Writer Neil Guiman once said that the purpose of a second draft is to make it look like you knew what you were doing in the first draft.  I know exactly what he meant.  I tend to write quite quickly, with a book roughly every four months, each of about a hundred and ten thousand words.  Writing the first draft takes about two months, with the other two months used for re-drafting, the book’s trip to the editor then the final production of a rechecked, uploadable version with contents page, links and so on.

When you think about it, taking two months to write a book you might read in a day means that I get the story in distinctly slow motion, so it’s very easy to lose track of the flow a bit.  You know the way some half remembered throwaway line on page three suddenly turns out to be vital when you get to page 200?  Well it’s important I do remember, and my stories tend to have a lot of intertwining threads.  They also work to a definite timeline.

To make matters worse, I’m seldom working on just one book at once.  Nor do I tend to write a book as you would read it: begin at the beginning and write through to the end.  I write in modules as ideas occur, then assemble them into a coherent order later.

The result is (and I confess it readily), my first complete drafts tend to be a mess.  They are barely in English.  They contain repetitions.  I might find that some of the threads never resolve.  By the time it’s complete I might have forgotten bits of the story anyway, and will have no idea if it’s any good.

The book I’m writing at the moment, ‘Debbie’s Gift’ has had even more complicated issues.  I was ill for some weeks in the middle and couldn’t work.  But I also got ideas for two new books and wrote first drafts of both before returning to what I was supposed to be doing.  As a result this one has taken almost twice as long as usual.  Fortunately my readers won’t be able to tell as I’d been a bit ahead of my normal publication schedule, but it did mean the first draft was even more confused than usual.

My second drafts are where this sort of mess is (hopefully) sorted out.  As you’d expect, I go through catching typos, bad grammar and the like.  More importantly I try to get the text to flow well, especially the dialogue.  I try to remove anything superfluous and add anything missing.  I hunt down repetitions and decide where the one that survives will go.  I filter out bits I decide I don’t like, and might adjust the order of events.

I also try to get a better feel for whether the story is actually good enough to publish and if it isn’t, I’ve been known to change things dramatically until it is.

Writing a series like the Saltbury Chronicles also brings one further issue.  My books have a range of narrators, but share many characters and places.  Threads in different books interact, which means the narrator of one book might appear as a character in another.  I even occasionally have the same event in more than one book, but seen through different eyes.  One of the biggest howlers I can commit then, is to have the same character be in two places at once in different books and that is another important check at second draft stage.

To do it I have a computer file called ‘Saltbury Timeline’ which records the exact time and date of everything every character does in the whole series.  It’s currently 80 pages long with each character colour coded.  So if I am having Hazel Fauld do something in one of her stories, I can and do check that I don’t already have her doing something else in one of Tina’s.  I update the file (and my character database) as I write.  It takes a lot of extra work and will only get bigger as the series goes on, but it’s saved me from making a fool of myself many times.  So to anyone else writing a series with recurring characters I can recommend something of the sort.  It’s very well worth the extra effort, because readers notice these things and you don’t want messages asking:

“How can Hazel be on stage in London, when she’s at her own birthday party in Saltbury?”