I got two things, that I brought up last time someone suggested this, first, there is no reasonable measurement you can come up with in which a day will fit evenly in a year while also being divisible by 10, so either your days lengths would have to vary or year lengths would have to vary making the concept moot unless you have make an exception and have an exception leads into point two. The universe doesn't care what base your measurement system is, the only value of the metric system over the imperial is that it is easier to convert given we also use a base 10 counting system (and even that is not much of an argument if people were taught the imperial system well as the human mind is more than flexible enough to not care about the difference when used properly) it is not inherently better. So if everything did not fit even then there would be no point to switch as its only advantage is gone.
The OP is talking about dividing the day decimally rather than expanding that to the year.
However the fact that unlike spatial measurements, where it's easy to extend the metre out far, things like seasons and lunar cycles exist does help explain why the metric standard didn't catch on.
Still the International Atomic Time does exist upon which all our clocks are based and it uses a defined second that
isn't a fraction of the Earth's spin.
So, dividing the day into a decimal amount is possible. It just needs to be practical enough for adoption by both farmers, industry, and scientists.
86400 seconds per day is close enough to 100000 for our sense of half and full hours to fit across, much like a lot of European countries could restandardise their pounds to half kilograms and pints to half litres.
E.g. 1 day = 1 hectochrone = 100 "chrones", 1 chrone (c14 mins) = 1000 millichrones.
Basically the standard unit needs to be easily relatable to common uses to be accepted.