I think this is one of those cases where it looks good on the white board but in actual practice probably doesn't come up too often. Thinking back to our last 3 sessions, using tactics like these might have been somewhat useful in one fight (against hordes of zombies). I think that is Mr. Crawford's point.
As for Cloudkill...it kind of proves the point that it is intended for these spells to do potentially double damage...Cloudkill is 5d8 and can double to 10d8 if the caster can keep the spell going. Cloudkill is straight up 8d8. 8d8 right now vs potentially 5d8 per round at the expense of concentration. Seems a fair trade, no?
Except that Level 5 Cloudkill does 5d8 because it affects multiple targets, not because it is concentration based. I assume you were talking about level 4
blight at 8d8. It does the higher damage because it only affects a single target.
The guidance for making spells is not nearly as robust as those for making a monster. If Cloudkill was following the DMG more closely, it would do (27 (6d8) damage to be more in line with the damage table on page 284 which calls for 28 (8d6) damage for a 5th level spell that affects multiple targets. My guess is that the lost 1d8 is because it is a concentration spell that can keep going.
But the spell damage table gives a rough idea of how much damage a spell should do to a target each round 45 (10d8) for a cloudkill that hits twice is significantly more than 28 (8d6). Granted, a cloudkill that hit twice would be possible, but very unlikely. Looking to the spell damage table, a 3rd level spell should do about 21 (6d6) damage.
spirit guardians does 13 (3d8) damage. It may be losing some damage because the caster may take actions normally the next turn, and it may also be losing 1d8 because of concentration. 18 (4d8) would certainly be closer to the suggested damage. If it hit a creature twice in a round, the 27 (6d8) damage is a bit better than the suggested damage, especially when it can happen again the next turn along with other actions by the caster.
When compared to one off spells like
blight, 36 (8d8) is quite close to the suggested 33 (6d10). Fireball 28 (8d6) is actually better than the suggested 21 (6d6), possibly because means of mitigating Dexterity based saving throws are so common.
The other concentration damage spell mechanics that I am aware of are
witch bolt (1d12 ongoing with 30' limit on 1 target), and
wall of fire (damage on casting plus damage on first time entering during turn or end turn in range). Both of these give the target the option of taking more damage
I have not gone through an extensive list of the spells, but it seems like most spells are very close to the suggested damage or lower. Given that most spells seem to not exceed the damage by much (Dexterity saves aside, and I am sure there are other examples), it seems strange that this small handful of spells could exceed the suggested damage by going through some easy extra hoops. Aside from the long run to hit lots of foes, the double damage could be triggered by casting the spell on some enemies (they enter the spell, then their turn comes around) or the caster moves the aura back a few steps and forward again to make them enter it for the first time that round.
Making so that the damage only occurs when the target enters the spell or starts its turn already there brings the damage potential in line with the other spells in the PHB. Moving the spell means the damage kicks in at the start of the target creature's next turn. Shoving a creature into the area of effect causes the damage to take effect at the start of the creature's next turn. If the creature moves into it If I had not ruled this way, then Spirit Guardians and Moonbeam would have come out almost all the time because they are then Too Good (tm) compared to other spells of the same level.
