Applying Affinio to Custom Datasets
Over the past few years, Affinio users have run thousands of analyses to further understand their audience. Leveraging the interest graph through social data queries has proved to be a powerful way to discover rich insights about people.
Recognizing the growing amount of data available today and the desire to better understand consumers, we’ve now made it possible to build audiences through custom data sources and use Affinio’s machine learning algorithms to cluster and gain insight to reason about them.
To do this, we’ve spent time modularizing and expanding the types of data that Affinio knows how to make sense of. To understand your audiences beyond the datasets provided, you can pair your custom data with insights from social networks to help make your dataset greater than the sum of its parts.
We’ve experimented with private and public datasets, with survey data, and with anonymized user data to draw conclusions about the things that make people distinct and the things that people have in common.
As an example of Affinio’s ability to analyze and understand custom multivariate datasets, we created an internal survey to collect some fun information about the Affinio team (affectionately named, Affinians). The following are examples of the clustered data and the insights we uncovered.
The clusters that exist within the Affinio team:
For this simple survey, we asked Affinians what music they have listened to most recently. As you can see below, Drake has fans across most teams at Affinio, while Kanye West only appears in the EBR, Sales, and Customer Success teams.
Learning more, we see that most of Affinio’s team is located in the Halifax, NS office (Affinio HQ). The most prevalent age is 24 and August, June, and December hold the most Affinian birthdays. Protein appears to be what fuels the team here, with Eggs and Breakfast Bars as the most popular breakfast choices.
If we want to understand what makes two teams similar, and what makes them different, we can leverage the Affinio cluster compare functionality. While footwear preference is what sets these teams apart, their love of Spotify is what brings them together.
As you can see, familiar elements from Affinio reports run on social networks like Twitter present themselves in our custom data reports. We believe this will give users the opportunity to explore their custom datasets in ways they haven’t been able to before: with Affinio’s data-driven insights at their fingertips.
Want to use Affinio to build an audience on your custom data? Let’s get you started today.