Alteryx can be used for ETL as long as you have an Alteryx Server. I've used it for a number of use-cases especially between cloud & database.
Some things that in my personal opinion make it clearly superior to SSIS:
- If input has column names (from database or from csv file with headers), it handles unexpected new columns or column order changes automatically, without requiring you to change the flows at all.
- You can build flows as "macros" which you can then unit test completely independently of your source/destination databases (try that in SSIS..)
- Ability to drop a browse tool anywhere in the flow and effectively debug.
- Build in assertions using "Test" tools.
Flows are runnable from the commandline on a server, and easiest way I've found (besides using Alteryx's own scheduler) is to save as an "App", and then run from the command line using the Alteryx engine executable, passing it parameters via xml file. You can save a sample xml parameter file from your flow by hitting the magic wand button (after saving the flow as a .yxwz (app)) This brings up a panel that lets you set the variables, and that panel has a handy "save" button which generates an xml file in the right format.
Within the flows themselves, parameterise things like environment settings either via action tools or module level parameters (User.*) - you can then for example set a database server on an input using %User.[Your variable name]% in the field.
Error logs are generally excellent (identify the tool that failed, useful error messages), and command line throws useful errorlevel numbers, so pretty trivial to schedule with some third party scheduler (or just use the Alteryx Server's own scheduler).
Obviously if you need to do any serious data manipulation, pivoting etc, then it's hands down the easiest tool I've used.