Improved the beginning of the README a bit

README.md

@@ -2,12 +2,17 @@
 
 ## Overview
 
-influx-spout consists of a number of components which receive [InfluxDB] [line
-protocol] measurements sent by agents such as [Telegraf], efficiently filter
-them and then forward them on to one or more InfluxDB backends. Much effort has
-been put in to ensuring that high volumes of measurements can be
-supported. [NATS] is used for messaging between the various influx-spout
-components.
+influx-spout is an open source messaging system that routes and processes 
+[InfluxDB] [line protocol] metrics from agents (For example [Telegraf]) to 
+processing and storage backends (InfluxDB, Kapacitor etc.)
+
+Key features:
+- Proven ability to handle high volumes of data (>500k points per second)
+- Horizontal scalability with a multithreaded data processing pipeline and multi-node distributed architecture based on the [NATS] messaging system
+- Ability to add and remove endpoints without disrupting existing data flows
+- Fine-grained control over routing of metrics data to specific destinations based on measurement names and, optionally, regular expressions
+- Sanity checking that prevents corrupt metrics data from reaching storage and processing backends
+- Batching of egress data to large chunks, making it easier for backends to process
 
 [InfluxDB]: https://www.influxdata.com/time-series-platform/influxdb/
 [line protocol]: https://docs.influxdata.com/influxdb/v1.4/write_protocols/line_protocol_tutorial/
@@ -53,7 +58,6 @@ All the influx-spout components may be run on a single host or may be
 spread across multiple hosts depending on scaling and operational
 requirements.
 
-
 ## Building
 
 Ensure the `GOPATH` environment is properly exported and simply run