The proliferation of digital traces generated by the activity of individuals or companies and the growing capacity to store and analyse them have given rise to what is known as the “Big Data” phenomenon. Using large quantities of individual data is obviously nothing new for official statistics, which already process data fromsurveys, censuses and a wide variety of administrative sources. However, the advent of big data does introduce two major shifts in the form of much larger volumes and almost-immediate access. There are a number of obstacles, nonetheless, to making the most of these advantages, as these data are themselves not always without their defects: they come in complex or unstructured formats, using them can require costly investments in specific technologies, they are not always guaranteed to be representative, and the same applies to their predictive power.

This focus proposes a review of the contribution of these data to three aspects of the measurement of the economy. First of all, short-term monitoring: does analysis of internet search behaviour or of the online press enable us to anticipate the short-term economic climate more effectively than survey data? As things stand, the answer to this question is still somewhat guarded.

The second aspect is tracking prices. In this respect, the contribution of big data is already shown to be muchmore tangible, whether in the form of prices collected online or scanner data provided by retail chains. Finally, we present some attempts to use big data to better capture the general phenomenon they are stemming from, the boom in the digital economy sector.