Since 2016, the calculation of turnover indices is based on the exhaustiveness of the monthly declarations of the tax source. This new index calculation methodology has been implemented in order to better take into account the demographics of companies and thus more accurately reflect the economic reality. Indeed, the turnover indexes of an N year previously calculated did not take into account the start-ups, discontinuations, entry and exit of fields and changes of sector that occurred during year N, the sample Being drawn at 31/12 of year N-1. As part of a project to revise turnover indices, solutions have been implemented to correct this bias and produce better sales turnover indices. On the one hand, it is necessary to free oneself from the annual samples by exploiting the totality of the monthly VAT returns and, on the other hand, to rely on the Sirus statistical register in place of the Sirene administrative register, to benefit from a much faster and automated provision of the most recent demographic events affecting businesses.

Revenue indices in services are monthly indexes in value (in current euros excluding tax). These indices are calculated in reference 100 in 2010 (the indexes are for average 100 in 2010). They are calculated as Laspeyres indices, with constant weights over the time of the finest indices, and are re-plated every 5 years.

The constant weightings of the basic indices are the sales of branches in 2010 at basic prices. By definition, productions at basic prices exclude taxes on products but are increased by subsidies on products; They do not include transport services. The weights were calculated from national accounts data.

The series are corrected for seasonal adjustments (SA) and calendar effects (working day adjustments WDA). An estimate of these effects is produced with the X12-Arima software in JDemetra+ (supplied by Eurostat) at the classes level of NAF rev. 2. SA-WDA indices for higher levels are obtained by aggregating the SA-WDA indices of the classes that make up the level.
The annual mean of the SA-WDA indices can slightly differ from that of the raw indices because it takes into account variations from one year to the next in the annual composition in working days: presence of leap years, position in the week of various public holidays, and so on.