Good point, but I disagree on a detail: just knowing that there's some threshold of savings that some people have achieved when switching to X Brand Insurance tells us that at least one person passed that threshold... but it doesn't tell us how likely we are to pass that threshold.

For that I think you'd need something like "10% of the people who got a quote from X Brand insurance later switched, saving on average Y dollars!" Except, no insurance company would run that ad, because 10% is an unimpressive "sounding" number even though in this context it would actually be really high.