Predictive marketing may sound like a new thing, but if you ask any retail buyer, they’ll tell you they’ve been doing it since fashion became a thing. The definition of a fashion buyer’s job is to figure out what customers will want next season–and in what quantity.
So what’s new?
Big data changes the game. Predictive analytics, the science of interpreting data to make informed marketing decisions, is on the rise–using data to figure out what you want before you know you want it.
If you think it sounds like magic, you’re not far off. Magic works precisely because magicians know exactly where the audience will look, or what number they are likely to “guess.” They may not call it predictive modeling, but that’s what it is.
How can predictive analytics improve your marketing?
Your customer data may be more valuable than you think. One obvious example: A woman who buys maternity clothes now is going to be shopping for tiny clothes and assorted baby paraphernalia in a few months. But that’s only the tip of the iceberg.
The overwhelming amount of data available today makes it possible to consider every angle to tailor offers with terrifying accuracy, and that’s not a bad thing.
Predictive analytics can look at buying habits of one customer and all customers, day of the week and time of day when people like your customer are likely to buy and the buying habits of a specific customer. It can factor in data from trends, seasonal information, update age or stage of life data and make incredibly sophisticated recommendations.
While it may sound creepy, in effect, predictive data analysis delivers ads that won’t annoy consumers. Young people who like to mountain bike will get offers for bike accessories, eco-friendly water bottles, the latest in spandex fashion and anti-chafing spray. Cat ladies will see offers for the latest miracle litter boxes, catnip toys and, presumably, Roombas and shark costumes, instead of random ads for adult diapers, Viagra or weight loss supplements.
Creating ‘a segment of one’
I read that phrase in a Marketing Week article and loved it so much I had to borrow it. In the post, Lucy Fisher quotes Ben Kay, a senior consultant in analytics at IBM:
“We’re moving from reactive social listening to looking at the ‘why’ behind it,” he says, claiming that effective analysis of social data can uncover personality traits, which, from a marketing perspective, can enable brands to tailor communications “to a segment of one.”
It’s the perfect description. Using predictive marketing and the massive array of information available, marketers today can predict with eerie accuracy exactly what I want. Not people in my general demographics group or people who live in my area, or even people who work the same job or have similar hobbies: Specifically me. What’s more, marketers can predict what I will want before I know I want it.
Know before they know …
Always ahead of the game, in 2014, Amazon took steps to ship goods before you order them. Call it anticipatory shipping. The theory is that most people will keep and pay for things they receive. I’m not sure how that will work out, but the lead-up is genius. In order to make this kind of thing happen, Amazon analyzes the buying habits in an area and ships the products it anticipates people will want to local warehouses, which facilitates next-day, or even same-day delivery.
If the prediction is wrong and the doodads don’t sell as quickly as anticipated, no problem. Time for a sale–or even a customer appreciation giveaway.
How accurate are the predictions?
In a study released in March, researchers from Ohio State University discovered that targeting ads based on behavioral cues improved click-through rates by up to 670 percent over ads not behaviorally targeted. People worry about their data being misused, but they like and respond to targeted marketing.