Crowd Riff is kind of like a Klout competitor with a focus on customer and community analysis for businesses. Seems like a perfect situation for a graph databse...
Previously using just one database, MongoDB, Crowd Riff decided to go with a graph and wide column store polyglot persistence solution. The stores they chose were Titan and Cassandra.
DataStax, the company that builds its solutions on Cassandra, interviewed the CTO of Crowd Riff, Abhinav Ajgaonkar. Here was the most interesting snippet I read from the interview:
The main difference from an Ops perspective is the sheer stability of a Cassandra cluster versus other systems that have run in a "single master" setup. Another huge bonus for us was the ability to add nodes to the cluster and have the storage, throughput and replication scale linearly.