Well, the scenarios that you should execute depend on the potential problems (risks) that may occur with the deployed release e.g.
1. A major functionality of the application does not work.
2. Portions of the application are not accessible.
3. The new features in the release are not available.
4. The application is very slow.
5. There are (obvious) defects in the deployed application.
I have simply provided some examples above. Once you identify the risks according to the business objectives of your web application, you may be in a better position to identify and re-use/ create scenarios that you should run in production. Then,
you should plan how to manage the test data (if any) that you create while testing the application in production. Would it be possible to remove or ignore the test data that is created during the post deployment verification tests? You should also plan your
tests such that any users using the application concurrently are not inconvenienced.
