"It doesn't have any gameplay" isn't really a valid criticism. There's absolutely no problem with not having any gameplay (had it actually been true, Cage games do have gamplay). It is, however, indicative of problems with the game. One of the most important lessons to learn when doing anything creative is to not take criticism by face-value but to try to understand why that criticism is being made. Sometimes the listener/player/viewer gets it right and correctly identifies their problems with a product but a lot of the time they're not enjoying the experience and try to find a reason why, but the reason they come up with isn't necessarily the problem and trying to fix it won't lead anywhere. See: plot holes in films. (Sure, a plot hole is a problem, but almost every film ever made has plot holes yet you usually don't notice them because you've suspended your disbelief and the film is engrossing. If you're latching onto plot holes, the real problem is probably that you're not engrossed in the experience.)
So, what does "there's no gameplay" actually mean? Well, it could mean that you've set up expectations in the player for how the game is going to be but then you don't fulfill those expectations. The problem isn't what the game is, but what the player thought it was going to be. So how do we make sure the player doesn't get the wrong expectations?
Example: I made a short game in which you play as a failed experiment, your only action is to type on the keyboard and send those messages, except due to communication errors, every letter you type turns into Z. The scientists try to communicate with you but eventually give up and shut you down, killing you in the process. The game is specifically about disempowerment. For the most part it worked but some players found it very frustrating, trying to look for alternative endings (that didn't exist). The problem wasn't the lack of alternative endings, the problem was that I had somehow given these players the faulty expectation that there would be alternative endings, which made the experience very frustrating for them.
But the problem can of course also be in what the game actually is, as I suspect it is in the case of Cage games. I haven't played any Cage game apart from a little bit of Indigo Prophecy, but I understand that they involve Quick Time Events.
QTEs are clearly gameplay. They require user responses and challenge the player, failing him/her if they don't manage to do what the game requires them to. But even if they didn't provide any challenge or fun, that wouldn't be a problem if they added something to the game. (In fact, sometimes a game should be a little "boring", it's called pacing.) The problem with QTEs is that they usually don't add to the game; rather, they distract from it.
So, what's the purpose of QTEs? What are they supposed to convey to the player? I can think of two potent use cases.
1. To convey disconnect and disempowerment.
In games, you usually want to feel like you're in control of a character. One of the most important things in establishing that relationship between player and in-game character is consistency in rules. Humans tend to love consistency. If A is the jump button, then the player should never be required to press B to jump. If button-to-action mapping was completely random (every time you press a button, the mappings are reconfigured, so now left-button is jump and right trigger is walk forward) you wouldn't feel like you were actually controlling that character.
This is pretty much the case with QTEs (as they are commonly implemented, not inherently, though). In the common implementation of QTEs the player is required to press a semi-random button or sequence of buttons to continue. This means that there is no clear, consistent connection between input and action. The effects of a button press are not easily predictable, leading to a disconnect between the gameplay and what is happening on screen, as if they don't really have anything to do with each other. This is why players sometimes complain that it feels like watching a movie that occassionally stops and asks you to press some buttons to continue.
This is great if you want to convey a disconnect or disempowerment.
2. To convey stress and urgency.
"QUICK! PRESS THE BUTTON OR SOMETHING BAD WILL HAPPEN! YOU NEED TO DO SOMETHING! NOOOW!!!!"
Ah! Oh God! Shiit! Where's that button?! Shiiiit!
This is how I feel most games want to use QTEs. And it certainly has the potential to be effective. It can, however, be very difficult to implement, for a number of reasons.
A. It can conflict with use case 1.
If you want the player to feel urgency and the need to do something, it usually helps if the player feels like their actions matter in some way, rather than be useless.
B. It can't actually be that urgent, or, what if the player sees through it?
You can't actually require the player to be all that fast. The player needs time to process what they're supposed to do, and you don't want players with poor reaction time to have to replay the same scene over and over again. (Because replaying the same thing completely and utterly ruins immersion.) So what if the player notices that it isn't actually urgent? If you play The Walking Dead and intentionally don't do the QTEs, you'll see that you actually have a pretty large window to do them.
This is a relatively easy (and yet difficult at the same time) problem to get around. The solution is simply: immerse the player. Convince him/her that it actually is urgent and that they do have to press the button even if it's not true. If the game is well designed and immersive, this shouldn't be too much of a problem.
C. You can never EVER, EVER use QTEs for something that isn't stressful and urgent.
You can't use QTEs to stroke the head of the main character's child. You can't use QTEs to light a cigarette. Why? Because you just established that QTEs aren't urgent or stressful. In one action alone, you completely ruined the potential of QTEs to be stressful or urgent. That's a deadly sin! The player will now forever associate QTEs with the scene in which the protagonists calmly and slowly lit a cigarette, with no pressure whatsoever.
Example of good use: The Walking Dead
The reasons QTEs actually work in The Walking Dead are multiple but they all stem from the same base: Telltale knows what QTEs mean for The Walking Dead and they consistently use them in that manner.
Problem A isn't much of a problem because The Walking Dead largely is about disempowerment. It is, to a large degree, about hopeless characters in a hopeless situation. The sense of disempowerment is entirely appropriate.
Problem B isn't that much of a problem since The Walking Dead is mostly well-designed and immersive.
Problem C isn't a problem because Telltale uses QTEs consistently. It was a little while since I played it, but I can't remember any QTEs in situations that weren't stressful or urgent.
Example of poor use: Battlefield 3 (Single player)
Battlefield 3 was, as I interpreted it, larglely supposed to be a power fantasy. That's the last place you'll want a sense of disconnect and disempowerment. If it wasn't supposed to be a power fantasy, then I suppose the QTEs were mostly appropriate, it's just every other aspect of the game that needs to change because every other element of the game screams power fantasy.
Battlefield 3 also has plenty of doses of Problem C. Not because it uses QTEs for non-stressful things, but because it uses other mechanics for stressful things as well. Battlefield 3 is constantly supposed to be exciting and stressful. This means there isn't much of a difference between the shooting gameplay and the QTEs. The QTEs don't seem to ramp up the stress-levels, they just seem to remain the same. In this way, Battlefield 3 has very poor pacing. For comparison, The Walking Dead has a much clearer difference between walking-around-adventure-style-gameplay and QTEs. It has better consistency.
Without having played Cage's game, I suspect they use QTEs in a way that's not consistent or to express things other than what QTEs are good at expressing.