Octopus Deploy is for pushing your built executables to a cluster of Windows machines. Once your build server is done with a new build, you can notify Octopus with its API or keep it looking for changes in a folder, and as soon as it is notified, the built objects are transferred and deployed (including setting up servers and configuring machines) to those remote machines in parallel. Doing this manually is a horror for DevOps and the products that existed before it (like those made by Microsoft as part of VS itself) were either too integrated with other products or way too simple for complex architectures.
1. Push new releases automatically or manual approval - Permissions model for deployment is top-notch here - Pushes can be automated fully if you are very confident about its reliability, or a person can be in charge of authorizing it.
2. Cluster Management - Arguably, the most important part of the deployment is the configuring the class of servers according to their role in your IT architecture and must be done beforehand. Its possible to assign groups, sub-groups and labels to different machines and configure them in batch mode. The visual management options is much better than the older command-line configurations.
3. Documentation - What Octopus does is itself not too complex, but it is all about how good it does the job. Its UI acts as active guide for the Deployment Engineer and the icing on the cake is the nice documentation. Both its UI options and API are explained well in the docs and is welcoming to new users.
I have had a few problems while on the beta build -
Installation failed initially due to missing DLL on our Windows 2008 server but after changing it to Octopus username and password (as suggested by their support) fixed the problem. But it still didn't work, showing an internal error message which I couldn't make much out of. At this point, their support just dropped the case asking us to revert back to the primary build, which obviously worked fine.
We handle a few .Net projects one for our clients, that need to be deployed across the country on their remote servers few times a year for maintenance purposes and this is a gigantic task which needs to be completed within two days with as less downtime as possible. All the servers need to be on the same version at the same time to make sure no intermittent bugs crop up and Octopus helped us make the switch live on most machines. We still micro-manage a few critical boxes but other things are quite automated.