The article is describing the ideal state of where we want to go. We’re not there yet, but making good progress.
In the end, the idea is not to avoid any intermediary deliverables, just to make sure we create them because they provide value. Value could be for the designer himself (“I want to iterate fast on lots of small variants of the same screen”), for the close team working on it (“I want to get a general sense of what landing on this page is gonna feel like, and the sketch doesn’t provide it”), for marketing purpose (“I want to prepare a newsletter to announce this new feature”).
All those are perfectly reasonable needs, and we should take them into account. The nice thing is the benefits of this system continue to pay off: easier to iterate, easier to build a page, etc.
However, the fact that we don’t for clients does eliminate things like formal signoffs.