Making the most of your narrative

In a previous post I mentioned a recent blockbuster, The Bestseller Code, which I’d found to offer some true gems of insight to writers. Jodie Archer and Matthew J. Jocker’s slim tome describes the findings of a computer program that analysed the features of bestselling titles. In case you think it may just be stabbing in the dark the program correctly predicted the top two bestselling titles of all time, suggesting they had features that it would benefit writers to know about. If the analysis of the program is at times difficult to – ahem – decode it doesn’t mean there aren’t a few tips that us writers can learn from the bestsellers. In this blog I’ll focus upon how we can make the most of our narrative, and The Bestseller Code has some interesting tips on the subject.

Serious creative writers are – with some justification – often sniffy about bestselling novels. Fifty Shades of Grey is often a target for such derisory remarks about poor writing. It is a novel which may break the ‘show don’t tell rule on the first page’ (telling us that character Christian is ‘attractive’), but it has sold by the shedload. Aha, I hear you say. But what value is selling by the shedload? What about genre fiction, and all those slushy romantic novels with tanned hunks and buxom beauties on the cover that sell by the trolley load? Surely there’s little a writer can learn from them?

Well, it’s complicated. There are so many novels published every day that a novel which connects with the masses, and crosses the boundaries of genre like Fifty Shades, may be tapping into a deep human prerequisite for a good story. When Archer and Jocker analysed two of the most mammoth selling titles of recent years – EL James’ Fifty Shades of Grey and Dan Brown’s The Da Vinci Code– they noticed some striking commonalities between the two books. They got their computer program to plot a graph- with the x axis plotting the percentage of the narrative that had progressed, and the y axis plotting the positive and negative emotional intensity at a given point in the story. Researchers in language call this ‘sentiment analysis.’ The results are striking.

screen-shot-2017-01-03-at-14-47-34
What this graph shows is that both stories have a very similar development of plot (except at the very end when we can see the lines diverge). Both novels show characters experiencing highs and lows at the same point and with the same smoothness in the narrative. Whether or not Dan Brown and EL James are aware of this, this proves that both authors are clearly attuned to a rhythm in the story that generic readers want to experience. The novels have totally different writers, themes, and genres, but their success evidences a connection with humans on a gut level.

Of course, we writers don’t have access to a computer program that can plot the narrative of our stories and see how smooth the ups and downs are. But even if we are not writing novels, the findings of the program strongly suggest that emotional lows (such as a conflict or a fight) followed by emotional highs (a warm or thrilling love scene) all unfurling in a smooth up and down manner that peaks around the centre of the story really connects a story with readers. With a novel I am working on – during a quiet moment – I decided to try and plot the ups and downs of a new story of mine; painfully aware that it is not a Da Vinci Code in the making. The graph was sketched on the back of a bank statement, which I think proves I didn’t intend to share it at the time. In fitting with this hi-tech way of analysing my creative writing, here is a screenshot of what my very rough graph found-

screen-shot-2017-01-03-at-14-59-52

The curves to my story are not very smooth. Although the story goes up and down in a potentially promising way, there are also some kinks and some minor bumps which suggest a smoothness not exactly worthy of Dan Brown. But I had another look at where those kinks were and realised they were scenes unnecessary to the novel. My novel also has a low which – according to my chart – may be too low. I have now removed these extra scenes to try and make the curves of my story that bit more attractive. You’ll notice as well that whereas the graph for Da Vinci Code shows a positive ending, the graph for Fifty Shades a negative one – suggesting that readers aren’t as bothered about happy endings (especially if a sequel in in the pipeline) as we might have imagined…

3 Comments

  1. debsriccio 20 January 2017 at 11:41 am

    Fascinating, Guy. Um… I may be a bit technologically unaware at times but precisely how does the decoder do the decoding? Does a human to do the initial ‘charting’ and feed this information in or does the computer actually read the book itself and produce the code accordingly? As they say these days: Hashtag justaskin’.

    Reply
  2. liz cashdan 21 January 2017 at 9:15 pm

    Fascinating. Has anybody charted this kind of thing for Jane Austen, or Charlotte Bronte for example, or for more recent novelists like Margaret Atwood or Jonathan Safran Foer.
    And how do you know you haven’t removed an innovative low that in fact might be more satisfying to readers than the curly graphs of P.D.James and Dan Brown? Just asking.

    Reply
  3. Guy Mankowski 23 January 2017 at 9:17 am

    Thanks for these intriguing comments. Yes, I think the program read (or followed its programming to pick out raw features, as a statistical program would do a factor analysis) raw text. Re Jane Austen, great question Liz. In a rather cold way I think the authors were interested in the top sellers in recent years. I recall that the program found that Austen’s work was powerful- in their cold view- with the specific use of a female protagonist, which they said was key to its appeal. They were focused on examining key features of the couple of highest sellers in recent years.

    Reply

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.