We just finished work on another feature yesterday. A lot of our day to day software development planning tends to be pretty adhoc - after a few years experience following more formal estimation processes, I've been trying to encourage my teams away from up-front planning (and so towards more team ownership of the delivered work). This means that stuff can sometimes fall through the cracks, and yesterday I noticed we hadn't done any testing around what things looked like for an initial signup ... which of course was a blank gray screen.
When you're deep in the guts of a building a feature it's easy to deprioritize the "zero state" (no items). They're usually pretty easy to build - static screen, no real interaction behavior. Maybe link to another static screen with some more context about the overall product. After enough time with the same product I don't even bat an eye clicking through screens and screens of zero states - "yeah this is what it looks like when there's nothing, let me get to the screen with that bug so I can make sure I fixed it". These screens I just sort of gloss over look really different to our customers - when they see that page says "You don't have any X now", they'll be asking what is an X, and what can it do for me?
It's easy to overlook the question of "why" when you've been working on the same product for a while, but it's a question that your customers are constantly asking. Why do I need to use your product? Why do I need to install an app? What benefit does any of this give me? These are questions that our designers are constantly thinking through, and it's a funny thing about software development that something so simple and basic to build is so important, and often the thing we'll do last.