In idle games and clickers the fail state, such as it is, is usually that you're using your own time inefficiently, as you build towards something. In the case of your game, the player is building to delay and hopefully stop something that is otherwise going to happen.
There's definitely something interesting there, but I think the challenge will be how to avoid it becoming an optimisation task, as the player's inefficiencies will eventually determine whether it's even possible for them to win (I suspect it won't be fun if the player has effectively mathematically lost after 5 minutes, but won't realise this for another 5 or 10 minutes).
It might be worth exploring some secondary loop that exists independently of the loop in the prototype. For instance, perhaps you can invest experience/some other currency in long-term upgrades that allow each failure in the underlying loop to set you up better for the future.
In terms of the prototype itself, I wonder whether you could find another dimension for interaction. In idle games there's often a tension between clicking and passive abilities, with upgrades being available for both.
Good luck with it :)