The difference between recommendations and predictions - and why it matters for marketers

Advertisement

Girl Shopping Sailthru

Getty

Predicting what customers will buy can be a powerful tool.

Most people understand how powerful online recommendations can be. You're shopping around for a lamp, and based on what you've already browsed or bought, the website recommends other related products you might like as well, such as lightbulbs. But predictions are just as important - and marketers need to know the difference.

Predictions are based on data models rather than logic. They leverage large volumes of data and allow marketers to know what you're going to do next - such as open email, purchase, or even opt out - so that they can control what messages you receive, predicting how you'll then act.

While both predictions and recommendations can influence buying habits in very different ways, too many marketers confuse the two or think they're the same thing, which can end up turning investments in technology into losses rather than gains. Even if a brand is using both recommendations and predictions, if they're not used together in the right way, it can cost companies money.

SailthruSightlinesPredictions_graphics4

Sailthru

If your data predicts that a customer will spend $123, don't just recommend an $80 sweater - you'll be leaving money on the table.

Case in point: recommending products a consumer may purchase, but overlooking the average order value (AOV) for a customer. (See illustration to the right for an example.)

The whole point behind both recommendations and predictions is that they focus on retaining existing customers. That's critical in this business climate because it's far more expensive to attract new customers than keep existing ones.

A report from Gartner shows that 80% of a company's future revenue will come from 20% of existing customers. Bain & Company also released research that shows a 5% increase in retention can boost profitability by 75%. So marketers need to focus on keeping customers by offering them new opportunities to buy - and not just that, but the right opportunities when they are individually ready to engage and buy.

Predictive intelligence is a crucial tool for today's marketers. By choosing the most meaningful predictions and strategically combining them with recommendations, they can prevent customers from opting out, encourage others to spend more, and build the long-term value of their brands.

Click here to download "The Definitive Guide to Predictive Marketing."

This post is sponsored by Sailthru.

Find out more about Sponsored Content.

Follow BI Studios on Twitter and Facebook.