation on the Establishment of a Community Survey of Industrial Production (N°3924/91).
The PRODCOM survey provides information on PRODuction of the European COMmunity and is compiled from data received from Belgian manufacturers.
The survey sample is approximately 3400 legal units and 3500 local units, covering 222 industries, derived from sub-sector, and 2067 products.
Available data include:
By product:
value and volume of product sales for own account
value and volume of products under sub-contracted operations (the value represents fee received for the sub-contracting operation)
actual production
3.2. Classification system
In the PRODCOM classification products are grouped by the NACE Rev.2, 4-digits industrial sectors, in which they are primarily produced (i.e. characteristics products).
Nace Rev.2 is the EU system for classifying economic activities.
The value and volume data shown for a product, therefore, includes not only the production by enterprises classified to the NACE sector where it is primarily produced but also its subsidiary production by enterprises classified to the other Nace Rev.2 sectors.
Classification of Products
The PRODCOM list is divided into "Classes" corresponding to the classes of NACE Rev.2. Within the individual classes, the headings of the list are set out in ascending order of their PRODCOM codes. PRODCOM uses an eight-digit product code of the form XX.XX.XX.YY.
The first four digits of the code correspond to the 4-digit classes of NACE Rev.2. The first six digits are the Community Classification of Products by Activity (CPA) codes. The CPA provides a detailed listing of the characteristic products for each 4-digit NACE Rev.2 economic activity. The last two digits provide a more detailed breakdown of the CPA classes into PRODCOM products headings.
There is a direct link between the PRODCOM classification and the EU foreign trade Combined Nomenclature (CN). Each PRODCOM 8-digit heading covers one or more CN 8-digit headings.
To view the national classification, please see the following link.
To view the different classification, please see the following link and search the nomenclature in the search module at the top left.
3.3. Coverage - sector
The survey covers local units of industrial enterprises, which are wholly or primarily engaged in industrial production and industrial services in the Mining, Quarrying, and Manufacturing industries, i) with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year; ii) with an industrial secondary activity with 20 persons employed or with a turnover of EUR 5.000.000 n the previous reference year.
Under the Belgian Prodcom survey, the NACE classification are covered as follows:
from NACE REV.2 : sections B and C, comprising sections 07-33, but excepting 09, 19, 1011 and slaughterhouse part of 1051
from NACE REV.1.1 : sections C, D and E, but excepting 1551 and slaughterhouse part of 1511.
Belgian Prodcom survey is a cut-off sample survey.
3.4. Statistical concepts and definitions
The purpose of the statistics is to report, for each product in the Prodcom List, how much has been produced in the reporting country during the reference period.
This means that Prodcom statistics relate to products (not to activities) and are therefore not strictly comparable with activity-based statistics such as Structural Business Statistics. The NACE codes on which Prodcom codes are based merely serve to identify the enterprises that should be surveyed in order to determine the amount of production of the product.
The products: the Prodcom List defines the products to be included in the survey. There are approximatly 4000 headings representing manufactured products and some industrial services.
The value and the volume of production: the value is expressed in Euro. The volume is expressed in a unit specific for each product.
The value and the volume of production represent
value and volume of product sales for own account
value and volume of products under sub-contracted operations (the value represents fee received for the sub-contracting operation)
By product:
value and volume of product sales for own account
value and volume of products under sub-contracted operations (the value represents fee received for the sub-contracting operation)
The volume unit: the measurement unit, such as kilograms, square meters etc. used to indicate the volume of goods produced.
Production type: either Sold Production (the value and volume of the product sold by the enterprise) or Total Production (the volume of all production of the product, including both the proportion that is sold and the proportion that is retained by the enterprise for adding to stocks, using in further processing etc.).
From 2006 onwards, for those products where Total Production is required, Sold Production by value and volume must be reported as well.
3.5. Statistical unit
The reporting unit is the local unit.
The observation unitis the local unit.
Results are not compiled at enterprise level because we have no information to neutralize flows between local units belonging to the same enterprise.
Sold production is only given by the local unit who made the sale. Total production is only given by the local unit that made this production.
3.6. Statistical population
The survey population is made up of all industrial enterprises in NACE07-33 (excluding 09, 19, 1011 and slaughterhouse part of 1051) :
with 20 persons employed or with a turnover of EUR 5.000.000 in the previous reference year;
with an industrial secondary activity with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year.
3.7. Reference area
All of the 3 regions of Belgium are covered (possibility of calculate regional results but there are some problems with confidentiality).
3.8. Coverage - Time
The period observed is one month before 2023 and yearly after 2024.
Belgian industrial products statistics date back to 1994.
This law stipulates that the national institute of statistics can only publish aggregate and anonymous results.
If a national data was previously indicated to Eurostat as being confidential, this should not be subsequently disseminate in any other public documents.
(to access from the main page choose yourlanguage and, click on the link "Release schedule calendar")
8.3. Release policy - user access
First provisional results are sent to Eurostat at T+6. These results are revised at T+12 and will be definitive at T+24.
Monthly national publication until 2024. From January 2024 onwards there is no monthly publication anymore.
Prodcom data are submitted annually to Eurostat and also published annually on our website.
10.1. Dissemination format - News release
Articles related to Prodcom results are published on this website.
10.2. Dissemination format - Publications
Data from 2004 to 2007 are available on pdf format and it can be provided on request.
Data from 2008 and 2009 are available on request.
Data (value of sold production) and data (volume of sold production) from 2010 are available on xls format at this webpage.
10.3. Dissemination format - online database
Not available.
10.4. Dissemination format - microdata access
Not available.
10.5. Dissemination format - other
The data are transmitted to Eurostat.
Data are also distributed to the national and international users on the 8 digits codes level and for the special request also on aggregated levels like 6,4,2 level.
10.6. Documentation on methodology
For users, the main documentation is the Prodcom Manual. This is available on under the label 'Métadonnées.
A number of internal papers and notes exists on methodological aspects and on description of the statistical production process.
10.7. Quality management - documentation
There is no user document available regarding the quality management.
11.1. Quality assurance
Prodcom is published with a high level of details, its accuracy therefore depends on data being checked in several ways.
Moreover, Prodcom is produced in close cooperation with the STS manufacturing indices, enabling cross-checks against each other.
11.2. Quality management - assessment
For each step of the production chain, the data are controlled.
For the electronic questionnaire, there are automatic checks. Error messages appear in the screen allowing the respondent to correct his data.
In our database, checks are made based on VAT data, based on the evolution of unit price and based on the past of the respondent.
During the routine calculation, checks are made on the aggregate data and microdata are controlled.
Before publication, data are validated by our validation service.
12.1. Relevance - User Needs
The main users of PRODCOM data can be presented in different goups, namely:
EXTERNAL/ INTERNATIONAL (OUTSIDE NSI), like:
Institutions at European level:
Eurostat: Prodcom provides statistics on the production of manufactured goods (Regulation ECC n°3924/91 and regulation EC n°912/2004); some Prodcom data are used for the European environmental economic accounts (Regulation EU n°691/2011)
Multi-national organizations:
United Nations: for the Industrial Commodity Statistics Questionnaire
FAO: for statistics on production for some agricultural and food products
EXTERNAL/ NATIONAL (OUTSIDE NSI), like:
Institutions at national level - governmental
Ministry of Economy: establishment of Prodcom statistics by Statistics Belgium
Institutions at national level - non-governamental
National Bank: establishment of national accounts
Trade associations: to have results concerning their sector
Media (the main channel of statistics to the general public)
Researchers and students at international or national level: some information are asked on certains products (ex: chocolate)
etc ...
INTERNAL (inside NSI), like:
National Bank: establishment of national accounts
Environmental statistics
Industrial output prices: Prodcom data are used for the weighting pattern
Industrial production: Prodcom is the principal source of data for the calculation of these indices
12.2. Relevance - User Satisfaction
No user satisfaction survey has been done for this statistics.
12.3. Completeness
All requested data according to the present Eurostat and EC EBS Regulation are provided.
13.1. Accuracy - overall
The survey sample is selected from business register (DBRIS).
The survey covers local units of industrial enterprises, which are wholly or primarily engaged in industrial production and industrial services in the Mining, Quarrying, and Manufacturing industries, i) with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year; ii) with an industrial secondary activity with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year.
If a company matches the selection criteria, a pre-survey is conducted to identify this company production local units.
According to the results of the pre-inquiry nace of the company can be changed in the registry.
Belgian Prodcom survey is a cut-off sample survey.
On the moment of production of the yearly PRODCOM data, the unweigthed response rate is approximately 90%.
For about 15% of local units, there are estimations for one of the variables.
The item non-response rates and imputation rates are:
Value of sold production: 14%
Volume of sold production: 17%
Value of products under sub-contracted operations: 20%
Volume of products under sub-contracted operations: 20%
Non respondents are re-contacted to ensure a response is received later on. Non-response is treated using several imputation techniques, depending on specific cases. If the correlation between data and VAT data is good then VAT data are used to estimate the turnover. Then this value is divided between all products codes which are relevant in the current period. When available montlhy production data on the NACE 4-digit level is used to impute variables based on past distributions of production.
If the correlation between data and VAT is not good then the past of the respondent unit or the data for the other respondent units are used.
In the case of respondents, estimates for missing volumes have been made in certain cases by using the average unit price from the respondent companies.
For users, the main documentation is the Prodcom Manual. This is available on under the label 'Métadonnées.
A number of internal papers and notes exists on methodological aspects and on description of the statistical production process.
There is a limited risk of multiple listing because the variables are not compiled at enterprise level before calculation of the results. We have no information to neutralize flows between local units belonging to the same enterprise and therefore it's also not possible to quantify the multiple listing.
13.2. Sampling error
The survey is done as a cut-off-sample, all units above a threshold are included.
There is no grossing up for units below the threshold. It is therefore not relevant to calculate sampling error.
There is an under-coverage because of the cutt-off sample used to reduce the burden on small enterprises.
Based on administrative data our sample covers 90% of the total turnover of the population.
13.3. Non-sampling error
There are several types of errors :
a respondent may give incorrect data;
the form/questionnaire may lead to the recording of wrong values;
error in manual entries for paper questionnaires;
in imputation, using ratios from responding enterprises to impute for non-respondents may have an effect on data quality but this is not quantifiable.
For each step of the production chain, the data are controlled.
For the electronic questionnaire, there are automatic checks. Error messages appear in the screen allowing the respondent to correct his data.
In our database, checks are made based on VAT data, based on the evolution of unit price and based on the past of the respondent.
During the routine calculation, checks are made on the aggregate data and microdata are controlled.
Before publication, data are validated by our validation service.
On the moment of production of the yearly PRODCOM data, the unweigthed response rate is approximately 90%.
For about 15% of local units, there are estimations for one of the variables.
The item non-response rates and imputation rates are:
Value of sold production: 14%
Volume of sold production: 17%
Value of products under sub-contracted operations: 20%
Volume of products under sub-contracted operations: 20%
14.1. Timeliness
Provisional yearly data are published on the national website, 5 months after the reference month.
This data for the reference year is revised in June when the prodcom annual data are sent to Eurostat. Once annual PRODCOM data has been validated by Eurostat, data is uploaded onto Eurostat’s website and also on Belgian website.
Data are recalcuated in T+12 and are considered as definitive in T+24.
14.2. Punctuality
Currently, there are no problem with the punctuality.
Annual data are sent to Eurostat before the end of June.
15.1. Comparability - geographical
Not available
15.2. Comparability - over time
An extensive revision of NACE in 2007 has resulted in changes to PRODCOM estimates for the 2008 survey onwards.
Every fifth year, there is a revision of the HS (Harmonized System) which will result also in more significant changes to CN and Prodcom. The latest major revision was in 2012. Revision in NACE will also affect Prodcom.
For STS manufacturing indices based on Prodcom data, data from 1995-2007 have been backcasted to NACE rev.2.
Since Belgium collected and published own and sc before 2021, there is no break.
15.3. Coherence - cross domain
Consistency checks are done with monthly industrial production and VAT data.
When there are differences, these cases are further scrutinised and corrected if necessary.
From January 2024 onwards data are no longer collected montlhy on the PRODCOM level and the monthly industrial production and VAT data will be replaced by yearly data and yearly cross-domain checks.
15.4. Coherence - internal
There is no problem of internal incoherence. Variables are checked for their internal coherence both at a national and European level.
Burden on respondents
The prodcom survey was simplified in 2008.
Year
Time to complete the survey
Hourly rate
Number of respondents
Periodicity (months)
Total (€)
2007
2,13
41
7.400
12
7.754.904
2013
1,49
41
4.500
12
3.298.860
2021
1,80
41
3.700
12
3.276.720
Cost associated with the collection and production of the industrial production index
~10 persons for ~15.000 hours/year
17.1. Data revision - policy
Revisions are made regularly on prodcom data. These corrections are made on data of current year, year-1 and year-2.
17.2. Data revision - practice
Data transmitted at the end of June are provisional (T+6). Data will be recalculated in T+12 and will be definitive in T+24.
18.1. Source data
The survey sample is selected from business register (DBRIS).
The survey covers local units of industrial enterprises, which are wholly or primarily engaged in industrial production and industrial services in the Mining, Quarrying, and Manufacturing industries :
with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year;
with an industrial secondary activity with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year.
A pre-inquiry is performed to determine the production local units of these enterprises.
Belgian Prodcom survey is a cut-off sample survey.
18.2. Frequency of data collection
Prodcom survey in Belgium is a yearly survey.
18.3. Data collection
Data are collected through web application based on XBRL and through paper questionnaires.
Web application
This web application is available at this webpage.
Information about this application and instructions for users are available on this webpage.
XBRL is an electronic language that has been specially developed for the exchange of business economic information through the internet.
XBRL (eXtensible Business Reporting Language) is an open standard based on XML (Extensible Markup Language) for the electronic collection and transfer of business economic data through the internet. The idea of the XBRL language is to identify each concept and add it to a 'taxonomy', which is similar to a dictionary.
These concepts, brought together in a structured way, can be recognised, processed and represented in diffirent ways, depending on the intended use.
The transmitted data is directly controlled by the system, the feedback is instantaneous.
Paper questionnaires
Responses of paper questionnaires are introduced in the operational system by an agent of our data collect center.
18.4. Data validation
Checks are carried out at the time of data collection and throughout the production of results. These are also validated by another service before distribution.
The web application performs automatic validation of reported data. If data are very different from those reported by the same company for the last period, the respondent will be prompted to check data again and to provide an explanation for the change.
Data (web and paper) are introduced in the operating system. In this operating system, there are some checks based on the VAT data, based on the variation of the data. Collected data are then daily transferred from the operating system in a datawarehouse. An error reporting with several information (included months for the different survey, included estimations) is sent to all users of DWH tables every day.
Then, a SAS program allows to check the variation of the different variables for a statistical unit and allows to identify problems. If a problem is detected, the collect center is contacted. In function of the response of our collect center, the declaration is maintained, re-estimated or removed for the calculation. Comparisons with VAT data are also made in this case.
Final tables containing results are then checked by our validation service: format, content, missing informations, layout of press release, etc.
If results appear abnormal, the validation service contacts the responsible for verification.
18.5. Data compilation
Non-response is treated using several imputation techniques, depending on specific cases. If the correlation between data and VAT data is good then VAT data are used to estimate the turnover. Then this value is divided between all products codes which are relevant in the current period.
If the correlation between data and VAT is not good then the past of the respondent unit or the data for the other respondent units are used.
The Prodcom survey is a census consequently grossing and weighting techniques are not used.
Results are aggregated to PRODCOM code level using sas codes. The confidentiality is also treated by semi-automatic sas procedures.
SAS codes allowing also to create the csv file that will be sent to Eurostat.
Some verification checks are made in this code.
18.6. Adjustment
At national level, data are published yearly and no adjustments are made.
No comments
The PRODCOM inquiry is conducted by Statistics Belgium. This is a yearly survey of the value and volume of products produced by industrial enterprises in Belgium and sold during the reference year. It is carried out in compliance with EU Council Regulation on the Establishment of a Community Survey of Industrial Production (N°3924/91).
The PRODCOM survey provides information on PRODuction of the European COMmunity and is compiled from data received from Belgian manufacturers.
The survey sample is approximately 3400 legal units and 3500 local units, covering 222 industries, derived from sub-sector, and 2067 products.
Available data include:
By product:
value and volume of product sales for own account
value and volume of products under sub-contracted operations (the value represents fee received for the sub-contracting operation)
actual production
29 August 2025
The purpose of the statistics is to report, for each product in the Prodcom List, how much has been produced in the reporting country during the reference period.
This means that Prodcom statistics relate to products (not to activities) and are therefore not strictly comparable with activity-based statistics such as Structural Business Statistics. The NACE codes on which Prodcom codes are based merely serve to identify the enterprises that should be surveyed in order to determine the amount of production of the product.
The products: the Prodcom List defines the products to be included in the survey. There are approximatly 4000 headings representing manufactured products and some industrial services.
The value and the volume of production: the value is expressed in Euro. The volume is expressed in a unit specific for each product.
The value and the volume of production represent
value and volume of product sales for own account
value and volume of products under sub-contracted operations (the value represents fee received for the sub-contracting operation)
By product:
value and volume of product sales for own account
value and volume of products under sub-contracted operations (the value represents fee received for the sub-contracting operation)
The volume unit: the measurement unit, such as kilograms, square meters etc. used to indicate the volume of goods produced.
Production type: either Sold Production (the value and volume of the product sold by the enterprise) or Total Production (the volume of all production of the product, including both the proportion that is sold and the proportion that is retained by the enterprise for adding to stocks, using in further processing etc.).
From 2006 onwards, for those products where Total Production is required, Sold Production by value and volume must be reported as well.
The reporting unit is the local unit.
The observation unitis the local unit.
Results are not compiled at enterprise level because we have no information to neutralize flows between local units belonging to the same enterprise.
Sold production is only given by the local unit who made the sale. Total production is only given by the local unit that made this production.
The survey population is made up of all industrial enterprises in NACE07-33 (excluding 09, 19, 1011 and slaughterhouse part of 1051) :
with 20 persons employed or with a turnover of EUR 5.000.000 in the previous reference year;
with an industrial secondary activity with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year.
All of the 3 regions of Belgium are covered (possibility of calculate regional results but there are some problems with confidentiality).
Before 2023: The month at which data are referring to.
Annual results are calculated on the basis of monthly data.
After 2023: The year the data are referring to.
The reference year is 2024.
The survey sample is selected from business register (DBRIS).
The survey covers local units of industrial enterprises, which are wholly or primarily engaged in industrial production and industrial services in the Mining, Quarrying, and Manufacturing industries, i) with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year; ii) with an industrial secondary activity with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year.
If a company matches the selection criteria, a pre-survey is conducted to identify this company production local units.
According to the results of the pre-inquiry nace of the company can be changed in the registry.
Belgian Prodcom survey is a cut-off sample survey.
On the moment of production of the yearly PRODCOM data, the unweigthed response rate is approximately 90%.
For about 15% of local units, there are estimations for one of the variables.
The item non-response rates and imputation rates are:
Value of sold production: 14%
Volume of sold production: 17%
Value of products under sub-contracted operations: 20%
Volume of products under sub-contracted operations: 20%
Non respondents are re-contacted to ensure a response is received later on. Non-response is treated using several imputation techniques, depending on specific cases. If the correlation between data and VAT data is good then VAT data are used to estimate the turnover. Then this value is divided between all products codes which are relevant in the current period. When available montlhy production data on the NACE 4-digit level is used to impute variables based on past distributions of production.
If the correlation between data and VAT is not good then the past of the respondent unit or the data for the other respondent units are used.
In the case of respondents, estimates for missing volumes have been made in certain cases by using the average unit price from the respondent companies.
For users, the main documentation is the Prodcom Manual. This is available on under the label 'Métadonnées.
A number of internal papers and notes exists on methodological aspects and on description of the statistical production process.
There is a limited risk of multiple listing because the variables are not compiled at enterprise level before calculation of the results. We have no information to neutralize flows between local units belonging to the same enterprise and therefore it's also not possible to quantify the multiple listing.
For volumes: units of measure are specific to each commodity code - for a comprehensive enumeration, please see the Prodcom list at this website.
For values: data are collected and stored in thousands of euros.
Non-response is treated using several imputation techniques, depending on specific cases. If the correlation between data and VAT data is good then VAT data are used to estimate the turnover. Then this value is divided between all products codes which are relevant in the current period.
If the correlation between data and VAT is not good then the past of the respondent unit or the data for the other respondent units are used.
The Prodcom survey is a census consequently grossing and weighting techniques are not used.
Results are aggregated to PRODCOM code level using sas codes. The confidentiality is also treated by semi-automatic sas procedures.
SAS codes allowing also to create the csv file that will be sent to Eurostat.
Some verification checks are made in this code.
The survey sample is selected from business register (DBRIS).
The survey covers local units of industrial enterprises, which are wholly or primarily engaged in industrial production and industrial services in the Mining, Quarrying, and Manufacturing industries :
with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year;
with an industrial secondary activity with 20 persons employed or with a turnover of EUR 5.000.000 or more in the previous reference year.
A pre-inquiry is performed to determine the production local units of these enterprises.
Belgian Prodcom survey is a cut-off sample survey.
Monthly national publication until 2024. From January 2024 onwards there is no monthly publication anymore.
Prodcom data are submitted annually to Eurostat and also published annually on our website.
Provisional yearly data are published on the national website, 5 months after the reference month.
This data for the reference year is revised in June when the prodcom annual data are sent to Eurostat. Once annual PRODCOM data has been validated by Eurostat, data is uploaded onto Eurostat’s website and also on Belgian website.
Data are recalcuated in T+12 and are considered as definitive in T+24.
Not available
An extensive revision of NACE in 2007 has resulted in changes to PRODCOM estimates for the 2008 survey onwards.
Every fifth year, there is a revision of the HS (Harmonized System) which will result also in more significant changes to CN and Prodcom. The latest major revision was in 2012. Revision in NACE will also affect Prodcom.
For STS manufacturing indices based on Prodcom data, data from 1995-2007 have been backcasted to NACE rev.2.
Since Belgium collected and published own and sc before 2021, there is no break.
