A $21B Dataset? Why Uber is eyeing Expedia

Uber is a canonical example of a data company. Their engineering blog is lore to engineers who follow their sheer scale and size. So when rumours started swirling around this week about Uber contemplating a bid to purchase Expedia Group, I was curious. It would be ignorant to exclude the fact that Uber’s current CEO, Dara Khosrowshahi (who has been at the helm since 2017), was the CEO of Expedia group from 2005 to 2017, right before he left to join Uber. Nonetheless, it is an interesting acquisition option, so I turned my finance brain back on and went digging.

So why would Uber be interested in a company like Expedia? There are a couple of reasons, which we will get into, but one of the main ones is Expedia's richness of data—all 70 petabytes of it. I could go into some of the financial reasons why this could be relevant or impactful for Uber, but this is a blog about data for data people, so let's stick to what we know.

Expedia

Expedia started as a division of Microsoft and was spun off in 1999. Then, in 2001, Barry Diller of IAC purchased a controlling interest in the company for $1.5B. In 2005, it was spun out as an independent company when current Uber CEO Dara became the CEO of Expedia. The company has done moderately well since going public. Their stock is up ~250% over the last 19 years, but it’s underperformed the S&P 500, which rose ~388% over that same period. From a financial standpoint, things have predictably been not so great since Covid. They currently trade at a $21 Billion market cap but are loaded with debt, with roughly $5.3 Billion on the balance sheet.

From a competition standpoint, the landscape is only getting tougher. Google and Bookings hold dominant market share and are increasingly encroaching into the space. The famous Barry Diller emails about Google messing with Expedia by promoting their inferior products only prove the point of internal turmoil.

It’s safe to say that Expedia survived the pandemic, but it has not been an awesome ride.

Uber

Uber, on the other hand, has been having quite a year. Its stock is up ~85% in the past 12 months, and its revenue has grown rapidly to almost $40 billion, far surpassing its scale pre-pandemic of just $13 billion. While profitability is still being figured out, the sentiment is that Uber is running away with the ride-share market and is only poised for more growth.

On the data side, Uber is known for the vast amounts of data it ingests every year. Their ML and data infrastructure teams are some of the best in the world. I mean, look at this diagram. This is what you call enterprise data infrastructure.

To highlight Uber’s sheer scale of data, here is a quote from an IBM article that takes about Presto:

The scale of Uber’s analytical endeavor requires careful selection of data platforms with high regard for limitless analytical processing. Consider the magnitude of Uber’s footprint. 1 The company operates in more than 10,000 cities with more than 18 million trips per day. To maintain analytical superiority, Uber keeps 256 petabytes of data in store and processes 35 petabytes of data every day. They support 12,000 monthly active users of analytics running more than 500,000 queries every single day.

So it’s safe to say they have an impressive but borderline insane scale of data.

Why this deal could make sense

The Data Play

As I mentioned, Uber is the textbook definition of a data company. It is reported that Expedia has roughly 70 Petabytes of data stored on its customers and their interactions. The amount of data would highlight how people search for travel, how they book, and their decision patterns around travel. These insights are precious to a company that wants to facilitate travel. Dara said that if something moves, they want to wire it up, and Expedia is where many people go to figure out where to start moving.

Now, the bear case for this is good luck trying to merge these data platforms. The technical strain and load this will take is massive and will be a multi-year process at best. However, if done correctly, it can unlock the ability to spot trends, gain deeper customer profiling, and unlock new customer segments.

The Ad Network

Uber makes roughly $1 billion from its app's ad network. Do you know how it is currently run? They have a $1 billion business run through an Asana form. I kid you not—go check it out. Expedia has a great (but smaller) ads business that is diversified and running in real time. Uber could make a data play here by merging the ad networks and increasing the value and price of the ads across their network as the targeting improves with better and more complete data.

Do I think this deal will happen? Probably not. However, I think there is maybe a 15-20% chance, which is something. There are opportunities to make this a home run, and with Uber being more aggressive, it could either be the thing to slingshot their success or bog them down. What I do know is that if this deal happens, I am going to pour a large glass of scotch to honour those engineers who need to merge these data platforms. (and I hope you will too). God Speed.

Awesome graphics from my friends at App Economy Insights if you want a side by side comparison.