That's the point of the testing pyramid at the end of the article. My rule of thumb is this: if you can test something at the unit testing level with mocks and feel confident about the behavior then do so. Otherwise, start moving up the pyramid (isolated integration tests > end of end integration tests > UI tests).

The higher you get the more time it takes for tests to run, the more difficult it is to identify where a failure is, the more brittle those tests will be because of inter-dependencies, and the more complex it is to write the test. The difficult part is understanding your system well enough to know where your unit tests aren't sufficient.