Gabby Jones/Bloomberg/Getty Images
Maybe you're actually special, and the perks you get with Starbucks drinks may not be available to all customers.
new york
CNN
—
It's hard not to feel angry when you think about how much more expensive everything has become compared to just a few years ago, but the people who make the exact same purchases as you don't necessarily pay the exact same prices as you.
This became clear to me a few weeks ago when a friend texted me to let me know that Starbucks was running a buy-one-get-one-free promotion, but when I logged into the app, the offer was nowhere to be found.
Why was my friend being treated differently?
“Starbucks likely used artificial intelligence to determine that my friend would buy an item if offered a promotion that she otherwise wouldn't, but that I would make a purchase with or without a promotion,” said Shikha Jain, lead partner in North America consumer and retail at consultancy Simon-Kucher.
The system worked well for me, I simply opened the app to see if there was a promotion, was able to place the order, and pay full price.
She said it's impossible to say exactly how Starbucks' system made the right decision about me, but she was pretty sure my past shopping behavior and that of similar customers played a factor.
The Seattle-based coffee chain declined to disclose what data goes into its AI model, called Deep Brew, but a spokesperson confirmed that it uses AI to personalize offers the company sends to customers.
This personalized pricing strategy isn't unique to Starbucks: Companies are increasingly using customer data, often from loyalty programs, in conjunction with machine learning models to set unique prices for goods and services based on an individual's willingness to pay.
The end goal, Jain said, is to “get them to buy more of the product, buy the same product again, or spend more on the same product.”
Companies still employ tried-and-true marketing tactics like segmenting customers with similar buying behaviors and preferences, taking into account shoppers' location, seasons and more.
Companies have used this information to effectively offer different prices to different consumer groups, but in the past, they had to cast a pretty wide net to do so, says Jim Pressley, senior vice president of U.S. consumer analytics at consumer insights company NielsenIQ. That might mean mailing coupons to homes in certain areas or offering specials tailored to people who click on certain types of ads or read certain online publications.
For years, companies have tried to target customers by distilling statistics into actual preferences.
What's changed now with advances in AI is the level of sophistication and accuracy with which these types of predictions can be made, Pressley said.
Using AI, companies can answer questions like, “What is this person going to buy next? How much are they willing to pay? Where are they going to buy it from? When are they going to buy it?” Jain says.
Matt Pavich, senior director of strategy and innovation at Revionics, an AI company that specializes in helping retailers set prices, said the company's goal isn't to tell retailers exactly how much to charge each customer. Rather, its main job is to provide “all the analytics and predictive scenarios” to help companies determine prices themselves.
David Paul Morris/Bloomberg/Getty Images
Tractor Supply recently announced a partnership with Revionics, an AI company that helps retailers set prices.
Instead of waiting for customers to react to price changes in real time, Revionics' clients get a toolbox to test prices in advance, and by predicting how much consumers will buy at different price points, Revionics helps retailers manage inventory.
Personalization doesn't just stop at pricing, it also applies to customer targeting: if you receive a notification about a sale, for example, the wording of that notification may be completely different from the wording of a notification another customer receives about the same sale.
Mary Wynne Pilkington, senior vice president of investor relations and public affairs at Tractor Supply Co., told CNN that the company recently partnered with Revionics because it wanted to more effectively align prices with an “ever-changing marketplace” and “attract and retain customers.”
The goal of partnering with Revionics, she says, isn't to see how far they can raise prices without losing too many customers, but rather to “curate individual offers for each customer,” which she says “often results in lower prices and higher value for the products and services that customers need.”
Of course, as in my experience with Starbucks, it can also lead to identifying customers who don’t need promotions at all.
