As measuring devices, us humans are quite sucky when it comes to quantifying change in absolute terms, but we can be surprisingly good at noticing relative change.
I've got 3 sets of tires and two sets of wheels that I commute on, one rather nice build, and one bog-standard. Tires range from 1.2 slicks to mid-count(106-160) studded winter tires to high count(288-294) winter tires.
While the difference in feel between nice wheels in slicks, and crummy wheels in either mid, or high stud count is absolutely massive(the high-counts I think are 800+ grams each), the difference in time really isn't that much. I can easily be faster on the heavy gear, if I have someone to pace, than what I am on a solo ride on the nice wheels. There are plenty of times where a winter ride has felt like a really boring slog, but when I get home it's only added maybe 5% on the time. And that's for a change that is more radical than most riders will see on road bike wheel set.
But sure, for someone whose level of effort is more closely linked to how the bike feels, I can certainly see how a moderately lighter wheelset can translate into an unproportionally large improvement in riding.

(this is "dry" winter of course. Icy patches, a bit of sand on the roads. If I have to churn my way through inches of snow/slush, the speed drops radically)