Eh, a major consideration of Colossus is how many water tiles you really are controlling.
Each city that might benefit from Colossus by working Coast/Ocean(if FIN) also has to get a Lighthouse, else the tiles are not food neutral and really not even worth working compared to a cottage or specialist at that point. So the sunk cost increases by an additional 60 hammers per city that will be working water, which makes the 120+125 hammer cost (wouldn't really try at all without Bronze, maybe if IND) deceptive.
It's not that I dislike the wonder. It's fun to go thalassocratic and draw great economic power from the the sea, especially as Darius or Hannibal, and I have an inflated fondness for water tiles because of playing Civ Rev so much. But like many, many of the wonders, it's supplemental at best, which makes things like comparisons of tiles yields not very useful, IMO. Those direct comparisons also assume working the tiles constantly without whipping them, which is a whole other thing (which tiles to work when at "n" time in the game, etc).
You can however look at the hammer investments required to utilize it and the economic opportunity cost of not using them for Wealth. The stage of the game where Colossus is most relevant is where Wealth is often the best build project option until you're ready to roll units out, so the hammer tradeoff is not insignificant unless you plan to (or have to) settle all your cities with tons of water tiles. And then work those water tiles as much as possible, placing more limit on whipping or specialist running.
Other reasons for not caring much about it are:
-Little incentive to go for MC early unless IND, really. Maybe if you want to Machinery bulb
-usually one AI in the field will prioritize it. Remember AIs can always see at all times what wonder/tech option are open and will strive for their own monopoly on something, Metal Casting is always an option since every AI techs Iron Working
-Land tiles are generally more powerful due to food/whip or specialist relationships and as sources of hammer yields, water tiles are food neutral at best except for resources and lakes. You don't want to bend over backwards for a wonder to improve the yields of a handful of tiles, you want to get a larger benefit for that investment. FIN trait is the biggest help to this.
That said, there are situations where it can pay off wonderfully. The recent Joao noble's club game is an example, where there is is little land parceled to the human player yet a ton of islands and thus water tiles to work (and even then, because of all the "intercontinental" trade, GLH was MUCH more lucrative and important). Any map with a significantly reduced amount of usable land increasing the prevalence of working water ("island" type maps) can be immediately identified as Colossus-friendly.
Financial also makes it much more attractive, as it can then buff even Ocean into 5 yield tiles with Lighthouse, and Lakes become nearly endgame power if eligible for a Lighthouse too.