Mercy Stroke

[Grembo]

It's fine as it is apart from the lengthy cooldown.

I've posted this elsewhere but you should try and macro mercy stroke to each party of your enmity rotation with a small delay on it, like this all you have to do is spot an enemy with low enough hp and tab target or click them and boom the enemy dies and you get the healing bonus.

Works fine on my PLD, WAR and MNK even in end game.

[Shurrikhan]

It's fine as it is apart from the lengthy cooldown.

I've posted this elsewhere but you should try and macro mercy stroke to each party of your enmity rotation with a small delay on it, like this all you have to do is spot an enemy with low enough hp and tab target or click them and boom the enemy dies and you get the healing bonus.

Works fine on my PLD, WAR and MNK even in end game.

Without macros I still get the heal more often than not, though I'd also recommend Grembo's advice. I'd also agree that the main issue is its cooldown. It's a useful split-second saving-grace or buffer for a tank in long, large or rolling-dps pulls. It'd just be a lot more useful if it was actually up every 40 or 60 seconds instead.

But, that's the tank perspective. The dps perspective, if undamaged, is essentially to absolutely not use the ability when the tank could use it, so that the tank can grab the 1.4k heal. It doesn't hit for much, and you're unlikely to see it come up twice during the low enemy hp needed to use it. It's decent for groups that are good at being sloppy (the kinds where DRGs pull mobs through threat, but then just tank them without really taking damage). When that's no longer an option, it feels nearly pointless.

So:

Curse application (2-5) seconds: takes out the team management, where the party leaves the killing blow to the tank. Good, or a little bad, depending on your personal views. Certainly makes it a whole lot more useful during latency...
Making it actually worthwhile for dps with some manner of strategy attached, such as dealing a portion of the last 3-attacks' damage: wouldn't mind, though that will then seem more or less unnecessary for PLDs, who won't have the AoE to buff that damage even if AoE could contribute (more likely it would have to be the last 3 hits, rather than "abilities").
Increase base (potency): taken far enough, it could surpass the use of applying Fracture, if it comes down to a choice between the two, but for DRG the difference is already minimal.
Reduce CD: I honestly think this should happen in addition to whatever works from above. 60 seconds or less. Will make it kind of OP for leveling, I suppose, but... by that rationale the classes that wouldn't be using it are already largely advantaged for leveling. (Bard, BLM, etc.)

[Remmadar]

I still don't see much use for an execute ability on a tank that just heals you..
If your fighting a boss, its a wasted skill and has no use other then to push it for some extra dps

I highly doubt when creating dungeons, SE designed tanks to really need that extra health restoration when fighting. It's far to difficult to hit when dps are doing 2-3x that damage you are.

Changing it to some it some dot effect or some 5-10 sec debuff thing would be ok I guess.. but still just doesn't seem to have an impact on the game in a meaningful way atm