Less than 45 days was doable. :)
I'm considering to stop seaching for new kanji/vocab each day and repurpose that time for focussed reading practice, collecting new kanji/vocab on the way. This would, I expect, slow down my kanji progress significantly. But I feel that I've reached a point where the hassle of blindly seaching for new words to learn, solely for the purpose of getting those damn 常用漢字 done isn't really worth it anymore.
When reading Japanese it isn't the kanji that hold me back, it's grammar, phrases, compounds, etc. I presume simply reading a lot while looking up the stuff I don't understand will benefit my reading comprehension way better than continuing what I've done up until now. Everything new I come across this way will have the huge benefit of being learned with a context — which is kind of taking my 'learn kanji in the context of words' approach one step further: 'learn vocab and grammar in the context of continuous text'.
As I predicted, exams caused a noticable slowdown. Which would've even been more severe had I not tweaked my learning routine a bit.
To save time I decided to stop writing out readings for recognition cards while continuing to write kanji + furigana for recall cards. Curious side effect: I have the feeling this helps me paying more attention to the correct pronunciation of words, since I have to verbalize the word in my head precisely in order to be able to compare my answer to the written solution in Anki. I guess coming up with an answer in a "verbalizing context" is somehow different to writing it down.
As for recall cards I started trying to always think of the pronunciation of a word first instead of the kanji. Reason: in conversations I need the pronunciation of a word immediately. When writing it is no problem to halt for a few seconds.
Two more weeks of exams to go. Won't manage to get to 1800 in 30 days I guess, but less than 45 should be doable. :)
Pun intended. Semi interesting kanji related observation:
I learn kanji by breaking them down into their parts and building mnemonics out of those. Since not all kanji are either simple — in which case it's reasonable to remember the single strokes — or composed only of other kanji, I necessarily have to carry some additional baggage along the way.
I wondered how this turned out for me up until now and created a plot. For learning my first 800 kanji I memorized 97 additional non-kanji characters (12% overhead). For the next 800 it was only 13 (1.6% overhead, 6.8% overall).