Traders at a major bank noticed that it was possible to estimate NFP data with a high precision by observing how many new listings from job seekers appear on a major job search website.
Using Machine Learning tools, they investigated the relationship between the number of new job search postings and the employment data published later, and found a formula that worked very well.
It allowed them to estimate the size and direction of currency price movements a few days before the employment data was published. Rumor says that traders found this way back in 2012!
Knowing in advance what the key data might look like allows you to build a position under the expected movement, maintain an existing position or get out of it.
Some traders (but also many textbooks) recommend getting out of the markets before the publication of key data (“because the market can move hundreds of pips in any direction”). However, if we know what this data will be and what the market’s expectations of it are, then in practice we have a money-making machine.