Retailers must earn their stripes with fickle shoppers every single day. In the viral world of Facebook, Twitter and text messaging this is especially true, but word-of-mouth communication has always been the top influencer on shopper behavior. This is true today just as it was in the 1880s, when the following 10 oldest U.S. retailers were already making their marks on the retail scene.

What we currently call “smart retailing” typically refers to improving processes or systems. We lean out the supply chain, we respond to real-time data, we match labor to demand. But automated information processors, aka computers, can do much more. When computing power converges with retail logic and large databases, “smart” systems evolve into “thinking” systems, and they tap into the vast potential of advanced and predictive analytics.

The fact that retailing employs more workers than any other industry is the worst kept secret in business. The actual figure is 25 million workers, or one in every five jobs in the U.S., according to the NRF. But with an elephant in the room like that, how is it that most retailers have not yet maximized their labor efficiency nor optimized their workforce management tools? Here’s why and how to solve it.

Using analytics to review past performance is no longer enough to be competitive. Retailers need to shift from showing what happened (reporting) and what is happening now (alerts) to what will happen. They need to progress from getting explanations to predicting outcomes. Not long ago this level of complexity was beyond the capability of analytics software, but today predictive analytics solutions are maturing and becoming powerful engines in the hands of the select retailers that use them.