We couldn’t agree on the best dim sum, so we dug into the receipt data
Illustration by: Sensibill
We like food a lot at Sensibill. We talk about it a lot. We consume it a lot. We have a total of eight Slack channels dedicated to different foods. It's an occupational hazard that sometimes we look through our receipts and try to one-up each other...
This cultural tenet probably has something to do with our CEO, Corey Gross. He can advise on a lot of things—business, banking, technology—but what he loves to opine about is food. Ordering lunch? Going for dinner? Going to Rome? Ask Corey where to eat.
So when he insisted that Rosewood had the best dim sum in the area, we wanted to believe him. But not everybody was sold. To settle it, we decided to dig into our employees’ receipt data.*
The proof is in the pudding
We have a winner! 27% of receipts came from Rol San and only 9% came from Rosewood... Sorry, Corey. In terms of favourite dish, 50% of people ordered pork sui mai.
While this proved that Rol San is the most popular dim sum spot among our employees, it got us thinking about what other behaviours we could uncover from restaurant receipt data.
Did the day of the week have an effect on the types of dishes people ordered? Or the number of coffees they had? Or how they pay? Does everyone order McFlurries on Fridays? Asking for a friend.
Eating for the weekend
Something about Friday nights must make us hungry... People ordered 2X the amount of food items on Friday and Saturday versus Monday to Thursday.
People ordered 2X the amount of food items on Friday and Saturday versus Monday to Thursday. Animation by: Sensibill
Payment preferences
Cash is decidedly not king at Sensibill, and credit cards came out way ahead. This could possibly be because we're big fans of our fellow Toronto startup, Ritual. 1.7% of our spending was on loyalty cards—all of which was spent on coffee (hi Starbucks!).
70% - credit, 15% - debit, 3.3% - cash, 1.7% - loyalty cards (all for coffee). Animation by: Sensibill
Cheers to midweek beers
Another finding that surprised us: 45% of alcoholic drink receipts were from Wednesday nights. What can we say? Hump days are a thing.
Wednesday was the most common day to have a drink. Animation by: Sensibill
And what's in our cups? 70% of people ordered beer, 20% ordered cocktails, and only 10% ordered wine.
70% of people ordered beer, 20% ordered cocktails, and only 10% ordered wine. Animation by: Sensibill
The caffeination situation
When it came to our morning shot of caffeine, we kept it classic with 32% of coffee receipts coming from Starbucks.
32% of coffee receipts are from Starbucks. Animation by: Sensibill
The most popular beverage was a Pike Place drip coffee (ordered 32% of the time), followed by a latte (ordered 18% of the time).
The most popular beverage was a Pike drip coffee (ordered 32% of the time), followed by a latte (ordered 18% of the time). Animation by: Sensibill
But, interestingly, 33% of Starbucks orders were actually food items. Of these food items, 64% were desserts or sweet treats. Strawberry cake pops were the most popular order.
33% of orders from Starbucks were food items. Of these food items, 64% were desserts or sweet treats. Strawberry cake pops were the most popular order. Animation by: Sensibill
TGI fry-day
As we suspected, weekends were unhealthier than the rest of the week. We were more likely to order fast food on Fridays and Sundays than on any other day.
People were more likely to order fast food on Fridays and Sundays than on any other day. Animation by: Sensibill
37% of fast food orders were cheeseburgers, and 18% were french fries. Like good Torontonians, The High Priest burger from our local Burger's Priest was the most popular cheeseburger.
37% of fast food orders were cheeseburgers and 18% of fast food items ordered were french fries. Animation by: Sensibill
Most popular dishes by cuisine
We also broke down the most common cuisines to order and the most common dishes:
1. Japanese - California rolls
2. Mexican - Fish tacos
3. Italian - Cheese pizza
California rolls, fish tacos, cheese pizza were always on the table. Animation by: Sensibill
Unlocking the potential of receipt data
Ultimately, the objective of this exercise wasn’t to prove Corey wrong. We wanted to explore some of the ways that receipt data could explain and predict human behaviour—in this case, eating habits—with more accuracy.
For banks, those insights could translate into identifying customers' life events, finding out which products to recommend to them, and providing relevant financial wellness advice.
Find out how your bank could leverage receipt data for better personalization.
*Note: Information privacy is extremely important to us, and we’re hyper-diligent about how we collect, handle, and process data. Our philosophy? Data for data’s sake is creepy. Our employees have explicitly consented to having their receipt data analyzed for the purposes of this post. For more information about our company’s privacy policy, visit our privacy policy page.