Had a lot of fun figuring this one out. The "Limit" trait is surprisingly powerful.

(Potential spoiler hidden with ROT13)
Svthevat bhg ubj gb ernpu gur raqvat jvgu gur arj raqvat pevgrevn frrzf yvxr vg jvyy or dhvgr gur punyyratr...!

I keep playing this every so often since it's come out for like a long stretch of time each while listening to political debates on youtube. This is a good game for that sort of a thing. I like trying to find the optimal skill tree, as it were, of this game. I like how minimal the mechanics are and how you could apply these sorts of stats to a lot of styles of things. I at first was irritated that it felt like you had to click something as soon as it popped up, but I also like how it is, especially later in the game where you get upgrades more frequently and may not want to immediately upgrade constantly. I also really enjoy the difficulty ramp in this. There is sort of a 'it gets easier as it goes on' and 'your early decisions matter more than the later ones' and it puts sort of like a relaxing zen vibe over the last level and such if you've done pretty well. I also like that you want to make it so the cats just barely touch the tower at that last level, it feels super good just seeing them just touch it.

V good game, my high score was 980-something. I'll try to get over 1000 soon!

This is such an interesting puzzle! I won the game on my third attempt, but the "better ending with more than 1000 money" was a challenge I couldn't ignore! I had to get a lot better at it though- everythingstaken's comments on how the beginning choices matter the most drove my approach.

If I understand the math correctly, a value of 2 for a stat doubles the effect, a value of 3 is a 50% increase, etc- a sort of multiplier on the base value, rather than a fixed, incremental rate of gain. As such, getting each value to 2 is imo the most important beginning. But, which order to pick things in?

Obviously, picking Rate first off is terrible. Whenever you cannot fire because the limit is reached (which I thinking of as "waiting time"), the game is telling you "your build is not effective!". Speed and Limit are appealing choices, because they effectively, immediately halve the waiting time. Number is also an interesting first choice, counter-intuitively:

At first I thought that Limit meant that whenever you fired bubbles, the number that would be fired would be capped by it. ie. if there are 3 bubbles on screen, Limit is 4, and Number is 2, what I expected is one bubble would be fired. However, it's not so- 2 bubbles are fired, but I cannot fire again until there are 4 or fewer bubbles on the screen! So the relation Number and Limit have wrt the waiting time is very interesting!

So it seems that an optimal first path could be something like Speed, Number, Limit, Limit, Rate. However, there is one more significant detail on Number that makes it the most important stat to think about. Number directly corresponds to your income per bubble launch. to get the next upgrade, it will take (upgrade cost - current money)/Number rounded up launches. the more launches it takes to get an upgrade, the slower your rate of improvement will be, which will quickly become disastrous!

with all of this in mind, and after a handful of attempts, I came upon this general approach (barring the first ~6 upgrades, which I don't have an optimal order for yet):
1) if the waiting time is ever more than a few frames, increase Speed
2) Limit should always be 1 higher than Number, so that two launches can be done in rapid succession
3) Number needs to increase a lot early on, to keep up a high rate of income
4) if there is consistently _no_ waiting time, increase Rate
5) never, ever, heal. it's a waste of money, and if you need to do it, it means your build order was not optimal

I found that I could sit comfortably on a Rate of 6 (which is a bit overkill, but I lost with 5 once!), Speed of ~18-20, Number 10, Limit 11. I deliberately sat on my money so that I would just barely get over 1000 at the end, which is enough of a goal for me :)