Our service is built upon a very robust technical infrastructure that allows us to handle enormous increases in our average daily click totals (millions of clicks), while maintaining 99.99% uptime.
We host our service in multiple data centers around the world that each contain several servers, and utilize a top-end AnyCast DNS solution to provide real-time/performance monitoring of each location from around the world. This ensures that we are always serving clicks from the fastest data center to the end user, and clicks are always routed to the server with the fastest response time. Geni.us links are processed in roughly one tenth of a second before the next hop to to (affiliate network, iTunes, App Store etc.) Other similar services take nearly 3x longer. Should an issue cause a given datacenter to be unavailable, all traffic is rerouted within 30 seconds.
We utilize multiple load balanced web servers, cache servers, as well as fail-over load balancers in each region to optimize region by region availability/performance.
Our current configuration allows us to easily handle a 100% increase in our current average daily traffic (millions of clicks per day).
Finally we are totally transparent about our uptime, which can be monitored here: http://stats.pingdom.com/a8w38imsfvbm/322734. At this point in time, we have maintained 99.99% uptime world wide.
We usually ask our clients to give us a heads up if they expect a very large burst in traffic, so we can make sure to ramp up more resources if needed. This has only been needed once for a 10x increase in traffic after brining a large client online.