Omniata review


This post is part of a series of posts about analytics platforms that I would use and recommend using for game analytics. None of these posts were asked for or sponsored by any of the companies that own the platforms. My knowledge of the platforms comes from demos they made by my request or hands on experience working with them. My objective is to state what each platform offers and what makes it unique so that it helps game developers in the process of picking a platform if they choose to do so.

It is rather easy for me to speak about Omniata. It is part of Miniclip’s game analytics stack as I write this post. We’ve been using it for more than one and a half years now and it has served us well.

When we chose Omniata, we had a specific constraint we wanted to address. That constraint was the inability to create complex datasets for analysis that would answer complex questions. We wanted datasets of higher complexity without losing the ability of creating structured and organised reporting dashboards. That was our use case back then.

Since then we grew in sophistication and complexity and Omniata grew too. Although the fundamental offer didn’t change substantially, their vision of how they integrate in a complex data driven organisation is the most sophisticated I know of. Let me explain you how and why.


There are three main use cases to use Omniata for.

The first, most obvious one is analysing and visualising data. It is, in a nutshell, the reporting dashboards part. Creating and managing dashboards is extremely easy but the end results are quite powerful. The building tools are very intuitive and dashboards are feature rich from filters, to full page zoom of visualisations, download of the underlying dataset, image exporting, etc.

The second part is engaging users. This is where we setup messages to be sent to the game. These can be in game messages delivered by json (and you then decide what to do with them in game) or push notifications for mobile devices. This feature is also the home of A/B testing. You can setup your A/B tests and Omniata will deliver messages to the users of each A/B test group and “tag” those users with the information of the test and the group the user is attributed to.

Last but not least, managing user acquisition. Omniata works directly with many user acquisition partners that will feed user acquisition data directly from them to Omniata’s databases.

It’s the closest thing to having a data engineering team without having one

This is something I said about Omniata for a long time and I stand by it to this day. Behind the three use cases I mentioned is a powerful concept: create your own datasets from your own events. Instead of a restrictive, predefined set of tables that accommodate the platform, Omniata builds services around the tables that we define and a very well built user state database that enriches events as they enter the system.

You can work with their preset tables. They call them Data Applications and, to put it simply, they are a predefined set of tables and visualisations that allow you to build powerful analytics solutions effortlessly.

Like I mentioned in the first paragraphs both companies grew. In our case, we have a BI department with data engineering and data science teams. This seems to contradict the idea of having a data engineering team without having one but Omniata adapted over time. They understand that complex data driven organisations need to be able to have different data sources and data systems interacting sometimes in real time. To my knowledge Omniata is the one that better understands this complexity.


Omniata provides, out of the box, solutions for the three main current use cases: reporting, targeted messaging (including A/B testing) and user acquisition optimisation.

However its strongest asset is flexibility. In a time where many analytics build their systems in house, Omniata allows system coexistence through APIs and database access. In simple terms you don’t need to chose to have Omniata or your home brewed platform. You can have a modern mixed model… you can have both.

And in case your googling skills fail you, here’s the link.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s