i think it was a ten thousandth of the distance from the north pole to the equator through paris, disregarding the southern hemisphere, which was more difficult to survey. i've been looking for a precise number for how far that distance actually turned out to be, but i haven't been able to find it; i would be delighted if you could!

whether 0.02% sounds ridiculously good or ridiculously bad depends on your frame of reference

https://en.wikipedia.org/wiki/History_of_the_metre#History_o... says that the original mètre des archives in 01799 was machined to within 50μm, which is 0.005%, four times smaller than the error in delambre and méchain's computation. so even at the time that was a pretty large error. since then the measurement uncertainty of the meter has improved by five more orders of magnitude, to about 0.1 part per billion. 0.02% is 200 parts per million, or 200000 parts per billion, which is a lot more than 0.1