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.
The main purpose of the statistical operation of Services Sector Activity Indicators (SSAI) is to provide short-term performance indicators of the economic activity of companies operating in the non-financial market services sector in Spain via two variables: turnover and persons employed. The objective of this operation is to fulfil Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics, repealing 10 legal acts in the field of business statistics (EBS-Regulation) and Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistic (General Implementing Act).
Before the EBS, the STS legal base was Council Regulation No 1165/98 of 19 May 1998 concerning short-term statistics, amended by Regulation No 1158/2005 of 6 July 2005 concerning short-term statistics. The Article 25.3 of Regulation (EU) 2019/2152 mentions that Regulation (EC) No 1165/98 is repealed with effect from 1 January 2024.
These Regulations aim to create a common framework for the production of Community statistics on the short-term evolution of supply, demand, production factors and prices. This entails an actual set of statistics with direct data collection. Results are presented as indices so as to measure variations relative to the 2021 base year.
This operation began to be carried out in the year 2002, although we have data from year 2000. In the year 2005, the sample was increased in order to disseminate regional data, and as of January 2009 SSAI provides information in NACE Rev.2. In January 2013, coinciding with the base change from 2005 to 2010, the formulation used is modified, from direct Laspeyres-type indices with fixed base year 2005, to chain-linked Laspeyres indices, with base year 2010 (monthly chain-linking during last December).
3.2. Classification system
Statistical Classification of Economic Activities in the European Communities (NACE Rev.2).
3.3. Coverage - sector
Sections H (Transport and Storage), I (Accommodation), J (Information and Communications), L (Real Estate Activities), M (Professional, Scientific and Technical Activities, excluded 70.1, 72 and 75) and N (Administrative and Support Services Activities) of the Statistical Classification of Economic Activities in the European Communities (NACE Rev.2).
3.4. Statistical concepts and definitions
Enterprise classification variables
Economic activity
The economic activity carried out by a company is defined as the creation of added value by means of the production of goods and services.
Each one of the statistical units studied (enterprises) frequently carries out several activities that should be classified in separate categories of the National Classification of the Statistical Classification of Economic Activities in the European Community (NACE Rev.2).
In general, the activities carried out by an economic unit may be of three types: main, secondary and auxiliary activities. The main activity differs from secondary activities because it generates the greatest added value and auxiliary activities are those that generate services that are not sold on the market and only serve the unit they depend on (administration departments, transport or storage services).
Due to the difficulties companies face when calculating added value when various activities are carried out, the activity which generates the greatest volume of business is considered the main activity or, failing that, that which employs the greatest number of persons.
Although statistical units are classified according to their main activity, the information requested from the informant units refers not only to the main activity under consideration, but also to all the secondary and auxiliary activities which are carried out.
Enterprise size
The size of enterprises is one of the most important variables when it comes to classifying them. This size can be determined in terms of the magnitude of turnover or production value, or by considering the number of employees.
In SSAI, this latter option is chosen to determine the size of the companies.
Definition of variables
Income or turnover
Income or turnover is the total invoicing for the enterprise. Its definition is derived from accounting definitions used by companies. This includes those amounts invoiced by the enterprise due to the provision of services and the sale of goods that are the traffic objective, including those carried out through subcontracting.
Expenses invoiced for packaging and transport are included; the sale of purchased goods for resale in the same conditions as received, and sales of subproducts; as well as hours worked invoiced to third parties solely for subcontracted work.
Turnover includes the taxes that are levied on goods and services and excludes VAT paid by the customer.
It does not cover subsidies received from public authorities or from the European Union, financial benefits or other operating profit such as subsidies, sales of shares and fixed assets, income from interest, dividends and patents, leasing of enterprise and production unit property and machinery. Income from courtesy services to staff (dining rooms, etc.) and the provision of goods or services within the observation unit are also not included.
3.5. Statistical unit
KAU (kind of activity unit).
3.6. Statistical population
Sections H (Transport and Storage), I (Accommodation), J (Information and Communications), L (Real Estate Activities), M (Professional, Scientific and Technical Activities, excluded 70.1, 72 and 75) and N (Administrative and Support Services Activities) of the Statistical Classification of Economic Activities in the European Communities (NACE Rev.2).
The population is around 1.300.000 enterprises included in Other Services of the Statistical Classification of Economic Activities in the European Communities (NACE Rev.2.). The SBS population includes this enterprises, although its scope is bigger.
3.7. Reference area
The geographical coverage of Services is the entire Spanish territory excluding the Autonomous Cities of Ceuta and Melilla.
Turnover of the units located outside the Spanish territory are not included in the aggregates.
The sampled units located in Spain provide their total turnover.
3.8. Coverage - Time
The historical series began in January 2000 except for those corresponding to the activities:
58 of CNAE-09 (Publishing activities), which began in January 2002.
59 (Motion picture, television programme production and music publishing activities), 60 of CNAE-09 (Programming and broadcasting) and the overall turnover and employment indices for the Autonomous Communities, which began in January 2005.
68, 702, 77 and Real Estate Activities, which began in January 2021.
3.9. Base period
In 2024, the indices begin to be published using the new base 2021, in order to comply with the requirements established in Eurostat's Regulation on short-term statistics.
Services Sector Activity Indicators (SSAI) use chain linking indexes:
The base year: year 2021 (the year when the index equals 100)
The reference period: December t-1 (our chain linking is based on a period shorter than a year)
The reference year: December t-1 (the year on which the weights are based)
Our base and reference periods released nationally are equal than data transmitted to Eurostat.
Indices and percentage changes (compared to the previous period or compared to the same period in the previous year).
Variable reference period The reference period is the month.
6.1. Institutional Mandate - legal acts and other agreements
Legal basis
The legal basis for the STS indicators are Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics, repealing 10 legal acts in the field of business statistics (EBS-Regulation) and Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistic (General Implementing Act).
Before the EBS, the STS legal base was Council Regulation No 1165/98 of 19 May 1998 concerning short-term statistics, amended by Regulation No 1158/2005 of 6 July 2005 concerning short-term statistics. The Article 25.3 of Regulation (EU) 2019/2152 mentions that Regulation (EC) No 1165/98 is repealed with effect from 1 January 2024.
According to the Spanish legislation, SSAI is a compulsory statistic.
The compilation and dissemination of data are governed by the Statistical Law No 12/1989 "Public Statistical Function" of May 9, 1989, and Law No 4/1990 of June 29 on "National Budget of State for the year 1990" amended by Law No 13/1996 "Fiscal, administrative and social measures" of December 30, 1996, makes compulsory all statistics included in the National Statistics Plan. The Plan contains the statistics that must be developed in the four year period by the State General Administration's services or any other entity dependent on it. All statistics included in the National Statistics Plan are statistics for state purposes and are obligatory. The National Statistics Plan 2021-2024, approved by Royal Decree 1110/2020, of December 15th, is the Plan currently implemented. This statistical operation has governmental purposes, and it is included in the National Statistics Plan 2021-2024, which is available in the following link.
6.2. Institutional Mandate - data sharing
The exchanges of information between the INE and the rest of the State statistical offices (Ministerial Departments, independent bodies and administrative bodies depending on the State General Administration), or between these offices and the Autonomous Communities Statistical Offices, are regulated in the Statistical Law No 12/1989 "Public Statistical Function". This law also regulates the mechanisms of statistical coordination and concludes cooperation agreements between the different offices when necessary.
Through the collaboration agreements signed with regional statistical institutes, monthly information exchange is done meeting the requirements established to preserve statistical confidentiality.
Data are sent to EUROSTAT.
7.1. Confidentiality - policy
SSAI is a statistical operation included in the National Statistics Plan 2021-2024, therefore, it is subject to Statistics Law No 12/1989 "Public Statistical Function" and, consequently, data is protected by the principle of statistical confidentiality in all the storages of production.
Regulation (EC) No 223/2009 of the European Parliament and of the Council on European statistics stipulates the need to establish a guideline and common principles to ensure confidentiality of data used during the production of European statistics and also when accessing the confidential data.
INE Spain Confidentiality Policy can be found in the following website.
7.2. Confidentiality - data treatment
INE Spain takes the necessary logical, physical and administrative measures to protect confidential data, from its data collection to its dissemination.
Questionnaires of the survey include a legal clause related to protection and preservation of data collected.
In the case of this survey, only aggregate information is disseminated. Furthermore, data are only published down to an activity level where they are not confidential, and no confidential data can be obtained.
8.1. Release calendar
The advance calendar is published in the last quarter of each year and shows the exact publication dates.
All users have access to the data since 9 am on the press release day. The data are disseminated on INE Spain website. Users are informed that the data are being released via social network ‘X’ and also in the "News flash" section of the INE Spain website.
In accordance with the European Statistics Code of Practice, all users have access to the same statistical information at the same time, and any privileged pre-release access to any external user must be limited, controlled and made public.
In this regard, a few hours before its official publication, some statistical information of the INE is transmitted, under embargo, to the appointed body, the Secretariat of State for Economic Affairs and Business Support of the Ministry of Economic Affairs and Digital Transformation, in order to provide sufficient time for their responsible parties to prepare answers to any possible questions that journalists might ask.
Likewise, pursuant to the Multilateral Working Agreement for the development of a statistics partnership between the INE and the Central Statistics Offices of the Autonomous Communities (OCECA), seventeen OCECA receive information under embargo, relating to both their territory (at Autonomous Community level) and the national total (at a national scale), in order to compile and publish, in parallel with the INE, press releases related to their respective territorial scopes. Like the INE, the OCECA are governed by the laws and regulations that guarantee statistical confidentiality.
The release policy does not determine when the data will be released. The dissemination policy is available in the following link.
The same release policy is applied to national releases and to transmissions to Eurostat.
Data are transmitted in SDMX-ML format to Eurostat at the time of the national release.
Monthly. The publication is released approximately 50 days after the end of the reference period.
There are no paper publications; the information related to this indicator is published on the INE website.
In it is possible to find the following information:
Press release: where the components that most influence the evolution of the indicator in the reference month are highlighted.
Principal results: it contains the attached tables to the press release.
Detailed Results: with the information of the historical series of the indicators.
Methodology: where the methodological characteristics of the indicator are outlined with major detail.
The weighting used in the aggregated indices.
Results of the indicators in previous bases are attached .
EUROSTAT, the statistical office of the European Union, provide the European Union with statistics at European level that enable comparisons between countries and regions. This information is available in the Database - Short-term business statistics - Eurostat.
10.3. Dissemination format - online database
INEbase is the system that the INE uses to store and disseminate all the statistical information online. It contains all the information that the INE produces in electronic format. It is organized basically following the thematic classification of the Inventory of Statistical Operations of the General Administration of the State.
The basic unit is the statistical operation defined as the set of activities that lead to obtaining of statistical results for a certain sector or topic, based on information collected individually.
In 2024 the total number of accesses to Services Sector Activity Indicators (SSAI) website was AC1=142,129 and the total number of accesses to Services Sector Activity Indicators (SSAI) metadata was AC2=1860.
10.4. Dissemination format - microdata access
The microdata for the Services file Sector Activity Indicators (SSAI) are not accessible to the public. The microdata file is not disseminated anonymously; it is provided only to the “Oficinas Centrales de Estadística de las Comunidades Autónomas” (OCECAS) with which an agreement of collaboration has been signed for the reference of the above-mentioned information, by virtue of the Framework agreement of Collaboration between the INE and the OCECAS.
10.5. Dissemination format - other
Information can be requested from the INE’s customer service area. Confidentiality and accuracy are taken into account when processing these requests.
Our data are sent to Eurostat, either to be used in European aggregates or to be released also as national data.
10.6. Documentation on methodology
The documentation on the methodology is available at the link:
The statistics are prepared in accordance with EU guidelines in order to meet the Code of Practice implemented by Eurostat following the quality criteria on relevance, accuracy, timeliness and punctuality, accessibility and clarity, comparability, consistency and completeness.
To ensure the quality of statistics, every three years the Commission shall submit a report to the European Parliament and the Council on the statistics compiled pursuant to Regulation (EC) No 1165/1998 of the Council, in particular on their relevance and quality, as well as on the revision of indicators.
This report addresses the relevance, quality and review of the indicators, as well as the cost of the statistical system and the burden for businesses in relation to its benefits.
There are not additional national quality reports.
Report from the commission to the European Parliament and the Council concerning short term statistics as required by Article 14(2) of Council Regulation (EC) No 1165/1998.
11.1. Quality assurance
The frame of quality guarantee for the statistics of the INE is based on the ESCOP, the Code of Good practices of EUROSTAT'S European Statistics. The ESCOP consists of 15 principles grouped in three areas: Institutional Environment, Processes and Products. Every principle has associated a series of indicators that allow its measure. To assess quality, various tools provided by EUROSTAT are use: the aforementioned indicators, the self-evaluation (DESAP), the Review for counterparts (Peer Review), the Audits of quality and the user satisfaction surveys.
The Services Sector Activity Indicators (SSAI) statistical operation is carried out in accordance with the EU directives, following the Code of Good Practices implemented by Eurostat and taking the necessary measurements to ensure that the published indicators fulfill all the requirements of quality needed.
11.2. Quality management - assessment
The statistical operation of the Service Sector Activity Indicators (ISAS) constitutes a high-quality product. Its sample size allows for information down to the autonomous level within reasonable margins of sampling error. The harmonized methodology used allows for international comparisons with a high level of reliability. And it provides an extremely valuable measure of the services sector, which no other statistical source can provide with the same level of detail.
Regarding the limitations of the survey, it is necessary to indicate those inherent to statistical sampling operations, such as nonresponse and the already mentioned sampling errors or the coefficients of variation of the estimations. In both cases, they remain within reasonable limits.
The Regulation n º 1158/2005 of the European Parliament and of the Advice of July 6, 2005, establishes that the quality of the variables will be from time to time verify by comparing them with other statistical information. The above-mentioned comparison will be carried out by each of the members states and by the Commission (Eurostat). In addition, there will be internal coherence.
Likewise, it fulfills the quality criteria enumerated in the article 12, paragraph 1, of the Regulation (CE) n º 223/2009 of the European Parliament and of the Advice, namely: relevancy, precision, current importance, punctuality, accessibility, clarity, comparability and coherence.
12.1. Relevance - User Needs
The statistical operation of the Services Sector Activity Indicators (SSAI) covers the needs of statistical information of different user groups:
The INE itself, which uses this statistic as a basis to compile the national accounts.
Departments and other public bodies that use the results of this statistical operation to make economic analysis or predictions, to elaborate synthetic indicators of activity, etc.
Territorial administrations like the autonomous communities. All Institutes of statistics in the autonomous communities, by virtue of the agreements established (delete) between the INE and these institutions, they (delete) receive monthly the (delete) information (microdata or already compiled regional indexes) corresponding to enterprises of the sample who belong to their region (to the sampled companies in their region). With this information, the Services Sector Activity Indicators there (delete) can be disseminated (SSAI) to a major level of detail (with greater detail) for their Community, which is useful for measuring the economic cycle and as an instrument of regional economic analysis.
Investigators and academic world, enterprises and non-profit institutions (services of study of enterprises, foundations, associations, etc.), unions and employer organizations, press and specialized media, as well as foreign users, both institutional and individuals. Among the first of these latter stand out, Eurostat, the European Central Bank and the OECD.
12.2. Relevance - User Satisfaction
The INE has made general user satisfaction surveys in 2007, 2010, 2013, 2016 and 2019 and plans to continue doing so every three years. Their purpose is to know statistics users' perceptions of the quality of the information they use and the extent to which their needs are met. In addition, other special surveys are conducted to address specific topics: information dissemination, the quality of certain publications, etc. In addition to the surveys, experts are in constant contact with main users to address any specific information needs.
In the latest user satisfaction survey conducted in 2019, the " Short-term of the sectorial and labour activity " sector, which includes this operation, received a user rating of 3.80 out of 5.
The national user surveys' results are available in the website.
12.3. Completeness
The survey responds to all requests for information, both national and international regulations.
Available statistical results rate is R1= 100%.
13.1. Accuracy - overall
The sample design attempts to minimize sampling errors. Also, errors are reduced as far as possible during the survey process: in data collection (monitoring and controlling response rate of the edition) and later in the non-response imputation phase and calculation of aggregate indices. This allows a high degree of reliability of the statistical operation.
Random stratified sampling has been used. In order to set out a measurement of the quality of the indices, an approximate relative sampling error is calculated for the interannual variation rates.
CV Annual Average 2024: 1.19%
Non sampling errors are also small: over-coverage, multiple listings, non-response, imputation, etc., can be considered minor.
Our data are final after five periods from the first release, and the difference between the first and last publications is small.
13.2. Sampling error
Stratified Random Sampling has been used. The distribution of the sample among the different strata is carried out by Neyman allocation. A random sample is obtained in each stratum except in companies with 200 or more employees, in which all are part of the sample. In certain branches and communities, the strata are smaller due to a low population. In these cases, the strata are also exhaustive.
The enterprise framework is stratified taking the following variables into account: branch of activity, Autonomous Community and size, measured by number of employees.
In order to set out a measurement of the quality of the indices, an approximate relative sampling error is calculated for the interannual variation rates of the variable “number of employees and self-employed workers”. The general expression for the estimated relative error, a supposedly negligible bias, is given by:
CV*(R*)=100×(V(R*))1/2/R* , being R*=Y*t/Y*t-1
R* is the ratio estimator, given by the result of the estimation of the total of variable Y (number of employees and self_employed persons) in month m of the year t.
Y*t , and the estimation of Y obtained in the same month m of the year t-1, Y*t-1.
The method used for calculating the estimator of variance is linearization via Taylor series. The expression for the estimator of variance is:
Where V*(Y*t) shows the estimator of variance of Y*t, V*(Y*t-1) shows the estimator of variance of Y*t-1 and Cov*(Y*t,Y*t-1) shows the estimator of the covariance between Y*t and Y*t-1.
Month
CV
CV0124
1.39622425525143
CV0224
1.17077917832587
CV0324
1.13226739480797
CV0424
1.18509589021428
CV0524
1.30025676101128
CV0624
1.23330365727252
CV0724
1.00975070972802
CV0824
0.946669411677087
CV0924
1.19291218549
CV1024
1.13221355899445
CV1124
1.29434416819894
CV1224
1.27712653169427
Average 2024
1.19
13.3. Non-sampling error
Impact of over-coverage and multiple listings of the frame
We have studied the impact of over-coverage and multiple listings of the frame through new units that are selected to enter our sample.
In September of each year, we select units for an annual renewal: 20-25% of sampling units are rotated in non-exhaustive strata, and we select new big enterprises who belong exhaustive strata. In January of each year, part of the sample is renovated.
These results have been calculated with the new units selected in September 2024:
Over-coverage
Units that do not belong to the target population: 0.3%
Units that have closed: 2.22%
Units without any data: 12%
Multiple listings: Units are present more than once in the frame
% Impact of multiple listings: 0.01%
Under-coverage
We do not have information about the unit.
Impact of survey instrument, respondent and interviewer
Our survey uses simple questionnaires with a contact person who could solve any doubts respondents may have. These contact persons have been working at the INE for a long time and are prepared to answer any question.
Furthermore, most questionnaires are collected online. We use a selective editing system for web-based questionnaires, as most have undergone multiple editing checks and respondents must correct them or provide us with an explanation about their data.
When we receive all data, we also review them through selective editing. If an important error is detected, we recontact the respondent again and we correct it.
Impact of unit and non-response item
In 2024, our sample consisted of 30,116 units. The average non-response rate was low.
First release:
Average rate, A4. Year 2024 (January-December): 6.45%
Last release:
Average rate, A4. Year 2024 (January-October): 4.29%
Weighted average rate, A4. Year 2024: 2.72% in first release.
Weighted average rate, A4. Year 2024(January-October): 1.21% in last release.
Impact of data editing, coding and imputation
Data editing: a first editing phase, at micro-data level takes place when the respondent fills in the e-questionnaire. We use selective editing in them. Data collected through other channels (paper, telephone, fax) are also edited using selective editing at the time of recording. During data collection another editing phase occurs at the microdata level which includes all data collection methods questionnaires (all the questionnaires from the data collection methods). After that, the selective editing is carried out. Finally, the macro-data editing phase takes place.
Imputation: monthly and annual rates of units filling in the questionnaire in each elementary aggregates are imputed to those units in the same elementary aggregate that did not answer during the reference period. Elementary aggregates are the most detailed levels for which indices are calculated.
% data editing, coding and imputation (October 2024):
editing: 1.22%
imputation: 9.27%
Impact of specific models used in estimation
In order to ensure the quality of the adjusted data, the INE validates the data using the INE Standard quality measures (chapter 4): verification of the adequacy of the decomposition (specification) through the diagnoses provided by the SEATS program and verification of the absence of autocorrelation, absence of residual seasonality in the seasonally adjusted series and absence of residual calendar effects in the seasonally adjusted series.
Our questionnaire could be found in the Annex of this document.
14.1. Timeliness
The publication is disseminated approximately 50 days after the end of the reference period.
14.2. Punctuality
The results are published punctually according to the calendar of availability of statistical relating to the moment publications elaborated and spread by the INE.
The above mentioned calendar is fulfilled with complete punctuality.
15.1. Comparability - geographical
The survey is designed to be consistent across the entire geographic area in terms of its methodology, design, data collection process, and data editing.
Comparability among EU countries is ensured by the EBS-Regulation (Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics, repealing 10 legal acts in the field of business statistics) and its General Implementing Act (Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistic). They provide these statistics with homogeneity regarding the rest of European Union countries, which disseminate the same information.
15.2. Comparability - over time
All published historical series of the Service Sector Activity Indicators (SSAI) are homogeneous and, therefore, comparable over time.
Since the survey was implemented, there have been base changes of the indices and of the classification of economic activities. However, to ensure the comparability of the information over time, the INE has always published the retrospective series of the Service Sector Activity Indicators (SSAI) using the new base and the new classification.
The number of comparable data, CC2, from January 2000 to December 2024 is: 300.
15.3. Coherence - cross domain
In the data validation we take into account data of other statistics to ensure consistency between them, such as:
Structural Business Statistics: Services Sector (SBS) of INE
SBS and STS use the same Population framework. Our sampling frame is the Statistical Business Register. The sampling design is also quite similar. Both apply stratified sampling with optimum sample allocation. Strata are formed using the following variables: Main economic activity, number of employee and region (NUTS2).
In the SBS, companies with 50 or more employees belong to the take-all stratum. In STS, for services, companies with 200 or more employees belong to take-all stratum.
Both indices use Horvitz-Thompson estimators but there are some differences in the estimation process. SBS estimators are adjusted for stratum changes and the total population, N, is estimated using their own data. In STS, N comes from the updated sampling frame.
Besides, SBS is designed to obtain level turnover, and STS is designed to study increases or decreases of turnover.
Quarterly Spanish National Accounts of INE
STS Services is used as input for the Quarterly Spanish National Accounts (QSNA).
STS Services index is use in aggregates:
Household final consumption expenditure
Gross Value Added on Services activities:
Wholesale and Retail trade, Transports and Hotels;
Information and communication;
Professional activities.
Gross Fixed Capital Formation Intellectual Property Products
Although in these aggregates there are more indicators. We should have into account that QSNA uses more than 250 short term indicators from different institutions.
Sales of the Big enterprises of the Tax Authority
The Tax Authority publishes a monthly publication on the turnover of large companies based on VAT data. Large companies are considered to be individuals or legal entities whose turnover exceeded 6.01 million euros during the immediately preceding calendar year. This definition differs from that used in European legislation.
Large companies operate in all sectors; in some, their representation is practically exhaustive, while in others, where small companies predominate, their weight is lower. The number of large companies varies each year. They use a Unit Value Index (STS). The STS includes all units, large and small. Our sample has a turnover rate of between 20% and 25%; we use a Laspeyres Chain Index.
15.4. Coherence - internal
Coherence is a key issue both in the presentation of the survey methodology and throughout its production process. The coherence between the variables is confirmed in all the phases of the statistical process and no incoherences are detected.
In order to significantly reduce the burden on informants from reporting units, the questionnaire has been designed to contain all the necessary information while keeping its design as simple as possible to facilitate the response of respondents, thus minimizing the burden on certain informants who have to complete multiple surveys in a single year.
To avoid informant fatigue and sample aging, and to allow the selection of newly created companies, an annual rotation of 25% is carried out.
Regarding the cost associated with the collection and production of statistical products, in case of the SSAI, the budget Forecasts in the 2024 Annual Program reach an amount of 3,390.40 (thousands of Euros). (space) This budget includes turnover and employment.
Burden (respondents) hours per year: 9,797 h
Cost (NSI) hours per year: 51,650 h
17.1. Data revision - policy
Revised data are used both for the National dissemination and to obtain the STS data transmitted to Eurostat, so the same policy is applied to the STS data released nationally and sent to Eurostat.
The revision policy adopted for the raw data at INE Spain encompasses several types of revisions:
Periodic monthly reviews are conducted. This review consists of verifying new information available on the reference month's data, in addition to previously published data from previous months. Seasonal adjustment models are reviewed annually in order to update them.
Major revisions are made when methodological or base changes occur, such as changes in definitions and classifications, as well as legal changes such as those due to new regulations or updates to existing ones.
Non-scheduled revisions rarely take place in the case an error is detected in the statistical process.
The revision policy regarding seasonally adjusted data comprises two types of revisions:
Concurrent adjustment: the partially concurrent adjustment is carried out monthly. With the new data received:
Both the parameters of the ARIMA model and the parameters of the regressors for the correction of calendar effects and of the intervention variables of the outliers are re-estimated.
The filters to obtain the seasonally adjusted series are re-calculated.
Current adjustment: annually, the model (including the intervention variables of the outliers and the regressors for the correction of calendar effects) is identified, the parameters (of ARIMA, of the regressors for the correction of calendar effects and of the intervention variables of the outliers) are estimated, and the filters are calculated and set.
The link to the revision policy for raw data adopted by the INE Spain is as follows: Revision policy.
There is no official revision calendar, each indicator has its own revision calendar.
17.2. Data revision - practice
The first time the index is published for a given month, provisional data are provided, which are not final until four months later. In the meantime, the index is revised using the same data modifications that have occurred since the date of the first publication due, for example, to the replacement of the estimated data (for management issues) with the updated data completed by the company. However, the series corrected for calendar effects and seasonal effects are fully reviewed each month.
Indicator A6, average size of revisions, where revision is defined as the difference between final and preliminary estimators, is used for data revision practices. Estimators are calculated over several of reference periods: MAR (size of revisions) and RMAR (size of each revision (delete) compared to the last estimate).
Original series:
A6_MAR (December 2024) =0.082797
A6_RMAR (December 2024) =0.945472%
A6_MR (December 2024)=-0.041484
A6_MaxARt =0,297536 (Annual variation rate July 2023_ July 2024)
Series adjusted for calendar effects:
A6_MAR (December 2024) = 0.086949
A6_RMAR (December 2024) = 1.035540%
A6_MR (December 2024)=-0.045307
A6_MaxARt = 0.353321 (Annual variation rate March 2021_ March 2022)
MAR, RMAR, MR and MaxARt values were obtained with the information of its last 36 final months being December 2024 the last one.
Benchmarking is not performed. Nor are the company's quarterly/annual accounts, national accounts, or structural business statistics reviewed.
Major revisions are informed in advance in the press release on the day of their dissemination, although main users are informed in advance. Additionally, since 2015 we communicate major updates on our website.
18.1. Source data
Type of source
Statistical survey.
Frame on which the source is based
The source is the DIRCE (INE's business register).
The DIRCE has been used as the framework for the survey, which contains information on the main economic activity and on the number of employees in companies, allowing for stratification according to these concepts This directory also includes other data on the identification and location of statistical units, which are necessary for the correct collection of information.
The data contained in the DIRCE are obtained using administrative sources, primarily from the Tax Agency and Social Security, and are completed with information from the statistical operations of the INE.
Sample or census
Some strata are investigated exhaustively (strata with enterprises with 200 or more employees) and in the remaining strata, a systematic selection with a random start is carried out, having previously ordered enterprises by size.
Criteria for stratification
A stratified one-stage sampling is used, where, as a general rule, the enterprise population has been stratified according to the cross of three variables: principal activity at 4-digit level of NACE Rev 2, the number of employees and region. Information from the Structural Business Survey has been used in order to obtain the sample sizes taking into account the weight of the different strata in the total turnover for each activity and information of variance of the variable number of employees by stratum from the Business Register.
Frequency of updating the sample
A fixed sample is used every year. At the end of the year the new firms belonging to the exhaustive stratum are included. One quarter (25%) of the rest of the sample is replaced by other firms of the frame. With the change of base year, a complete update is carried out.
18.2. Frequency of data collection
Data are collected monthly.
18.3. Data collection
The questionnaire used is a specific monthly questionnaire named “Indicadores de Actividad del Sector Servicios”.
The questionnaire is sent by postal mail and there are also telephone and fax contacts, and companies can fill the questionnaire online.
18.4. Data validation
During the questionnaire completion stage, a system is used to detect inconsistencies so that the respondent is advised and required to confirm or modify the information provided.
When INE staff records this information, a monitoring system is in place to alert them to possible inconsistencies. During this process, in the event of discrepancies in the data, the staff responsible for collection contacts the informant to confirm or modify the information provided.
Once all the information has been collected, a coverage check is performed to ensure the completeness of the recorded data and to detect duplicates and coverage errors, non-response, etc.
18.5. Data compilation
The questionnaires are collected by the INE’s data collection staff and there is an initial cleansing of errors according to a set of edits. Microediting consists of validating data for each local unit, attending to its historical values. The analysis establishes monthly and annual comparisons so that the data can be validated without contacting the informants when the evolution repeats periodically.
Next, at the centralized level, the selective editing phase, imputation (nonresponse is imputed using the data of the previous month of the non-responding units and the variation rate of the units in the same stratum) and the macro-editing process take place (are carried out). In some cases, manual editing is required, such as, for large nonresponding units.
Finally, the estimators are calculated by aggregating the elementary indices taking into account the weightings. In the final phase, the dissemination files are prepared, along with the press release according to the dissemination breakdowns.
The indices are obtained using a chaining-linking. The elementary indices are calculated from the respondents (imputed in case of non-response) common to the reference month and the previous December.
The weighting variable is calculated from the SBS 2021 employment, and these weights are updated by Services indexes in base 2021 to obtain weights in December (t-1) in order to calculate the chain linking index.
18.6. Adjustment
Calendar effects: the impact on a variable’s time series due to the different structure of months in different years (both in length and composition) will remain constant if other factors influence this variable. The SSAI is a very sensitive indicator of calendar effects since a holiday in a particular month represents a decrease in the proportional turnover of such month.
To overcome this problem and obtain a number that uniformly compares production over time, the series are corrected taking into account the following schedule: trading days, Easter and leap year.
After editing the calendar effects, seasonal effects are removed. Seasonal fluctuations are movements that occur with similar intensity each month and are expected to continue.
The seasonally adjusted series, i.e. adjusted for seasonal and calendar effects, provide an estimate of the "new" factors in a series (changes in trend, cycle and irregular component).
The method is based on regression models involving variables from the three effects previously mentioned. We use JDEMETRA program.
The adjusted series are published together with the uncorrected series to complete the analysis of the situation.
To understand the impact of the COVID crisis in 2020, we had to analyze outliers on a monthly basis.
The main purpose of the statistical operation of Services Sector Activity Indicators (SSAI) is to provide short-term performance indicators of the economic activity of companies operating in the non-financial market services sector in Spain via two variables: turnover and persons employed. The objective of this operation is to fulfil Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics, repealing 10 legal acts in the field of business statistics (EBS-Regulation) and Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistic (General Implementing Act).
Before the EBS, the STS legal base was Council Regulation No 1165/98 of 19 May 1998 concerning short-term statistics, amended by Regulation No 1158/2005 of 6 July 2005 concerning short-term statistics. The Article 25.3 of Regulation (EU) 2019/2152 mentions that Regulation (EC) No 1165/98 is repealed with effect from 1 January 2024.
These Regulations aim to create a common framework for the production of Community statistics on the short-term evolution of supply, demand, production factors and prices. This entails an actual set of statistics with direct data collection. Results are presented as indices so as to measure variations relative to the 2021 base year.
This operation began to be carried out in the year 2002, although we have data from year 2000. In the year 2005, the sample was increased in order to disseminate regional data, and as of January 2009 SSAI provides information in NACE Rev.2. In January 2013, coinciding with the base change from 2005 to 2010, the formulation used is modified, from direct Laspeyres-type indices with fixed base year 2005, to chain-linked Laspeyres indices, with base year 2010 (monthly chain-linking during last December).
29 May 2025
Enterprise classification variables
Economic activity
The economic activity carried out by a company is defined as the creation of added value by means of the production of goods and services.
Each one of the statistical units studied (enterprises) frequently carries out several activities that should be classified in separate categories of the National Classification of the Statistical Classification of Economic Activities in the European Community (NACE Rev.2).
In general, the activities carried out by an economic unit may be of three types: main, secondary and auxiliary activities. The main activity differs from secondary activities because it generates the greatest added value and auxiliary activities are those that generate services that are not sold on the market and only serve the unit they depend on (administration departments, transport or storage services).
Due to the difficulties companies face when calculating added value when various activities are carried out, the activity which generates the greatest volume of business is considered the main activity or, failing that, that which employs the greatest number of persons.
Although statistical units are classified according to their main activity, the information requested from the informant units refers not only to the main activity under consideration, but also to all the secondary and auxiliary activities which are carried out.
Enterprise size
The size of enterprises is one of the most important variables when it comes to classifying them. This size can be determined in terms of the magnitude of turnover or production value, or by considering the number of employees.
In SSAI, this latter option is chosen to determine the size of the companies.
Definition of variables
Income or turnover
Income or turnover is the total invoicing for the enterprise. Its definition is derived from accounting definitions used by companies. This includes those amounts invoiced by the enterprise due to the provision of services and the sale of goods that are the traffic objective, including those carried out through subcontracting.
Expenses invoiced for packaging and transport are included; the sale of purchased goods for resale in the same conditions as received, and sales of subproducts; as well as hours worked invoiced to third parties solely for subcontracted work.
Turnover includes the taxes that are levied on goods and services and excludes VAT paid by the customer.
It does not cover subsidies received from public authorities or from the European Union, financial benefits or other operating profit such as subsidies, sales of shares and fixed assets, income from interest, dividends and patents, leasing of enterprise and production unit property and machinery. Income from courtesy services to staff (dining rooms, etc.) and the provision of goods or services within the observation unit are also not included.
KAU (kind of activity unit).
Sections H (Transport and Storage), I (Accommodation), J (Information and Communications), L (Real Estate Activities), M (Professional, Scientific and Technical Activities, excluded 70.1, 72 and 75) and N (Administrative and Support Services Activities) of the Statistical Classification of Economic Activities in the European Communities (NACE Rev.2).
The population is around 1.300.000 enterprises included in Other Services of the Statistical Classification of Economic Activities in the European Communities (NACE Rev.2.). The SBS population includes this enterprises, although its scope is bigger.
The geographical coverage of Services is the entire Spanish territory excluding the Autonomous Cities of Ceuta and Melilla.
Turnover of the units located outside the Spanish territory are not included in the aggregates.
The sampled units located in Spain provide their total turnover.
Variable reference period The reference period is the month.
The sample design attempts to minimize sampling errors. Also, errors are reduced as far as possible during the survey process: in data collection (monitoring and controlling response rate of the edition) and later in the non-response imputation phase and calculation of aggregate indices. This allows a high degree of reliability of the statistical operation.
Random stratified sampling has been used. In order to set out a measurement of the quality of the indices, an approximate relative sampling error is calculated for the interannual variation rates.
CV Annual Average 2024: 1.19%
Non sampling errors are also small: over-coverage, multiple listings, non-response, imputation, etc., can be considered minor.
Our data are final after five periods from the first release, and the difference between the first and last publications is small.
Indices and percentage changes (compared to the previous period or compared to the same period in the previous year).
The questionnaires are collected by the INE’s data collection staff and there is an initial cleansing of errors according to a set of edits. Microediting consists of validating data for each local unit, attending to its historical values. The analysis establishes monthly and annual comparisons so that the data can be validated without contacting the informants when the evolution repeats periodically.
Next, at the centralized level, the selective editing phase, imputation (nonresponse is imputed using the data of the previous month of the non-responding units and the variation rate of the units in the same stratum) and the macro-editing process take place (are carried out). In some cases, manual editing is required, such as, for large nonresponding units.
Finally, the estimators are calculated by aggregating the elementary indices taking into account the weightings. In the final phase, the dissemination files are prepared, along with the press release according to the dissemination breakdowns.
The indices are obtained using a chaining-linking. The elementary indices are calculated from the respondents (imputed in case of non-response) common to the reference month and the previous December.
The weighting variable is calculated from the SBS 2021 employment, and these weights are updated by Services indexes in base 2021 to obtain weights in December (t-1) in order to calculate the chain linking index.
Type of source
Statistical survey.
Frame on which the source is based
The source is the DIRCE (INE's business register).
The DIRCE has been used as the framework for the survey, which contains information on the main economic activity and on the number of employees in companies, allowing for stratification according to these concepts This directory also includes other data on the identification and location of statistical units, which are necessary for the correct collection of information.
The data contained in the DIRCE are obtained using administrative sources, primarily from the Tax Agency and Social Security, and are completed with information from the statistical operations of the INE.
Sample or census
Some strata are investigated exhaustively (strata with enterprises with 200 or more employees) and in the remaining strata, a systematic selection with a random start is carried out, having previously ordered enterprises by size.
Criteria for stratification
A stratified one-stage sampling is used, where, as a general rule, the enterprise population has been stratified according to the cross of three variables: principal activity at 4-digit level of NACE Rev 2, the number of employees and region. Information from the Structural Business Survey has been used in order to obtain the sample sizes taking into account the weight of the different strata in the total turnover for each activity and information of variance of the variable number of employees by stratum from the Business Register.
Frequency of updating the sample
A fixed sample is used every year. At the end of the year the new firms belonging to the exhaustive stratum are included. One quarter (25%) of the rest of the sample is replaced by other firms of the frame. With the change of base year, a complete update is carried out.
Monthly. The publication is released approximately 50 days after the end of the reference period.
The publication is disseminated approximately 50 days after the end of the reference period.
The survey is designed to be consistent across the entire geographic area in terms of its methodology, design, data collection process, and data editing.
Comparability among EU countries is ensured by the EBS-Regulation (Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics, repealing 10 legal acts in the field of business statistics) and its General Implementing Act (Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistic). They provide these statistics with homogeneity regarding the rest of European Union countries, which disseminate the same information.
All published historical series of the Service Sector Activity Indicators (SSAI) are homogeneous and, therefore, comparable over time.
Since the survey was implemented, there have been base changes of the indices and of the classification of economic activities. However, to ensure the comparability of the information over time, the INE has always published the retrospective series of the Service Sector Activity Indicators (SSAI) using the new base and the new classification.
The number of comparable data, CC2, from January 2000 to December 2024 is: 300.
