How AppsFlyer manages scale without sacrificing performance

AppsFlyer with Firebolt


Listen on Spotify, Apple Podcasts or watch on Youtube.

Recently on The Data Engineering Show, we had the pleasure of speaking with Alexandra Sudilovski, Senior BI Expert & Looker Guild Master at AppsFlyer. For those who don’t know, Appsflyer is a leading SaaS mobile marketing analytics and attribution platform.

Whenever you download and interact with an app, data is sent through their servers. As you can imagine, this amounts to an enormous amount of data.

Appsflyer processes 120 billion events daily and saves 90 terabytes of data in AWS and 40 petabytes in BigQuery daily — and they’re still growing.

Appsflyer has exploded in size, growing from a small company of 200 people to 1000 people in just three years. They went from nothing to having thousands of users and hundreds of thousands of dashboards in a matter of a couple weeks.

Dealing not only with a huge amount of data on a daily basis but doing so while growing quickly as a company can come with many challenges. Not to mention, the definition of “huge data” seems to only grow year after year.

We asked Alexandra how she and her team handle these challenges:

We’ve learned that you need to adapt quickly and be ready to change your mindset and approach. After we went into production, we had all these people who were hungry for more data, and it became hard to keep up with everything from delivery of new projects, system maintenance, and even bringing on the new employees that we needed.

But at Appsflyer, every single department and internal process is measured — and not just the R&D side of things but also functions like HR, marketing, legal, finance, and more. We want to see everyday how we work as a company and if there are any processes we can improve. We constantly measure ourselves and are improving, and we have seen great results from this.

We also use every code that is known on the market to be able to adapt and be equipped to always use the best solution in any situation.

With so much data to manage, it would be easy for mistakes to fall between the cracks. But Appsflyer takes data accuracy to a whole new level — which is one of the reasons we love this company so much. Alexandra is a leading force behind this drive and mindset:

If I saw data that was not accurate and not showing the truth, this would make me not believe in my own product. If I don’t believe in it, why should others? So my priority is always data accuracy and making sure people can trust us and our product because people rely on that data to make decisions.

At Appsflyer, we constantly measure ourselves and double-check our work. There are always multiple pairs of eyes on everything, and we have people dedicated to testing and QA in R&D. We want to catch any errors immediately instead of waiting to get this feedback from stakeholders and users.

Alexandra is truly an amazing leader in the BI space, and of course, Appslyer is a company that we believe in and love to follow. We talked to Alexandra about so much more, including how she has blended the roles of BIand data engineering in her work and how she uses some of the hottest tools in the industry today, so be sure to check out the full episode with her.  

Read all the posts

Intrigued? Want to read some more?

Curious to learn more?

Discover how Firebolt achieves extreme performance at a fraction of the cost