Treasure Data (TD) was chosen to leverage its data integration tools, elastic computing power, and fast segmentation and activation engine.
The data from each label was easily ingested into TD using their open-source bulk data loader called Embulk.
Workflows were then developed to deduplicate and cleanse the data. Following this, the team crafted an efficient ID unification algorithm to unify users.
Lastly, we used the segmentation engine to activate segments for the client to leverage for their advertising efforts.