Though it is usually difficult to obtain data covering all of Facebook’s, LinkedIn’s or Twitter’s users, some aggregate data is provided to advertisers using these platforms. Concretely, all of the internet giants provide advertisers with information on how many of their users match certain criteria. As an example, before an ad is launched and before any cost is incurred, potential advertisers can ask Facebook “how many male Facebook users living in Massachusetts,  aged 30-50 with a PhD are interested in German cuisine?”. This information is intended to be used to estimate the required budget for running a campaign, but it can also be used as a kind of “digital census” for research purposes. In this talk, I’ll present some our work using this kind of data for projects ranging from estimating stocks of migrants around the world, to tracking global internet access gender gaps, to ongoing work on modeling urban crime rates.