DoorDash Data Analysis in Excel

We all have to eat. Yes, it's cheaper to cook and eat at home or to get in the car and pick up your own food. But that's not always the best option. Sometimes, convenience is king.

With my experience in marketing, I'm always wondering why people make purchases, and why they don't. Data about purchasing decisions can reveal trends, uncover customer motivation, and guide future marketing campaigns.

When I saw this data set from iFood, the Brazilian equivalent of DoorDash. I had some questions...

🤔 We're in the midst of the busy holiday season, a time when ordering out would make sense. But do more people start getting food deliveries during the holidays, or in the summer when it's too hot to cook?

🤔 Lot's of people cut corners to save money. But we all still have to eat. Do people with more money order out more often than people with lower incomes?

🤔 Once you start ordering food, it's easy to do it again and again. What's the average amount that individual customers spend on food orders over time?

I dug into this data set so I could:

✔ See the big picture of the data

✔ Zoom in and discover business opportunities & insights

✔ Formulate data-driven strategies to optimize the results of future marketing campaigns & generate value for the company

Data Set Info

In my work as both a teacher and copywriter/marketing strategist, I've seen first-hand that accurate data analysis is pivotal in long-term results. And as I've dug deeper into data analytics, I've realized how much I love it.

So I decided it was time to level up in my data analytics skills. So I signed up for a data analysis boot camp.

My boot camp instructor Avery Smith shared this data, derived from the Brazilian food delivery company iFood, for a course project.

This data set includes customer information like marital status, income, number of children, the types of food they order, and the money they've spent at iFood over the course of a year. I looked at the data set as a whole, but I was also able to filter the data based on specific advertising campaigns.

The Juicy & Delicious Results

When do people join iFood?

I wanted to see trends in when people first joined iFood. So I started with a bar chart:

Unfiltered data reflecting new customers by month

I'd hypothesized that summer might see an uptick in new iFood customers. And sure enough, those hotter summer months saw more new customers. Keep in mind that Brazil is in the southern hemisphere, so summer runs from December to March.

I decided to zoom in and look at the data from one specific marketing campaign.

Campaign 6 new customers by month

Campaign 6 clearly peaked in January as well. But the summer trend didn't show up as obviously.

This raised questions about how the campaign was run.

🤔 When did Campaign 6 begin?

🤔 How much A/B testing was done with ad frequency, target audience, and copy (wording)?

Unfortunately, I don't have access to that data so my questions remain unanswered...

How does income affect ordering?

To answer this question, I created a scatterplot that compared customer income with the amount spent on iFood deliveries.

Customer income and amount spent on food deliveries

As I expected, a clear correlation between income and spending emerged. I was able to draw a line that showed the trend. The higher the income, the more a customer tends to spend on food deliveries.

While this finding aligns with common sense, it would be worth looking at the campaign target audience and making sure that people with higher incomes are being targeted.

How much do people spend on iFood orders on average?

Using the data on the 2205 iFood customers, I discovered that the average person spent $562.76 over the course of one year.

And in case I wanted to find out the spending totals of specific customers, I also created a tool that returns the amount spent when I enter the customer ID number.

I also created a pivot table so I could compare customer spending based on age and the size of their family.

Next Steps

If I were to take next steps with this project, I'd analyze each campaign specifically to determine what was most effective for each age group in each month.

If you'd like to take next steps with me — a virtual coffee meet-up, a chat about data analysis and marketing strategy, or an opportunity you think I'd be a fit for — feel free to connect with me!