BY JESS BARDIN – SR. ANALYTICS DEVELOPER
Since I predominately work from home, I find myself working from a coffee shop whenever I need to get out of the house. For anyone who spends any time in coffee shops around this time of the year, the inundation of pumpkin spice lattes is unavoidable. I think it’s safe to say that it comes standard with the season.
Wouldn’t it be nice if we could predict what customers were going to buy as easily as we could predict when the scent of pumpkin spice lattes would fill the streets? Unfortunately, customers aren’t always driven by seasons or holidays. Sometimes a customer’s purchase pattern is specific to only them.
Let’s take a look at a few data points from my transaction history at our neighborhood grocery store over the past few years.
It’s a relatively sporadic stream of transactions, but do you see any year-over-year patterns?
Ok, I’ll make it easier for you by highlighting transactions that repeat every year within a few days of each other.
See the pattern now? So why was I shopping at this grocery store during the first week of August the last three years? My wife typically does our grocery shopping, so my trips to the grocery store are normally pretty random. There must be something specific about early August. Let’s take a closer look at what I’m actually buying:
SKU #351650000078 is a jumbo size Toblerone candy bar, which happens to be my wife’s favorite candy. SKU #996476000452 is an arrangement of hydrangeas, which are my wife’s favorite flowers. At this point, it’s pretty safe to assume I’m buying her presents. As it turns out, our wedding anniversary just happens to be August 6th.
Let’s say we applied the same approach to all customer data for this grocery store. We’d certainly find many other “anniversary” shoppers for things like birthdays, wedding anniversaries, and other special occasions. We could do the same thing for other types of businesses, too. For example, if we looked at a department store’s customer data, we might identify a customer that buys their mother something from the store every year around the last week of February, because their mother’s birthday is February 26th and she can always find a petite dress she likes at that department store.
Why do we care? Because we can anticipate a customer’s need for something before they show up at the store. We can remind them, and incentivize them to come back, so they don’t end up buying what they need somewhere else this year. It’s also an opportunity to show the customer that we’re paying attention and we’re here to help.
Now, am I proposing that our neighborhood grocery store can dramatically increase their profits by ensuring that I return every year to buy my wife a jumbo Toblerone and an arrangement of hydrangeas? No. I’m suggesting that it’s a pattern of behavior that can be identified from customer data, and used to communicate with those customers when they’re very likely to listen. In my case specifically, it’s a great opportunity to let me know that that they’ve recently started a MyFlowers loyalty program, for example.
So as I sit here in this coffee shop, smelling the pumpkin-spiced scents of the season, I can’t help but wonder what other shopping patterns I have made. Better yet, I wonder if those patterns have been discovered and I’ve been steered toward repeating those patterns this entire time!
Timing is Everything
BY JESS BARDIN – SR. ANALYTICS DEVELOPER
Since I predominately work from home, I find myself working from a coffee shop whenever I need to get out of the house. For anyone who spends any time in coffee shops around this time of the year, the inundation of pumpkin spice lattes is unavoidable. I think it’s safe to say that it comes standard with the season.
Wouldn’t it be nice if we could predict what customers were going to buy as easily as we could predict when the scent of pumpkin spice lattes would fill the streets? Unfortunately, customers aren’t always driven by seasons or holidays. Sometimes a customer’s purchase pattern is specific to only them.
Let’s take a look at a few data points from my transaction history at our neighborhood grocery store over the past few years.
It’s a relatively sporadic stream of transactions, but do you see any year-over-year patterns?
Ok, I’ll make it easier for you by highlighting transactions that repeat every year within a few days of each other.
See the pattern now? So why was I shopping at this grocery store during the first week of August the last three years? My wife typically does our grocery shopping, so my trips to the grocery store are normally pretty random. There must be something specific about early August. Let’s take a closer look at what I’m actually buying:
SKU #351650000078 is a jumbo size Toblerone candy bar, which happens to be my wife’s favorite candy. SKU #996476000452 is an arrangement of hydrangeas, which are my wife’s favorite flowers. At this point, it’s pretty safe to assume I’m buying her presents. As it turns out, our wedding anniversary just happens to be August 6th.
Let’s say we applied the same approach to all customer data for this grocery store. We’d certainly find many other “anniversary” shoppers for things like birthdays, wedding anniversaries, and other special occasions. We could do the same thing for other types of businesses, too. For example, if we looked at a department store’s customer data, we might identify a customer that buys their mother something from the store every year around the last week of February, because their mother’s birthday is February 26th and she can always find a petite dress she likes at that department store.
Why do we care? Because we can anticipate a customer’s need for something before they show up at the store. We can remind them, and incentivize them to come back, so they don’t end up buying what they need somewhere else this year. It’s also an opportunity to show the customer that we’re paying attention and we’re here to help.
Now, am I proposing that our neighborhood grocery store can dramatically increase their profits by ensuring that I return every year to buy my wife a jumbo Toblerone and an arrangement of hydrangeas? No. I’m suggesting that it’s a pattern of behavior that can be identified from customer data, and used to communicate with those customers when they’re very likely to listen. In my case specifically, it’s a great opportunity to let me know that that they’ve recently started a MyFlowers loyalty program, for example.
So as I sit here in this coffee shop, smelling the pumpkin-spiced scents of the season, I can’t help but wonder what other shopping patterns I have made. Better yet, I wonder if those patterns have been discovered and I’ve been steered toward repeating those patterns this entire time!
Tags: