DEV Community

Cover image for How AWS helps Domino’s to deliver their pizzas in record timings?
Aditya Gaur for AWS Community Builders

Posted on • Originally published at aadigaur.Medium

How AWS helps Domino’s to deliver their pizzas in record timings?

You excitedly opened the box containing your delicious Double Cheese Margherita loaded with extra cheese. Sprinkled oregano and chili flakes over it, made a smiley with the tomato sauce. You were just about to put the first bite of your Double Cheese Margherita and your date interrupted in between and asked you to click a photograph first!

Doesn’t sounds like the ad-interruption of Spotify?

We are so much surrounded by technology,

Oops!

Let me say it again.

Saying ‘surrounded’ would be like looking at the tip of an iceberg.

We are so much submerged into the technology that sometimes it makes us compromise with our special moments.

I haven’t been to Domino’s since the lockdown, neither my house falls within their delivery range, so yes, I’m very much missing hanging out with my friends and enjoying those cheesy creamy pizzas.

What about you?

Well, talking about Pizzas, or to be more precise Domino’s, have you ever thought about how are they able to deliver your desired pizza at your doorstep in record timings?

How it feels like waiting for your food with the cravings brimming out of your mind and unwanted sounds coming out of your stomach?

Along with these things, some people have a slight intention of getting their order delayed so that they can enjoy their food free of cost.

So, which category do you belong to?

Even the 30 minutes sounds a great deal of time to an empty stomach.

Well, here’s good news for those who love to enjoy their meals at their homes. Domino’s aims to reduce this delivery time to 10 minutes or less under a collaborative effort with AWS under the name Project 3TEN.

Why this project is named 3TEN?

Any guesses?

No worries, let me break it down for you.

With this initiative, Dominos aims to have an ordered pizza ready within 3 minutes or safely delivered within 10 minutes to the customer. That’s ‘3TEN’.

Why Dominos is focusing so much on reducing the delivery time in spite of having so many outlets already?

This may sound crazy to you, but 70% of sales of Domino’s come solely from online orders!

They’ve already started to deploy the 3TEN solutions to stores located in New Zealand, France, the Netherlands, Japan, and Germany. Now, this brings hope for the people residing in the rest of the world too. Upon asking what’s the key ingredient in Dominos’s growth as a business, ‘Investment in technology’ was the reply of Michael Gillespie, Chief Digital and Technology Officer for Domino’s.

How Domino’s taking the leverage of technology in enhancing their delivery solutions?

Here comes the entry of AWS. With the help of AWS and an Advanced Consulting Partner from AWS Partner Network, Domino’s wished to create a predictive ordering solution. Doesn’t it sound astonishing that they would be knowing already what are you going to order even before you order! That’s what Gillespie saw in AWS machine learning technologies.

In order to build any predictive solution, one must have a huge amount of data relevant to the field in which the predictions are going to be made. That’s how machine learning works. You need data to get the desired data. Hence the first task for Domino’s was to accumulate the key information of their orders at a single place. For this, they took the advantage of Amazon Simple Storage Service (S3) to create a data lake of their own.

As the name suggests, S3 is just a simple storage service, and instead of having a traditional hierarchical structure for storing files, it has a flat structure of containers known as buckets in which the files are stored. You can’t just make complicated queries to get your desired result. Each file has its unique Key ID and URLs are used to access the contents in the bucket.

The next task was to perform analytics on the data stored in S3, that can’t be done directly. In this scenario, AWS Glue was the best option to use in between. To keep it simple, AWS Glue is a fully managed extract, transform, and load (ETL) service that makes it easy to prepare and load your data for analytics.

So, what next?

The next task was to predict the probability that an order will be placed. Glue along with Amazon SageMaker is capable of performing this job without any hustle. They had to just build and train the machine learning models for the same so that your pizza can visit the oven even before you place the order.

This was possible not only through AWS, but Dominos had to look at each and every factor that contributed to reducing the delivery time. This includes enhancing the cooking methods and transportation techniques and opening more stores in more proximity to the customers.

So next time when you’ll be having your Domino’s meal, don’t forget to say thanks to the technology that made it possible to deliver it in much less time. And to everyone who’s playing an immense role in this. Farmers, Cooks, Delivery Men, and other workers too, have their equal contribution!

References:

Fig. 1: https://d1.awsstatic.com/case-studies/Dominos_031019__1830.858a45d28fb2879b938cec7e0961767fe7a8e695.jpg
https://aws.amazon.com/solutions/case-studies/dominos-case-study/

Discussion (0)