Why has the Affinity Segmentation method in Insights changed?
Our recent acquisition of Affinio has spurred us to elevate our Insights tool by integrating the finest features from both platforms. This enhancement is one of the ways we aim to consolidate key functionalities into a unified platform, and have therefore adopted the same algorithm and methodology as Affinio to cluster audiences based on shared interests.
One of the main benefits of this enhancement is the freedom to select the number of clusters for segmentation, ranging from 2 to 20 and allowing for extra customization, which was one of the most requested features to be improved.
What is Affinities clustering, and why is it different to Interconnections?
Affinities provides unsupervised clustering, which breaks users out into clusters based on their unbiased interest behaviors due to their following patterns. Individual users are grouped into a single best cluster based on the handles they choose to connect with, identifying their shared interests.
Once the algorithm detects a group of users with a pattern of following a handful of the same accounts, that group becomes the signal for a particular cluster.
Interconnections clustering on the other hand is based on mutual relationships, clustering individuals based on "who knows who," i.e., how these individuals are interconnected. A much more granular look at the audience and the sub-communities that represent it.
Read our full Audience Segmentation article for more information.
How can I use the new Affinities segmentation?
- This is available to all Audiense Insights users.
- Simply navigate to the clustering selection step when creating an insights report.
- Select the Affinities clusters option
- Choose the desired number of clusters between 2 and 20, or go with the recommended amount, which is based on the estimated audience size.
- Follow the usual process to launch your report!
Affinities segmentation is available via the Public Audiense API.
We value your opinion to improve our product: If you would like to provide feedback on your experience, please email productrequest@audiense.com or use our Product Portal to leave your comments.