Make an early start at identifying a topic and then fixing a title, and talk it over with your supervisor. Keep discussing the dissertation with them as you work on it, too.
IME it's a good idea to get stuck into the research early. I find that the process of research gives me ideas, and in any case, the earlier you start the less of a hurry you'll be in. Research is always more enjoyable when you don't need to rush it IME.
Are you planning on doing primary research in archives? IME it's a good idea to familiarise yourself as far as possible with what you're likely to find in the archives. There are loads of published guides to various bodies of records. They should give you a better idea of what you can expect to get out of the records and where else you might look for information if you don't turn up quite what you're after.
Make decisions early on, on various stylistic points - e.g. the precise format of your footnotes - and then stick to what you've decided: it makes the final editing much easier.
As others have said, get at least one other person to read over the finished article, preferably someone with some knowledge of the subject. They'll almost certainly spot mistakes, duff sentences and places where the argument's not at clear as it might be, which you'll probably miss.
IME it's a good idea, if time allows, to have a few days' break between writing and editing: you come back to it with a clearer mind.
