if you are doing it so that you can consistently plant and account for a given flower period or maybe more consistently time your harvest with the same genetics etc etc...
best to always count from when you flip to 12/12. this way, your bias and perceptions don't impact the count. you won't always choose teh same point, which will cause all sorts of additional variance in your flower day counts. it'll be less useful information for planning, in other words.
if you run clones, it is absolutely beneficial to know this stuff. if you treat it the same way, it will be the same result with the same genetics. you can count on that... if it deviates, there's a factor that is different causing the difference. life is a set of dominos, no matter how complicated it seems... cause and effect... repeatable/consistent cause and effect.