Along with Microsoft's announcement for new service tiers, the company introduced performance levels that lets users easily scale up or down depending on their needs.
According to Microsoft, performance levels provide predictable performance across all levels, resulting in a more consistent and reliable experience.
Resources for each tier and performance level are expressed in Database Throughput Units (DTUs), which measure the capacity of a performance level on a mixture of CPU, memory, and read and write rates.
You can also now monitor your multiple SQL Database resource usages through an API and the Azure Management Portal through several different metrics, including CPU, reads and writes, and memory. Notably, users can track their performance usage relative to the available DTU resources within your chosen service tier level.
Once you've got a feel for how much resources you're eating up, then you can also dynamically adjust your service tier depending on your needs. If you're finding that you need more resources, then you can move up a service tier as your application becomes popular. Alternatively, if you're swimming in extra resources, it's just as easy to move down to a different service tier.
Also see the MSDN article, "Azure SQL Database Service Tiers and Performance Levels" for more information.
