There may be a little conundrum with this …
On the one hand we have a set of more or less “fixed” constraints, like 340 ADA Fix, VPS rental, domain, power, network, etc. Looking at these alone, one could be tempted to make absolute statements on sustainability, real cost etc. For example, not too long ago it was believed that an SPO needs not only those 340 ADA for the first block but also some additional margin to be able to run a pool securely.
On the other hand, we have a high variability in ADA value. What was true yesterday, may be non-sensical tomorrow. For example, given 6 x 340 => 2040 ADA p.m. fix and an ADA value of $2, it cannot possibly be claimed any more that this is really “needed”. No one actually needs $4080 p.m. to run two simple servers let alone a 2% margin on top. With an even further increase in ADA value, which I guess non of us would doubt, the above claim (i.e. an SPO needs …) becomes ever more funny - who can then still sell this to a delegator with a straight face - and who would?
I’d say, this is nearly impossible already - I mean with a straight face 
Some factors are fixed, others are not - the redistribution of pool rewards needs to become a little more flexible too - and it will of course. With Alonzo, we will get a programmable blockchain. Creativity will get unleashed and far more choice for the delegator will emerge.
Daedalus might not show it, but other communication channels will. The concept of effective cost per block might become an important metric for the delegator.