IMO "kWh per hour" is clearer, because when dealing with power generation those are the units you deal with when you are connected to the grid.
When people talk about their solar generation, they say "I generated X kWh today", not "today my system averaged Y kW over Z hours" and leave the math to the listener. kWh per unit time is common when talking about home energy generation.
OK, first, let me say that I have no objection to your "I generated X kWh
today". Neither do I have any strong objection to "kWh per day" or "kWh per
month". But I do object to "kWh per hour". I would have the same objection to,
"0.6 light-years per year". I would ask people to say, "0.6 times the speed of
light" instead.
It is common to need to refer to two different things that happen to have the
same "dimension" (electrical power being an example of a dimension and energy
being a different dimension). For example, a supermarket needs to distinguish
"gross revenue", i.e., how much money it collected by selling things, from "net
revenue", which is how much money is left over after subtracting the cost of
the supermarket's having bought the things it sold. Note that both gross
revenue and net revenue are best measured in dollars (if the market is located
in the US). So according to you, if I understand correctly, you think it would
be an okay idea for accountants to adopt the convention of using "dollars" as
the unit when talking about gross revenue and "dollar-months per month" to
signify that we are talking about net revenue. But what are you going to do
when we need a third measure that happens to also be denominated
in dollars (profit, say)? Are you going to adopt the convention that
"dollar-miles per mile" refers to a profit measure?
There was a famous physicist (Steven Weinberg, I think) who criticized
Windows for giving science and technology a bad reputation. In particular, he worried that
young people would react to the buggy behavior of Windows by becoming sour
on the whole edifice of science and technology and would try less hard to learn science.
I guess my main objection to "dollar-months per month" and to "kWh per hour" is like that: I worry that it will prevent (via confusion) some fraction of people from learning enough about accounting and the fundamentals of electricity to wield words like "gross revenue", "net revenue", "energy" and "electrical power" effectively. (One of the sources of my worry is my imagining how I personally would've reacted to coming across "kWh per hour" long ago before I had mastered the concepts of watts, energy and power.)
To bring this back to where we began, if there are two different quantities of interest, namely, the advertised "capacity" of the panel versus the power a consumer can realistically expect to get out of the panel on a typical day, I would ask that people settle on some short phrases for both -- similar to "gross revenue" and "net revenue" -- and to denominate both quantities in kW (or kWh per month maybe).
IMO "kWh per hour" is clearer, because when dealing with power generation those are the units you deal with when you are connected to the grid.
When people talk about their solar generation, they say "I generated X kWh today", not "today my system averaged Y kW over Z hours" and leave the math to the listener. kWh per unit time is common when talking about home energy generation.