That sounds pretty powerful. Do you mind sharing more information about the data storage/querying stack you use?
http://pragprog.com/book/rwdata/seven-databases-in-seven-wee...
and grep for the phrase "polyglot persistence" you'll find some great information to get you started.
a video promo for the book: https://www.youtube.com/watch?v=bSAc56YCOaE
Graphs are perhaps the only area where NoSQL may beat relational databases for expressiveness, rather than alleged scalability or ease of use. I've read the documentation for Neo4j and Cypher, but never used them in anger, so i'd be really interested to know more about how they are actually used.
In particular, i would love to be able to make a comparison between a graph database and a relational database queried using recursive common table expressions, in the context of a real-world problem.
Graph databases can scale write throughput fine when those writes are being fed to it from a single source so it's best to have a service who's sole purpose is to keep the graph database in sync with the SOR. The graph should only store the data you actually need in order to get the queries that you want.
The specific queries in my domain are exactly the kind of thing I'm not supposed to talk about, but I will say you can do things like:
Give me 50 users who who live near me (lat/long bounding box), who I'm not already following, who I have not already sent a message to, who have at least two friends who also live near me, and who have matching tags, order by number of followers.
there are a bunch of great examples in this awesome book: http://www.amazon.com/Graph-Databases-Ian-Robinson/dp/144935...
The underlying data structures involved in Neo4J do help with performance on these kinds of queries, but in addition to that I find that for me the Cypher query language feels like a more concise and elegant way to ask for the data.
With RCTE it feels like I'm first asking the DB to construct a custom data structure, then I'm asking the DB to query that custom data-structure.
With Cypher it feels like I'm simply asking for the exact data I need by specifying the relationships and attribute characteristics that I care about.
This might just be a matter of taste, but if you start playing around with Cypher it just might grow on you.