Book discovery sites offer authors and publishers the opportunity to promote their books to mailing lists comprising thousands of avid readers. Readers subscribe for free, while authors and publishers pay to be included in a listing.

The largest of these sites, by far, is BookBub. But, is it the best?

BookBub is a book recommendation site with between one and four million subscribers, depending on the genre. They claim to distribute to 3.82 million fans of crime fiction, for example, and 2.1 million readers of romance. In my genre of literary fiction, they claim 2.61 million subscribers.

Promoting your e-book through the site would seem to be a no-brainer. But it doesn’t come cheap. To promote to their e-mail list for literary fiction costs $768 if you reduce your price below $1.00 So, to recoup the cost at $0.99 a copy, you have to sell 776 copies.

Don’t decide on which book discovery site to use based on its price alone. If BookBub gains you 1,000 sales at a cost of $776, the cost per sale is 78 cents. If another site sells 100 books at a cost of $90, the cost per sale is 90 cents.

BookBub will not accept every book submitted for this service. They handpick their promotions, and accept only around 10%. Here are some of the factors they use to make a decision

• Competitive price, discounted on all retailers
• Number of reader reviews and rating
• Professionally designed cover optimized for the genre
• Optimised product description page

Is it worth it?

I tracked the performance of 102 lit fic titles promoted through BookBub between the beginning of September 2020 and the beginning of December, following each for 11 days.

BookBub claim that, on average, their promotions result in an additional 2,560 sales in the genre. The titles I tracked achieved less than half this on Amazon, at 1,090. But that’s still comfortably above the 776 needed for break-even. The effect is short-lived. After the boost on the first day, sales dropped sharply to an average of 10% of the initial boost by day five.

But, like anything, success is not evenly distributed. Twenty nine per cent of the titles experienced no impact on sales at all. For the remaining 71%, sales rose, on average, almost two-hundred-fold compared with the level before the promotion. So that’s pretty good, but for individual titles it varied from 12,910 additional sales (which was for Tracy Chevalier’s A Single Thread) to just 1.

Around half  (54%) of these titles achieved more than the break-even sales figure of 776. On average, they racked up 2,761 sales. The other half  sold an average of 364.

In summary, the odds of covering the costs of a BookBub promotion are 54% of 71%. In other words, 38%.

What kinds of books do better?

Analysing the performers (those that achieved sales) and non-performers (those without sales), some differences stand out

• Performers have a higher rate of reviews. On average, they each had 522 reviews, compared with 400 for non-performers
• Non-performers tend to be more recent publications. The average publication date of non-performers was 2008, compared with 1994 for performers
• Non-performers have a higher percentage of foreign and translated titles, accounting for almost a quarter of the total, compared with 13% for performers. This difference is even more pronounced when the performers are divided into those that recoup the costs of the promotion with those that don’t. Only 6% of those that covered their costs were foreign or translated, compared with 15% of those that didn’t cover their costs.

These differences were not repeated in other book discovery sites. A comparable analysis of BookGorilla showed that titles that achieved no sales had more reviews than the performers. And the group with no sales were, on average, published earlier. None of the BookGorilla titles were by foreign authors or in translation, so this comparison cannot be made. However the role of reviews did show up again in the distinction between the titles that covered their costs and the group with sales that did not cover their costs—those in profit had three times as many reviews as those who did not recoup their costs.

Are other sites better value for money?

BookBub is by far the largest book recommendation site, but there are many others. Many are too small to be of great value, but some of the intermediate-sized sites offer the prospect of significant impact on sales at costs lower than BookBub. I compared three of these sites, BookGorilla, Bargain Booksy, and Fussy Librarian. I also looked at one free site, Book Angel. In all cases, I compared the effects on titles classified as literature.

BookGorilla looks like a contender for BookBub’s crown. It is possible that some sales are due to “stacking”. This occurs when a title is advertised simultaneously on multiple sites. To check that sales on BookGorilla were not really reflections of promotion on the much larger BookBub, all titles that appeared simultaneously on both sites were stripped out of the BookGorilla figures. There were 10 such titles. Removing them reduced the average sales on BookGorilla to 1,498, reduced the number with no sales to 15% and increased the chances of break-even to 60%. BookGorilla’s performance was therefore not caused by overlap with BookBub.

The BookGorilla results were so surprising that I ran them again after the Christmas period on a smaller sample of 28 titles. The numbers were comparable for break-even (54%) and for no sales (21%), though sales were lower at 435. A repeat of the BookBub analysis for 30 titles showed no evidence of a post-Christmas slump, averaging 1,353 sales. The proportion who showed no sales was consistent at 30%, though only 30% broke even, a smaller proportion than previously.  

The numbers seem clear. BookGorilla offers an equivalent sales boost to BookBub, a dramatically lower price, and a better chance of achieving break-even on the cost. BargainBooksy and Fussy Librarian did not perform nearly as well.  The free site (Book Angel) achieved almost no growth in sales.

Two notes of caution about the comparison.

• The sales on BookGorilla may have been artificially high because the first numbers were tabulated in the run-up to Christmas, while the data for BookBub was collected a couple of months earlier. Sales did fall in the second sample of BookGorilla titles after Christmas. There was little evidence of a seasonal boost though for BargainBooksy or Fussy Librarian.
• The sales on BookGorilla, though apparently higher than BookBub, are boosted, in part, because many of the titles were already achieving high sales before the promotion, perhaps due to stacking.  As a result, while BookBub saw an elevation of sales by an average 179-fold over the baseline, BookGorilla yielded a more modest average boost of 17-fold.

“Stacking” book discovery sites

Up to this point, I’ve talked about book discovery sites as alternatives. But this isn’t necessarily the best way to use them. You can stack several promotions together. Don’t think of them as being simply additive. 1+1 can be more than 2. This is because of the way the Amazon algorithm works. To get best value from the promotions, you want to kick-start Amazon into recommending your book to more potential buyers. The algorithm notices a successful “spike” in sales. But, as seen, this tails off very quickly. So, if you can maintain sales over time (say 5-7 days), the algorithm will notice the “consistency” and interpret as organic buying behaviour. This will increase the chances Amazon recommends your book.

So, you want to plan a promotional campaign that creates an upward sales trend over the 7 days. You want to come off the promotion at your peak Amazon ranking. Combine all the firepower you have available. For example, you may start on day 1 with a promotion through social media, your website, newsletter or mailing list. You may want to combine this with advertising on Amazon and other places. Around day 3, bring in the first of the book discovery sites, and introduce the others over the succeeding days, ending with a bang on the penultimate day. If you’re lucky, Amazon will give you a long “tail” of sales after the promotion is over.

It’s also important not to think of an individual promotion as an end in itself. Think strategically. Your short term goal is to sell more books, sure. But, longer term, your success depends on building a sustained relationship with fans and subscribers. It also depends on learning what promotional tools work for you. Think about how an individual promotion can help you towards the goals of building your skills in promotion and reaching out to a wider audience.

Conclusion

Opting for the biggest and most expensive site may not be your best option. BookBub performs dramatically, but not for every title. BookGorilla performs equally well, for less cost and with a higher chance of recouping your costs. Think about “stacking” the sites to achieve greatest effect.

Since classical times and Aristotle’s Poetics, we have believed that stories must have a beginning, middle, and end. Even with distortions of the timeline such as flashbacks and flashforwards, such stories move with inexorable causality from their starting conditions to the final consequences. And this does create extremely satisfying tales. But what if we try to imagine a form of story-telling that is divorced from the iron hand of time and from the laws of cause and effect?

Looks like

What, for example, if the organizing principle of a story is homologies? Homology just means a likeness in structure. For example, we might say that confectionary with a liquid centre is a homologue of our planet with its molten core.

This idea of homology might seem unusual in our era of scientific understanding of cause and effect. But it’s a very old idea. For much of the Middle Ages, scholars attempted to understand the universe using this principle. For example, believing that there was a principle of homology between the earthly and the heavenly realm, healers felt that God had created a medicinal plant for every ailment, and that these plants could be recognized by their “signatures”.

The white spotted leaves of lungwort were used to treat tuberculosis because they were thought to look like diseased lungs.

Satin Island by Tom McCarthy is a good literary example of the use of this principle of homology. The narrator is charged by a mysterious multi-tentacled consultancy firm to create a report on the codes governing the present age. He finds connections and patterns everywhere, and therefore perhaps, nowhere.

Fractals

Connections and patterns everywhere. That’s a characteristic of fractals. Fractals are mathematical entities, repeating patterns. Whatever level of magnification you look at them, they go on and on forever.

How cool would that be for a story to exhibit the same pattern wherever you looked? To go on and on forever? Of course, stories have endings. But their resonances in the reader’s mind may persist as long as that brain exists.

There are various ways in which stories can have this fractal-like effect. Let’s consider them, one by one.

• The reader’s experience of structure. This is different from the writer’s conscious use of structure, in that it is not usually deliberative. Let’s look at this first in an analytical mode, the mode the writer uses. Different writers follow different conventions about how a story is structured. But common terms include scenes, sequences and acts. Some include the notion of beats as the smallest atomic unit of dramatic writing. What all of these have in common, is a change in tension. There is an incident, a rise in tension and a resolution. This creates change and movement. This general formula can be applied from the smallest unit of story (the beat or scene) all the way to the arc of the entire narrative. In a sense, this repeat pattern is fractal. Maslow’s famous triangle illustrates this at the level of the whole story.

Now let’s examine the reader’s experience of this structure. There’s a rhythmic rise and fall of tension building to a crescendo. Though the reader may not be aware of the units of the rhythm, the body experiences it.

• The most literal fractal story would be one which exactly repeats at every level. Nancy Fulda describes what this would be like: “Having a novel about a Father who loses his son, in which a father loses a son in each chapter, is not going to go over real well unless it’s some sort of an artsy/literary thing.” But she says, at the level of thematic similarity, something like this is possible: “A novel with a theme of loss might have an overarching storyline that addresses that theme, coupled with several subplots that address it in different ways, coupled with word choices at the sentence level that also emphasize the theme. That’s a pattern that starts to sound distinctly fractoid.” We might also explore an idea at different “scales”. Are children learning the same lessons as their parents? Is the family going through the same seismic shifts as the country they live in? I’m indebted to A.C. Blais for this idea.
• Symmetry. Our brains like and respond to symmetry. Stories that loop back to their beginnings (either as a circle or a spiral) are very satisfying. The Hobbit is a  classic “there and back” tale. The widely-used, if rather mechanical, Hero’s Journey format is full of symmetries in which different segments mirror each other.

A more complex mirroring occurs in David Mitchell’s Cloud Atlas, a nested series of two-part stories.

• Echoes and motifs. This is a more subtle (and less geometrical) example. When themes and images recur throughout a story, our brains register the pattern (though not always consciously). Pattern is one of the main ways in which we understand the world. In the case of echoes and motifs, the pattern is not an exact repeat, but it creates an atmosphere laden with meaning. Consider, for example, The Great Gatsby. Arguably, the novel charts the hollowness of The American Dream. Recurring motifs underly this—the parties, the conspicuous consumption, the transactional nature of relationships, the optometrist’s eyes, the valley of ashes.

Counterposed to these is the green light, which represents for Gatsby all that is unattainable and all that he has lost.

A fractal story grid

How might we apply these ideas to the construction of a story? This must start with the fundamental pattern that will be repeated and echoed. Let me take the premise of the novel I’m working on:  Sol must learn that boundaries do not provide safety, or he will never make it home.

With this, I can set up a grid: bounded vs unbounded and safe vs unsafe. This is the primary pattern that will repeat again and again at smaller and smaller scales.

Next, I identify some of the plot elements. Sol, the central character, is a clever boy who will grow to manhood over the course of the story. He will transgress boundaries and discover things. And he will struggle with his feelings of loss and detachment. So, within the basic grid I can place a further four elements: knowledge, activity, time and feeling.

And I can break each of these four elements down into four sub-elements, as you can see below. This approach is modified from the system promoted by Dramatica.

In each grid and sub-grid, the top left will have the valency of bounded and safe, the top right of bounded and unsafe, the bottom right unbounded and safe, and the bottom left unbounded and unsafe. So, for example under Feeling (which is predominantly unbounded and unsafe), secure is safe and bounded, while insecure is unsafe and unbounded. There is a dramatic tension between elements in each grid and sub-grid that are diagonally opposite each other. This tension will supply the rhythm of the story.

This is not the structure of the story. It’s a coding sheet. When the plot points are superimposed onto the grid, it shows the valency of each point and the connection with other points. This helps guide the sculpting of the piece to achieve the desired effect.

For example, in the beginning of the novel, Sol is evacuated during a war to his uncle’s house in the country. He desperately misses his parents and believes he has found the sign of a way to get back to them. So the story begins in the bottom right quadrant of feeling with a sense of detachment. He fails to bond with his uncle, Zand but, when he crosses the boundary wall of the estate, he encounters a beekeeper, Bernard, with whom he forges a relationship.

There is a recurring theme of boundaries. Zand tells him he is free to go anywhere in the house, except in the study. Sol is fascinated with the beehive with its many chambers. In the field by Bernard’s hives he discovers what he believes to be a shape under the ground.  We have moved to the top left set of quadrants dealing with knowledge.  Is the shape really there? Or is it an illusion? Can he reason it out, or will he need to apply the less secure means of intuition? The shape escapes the boundaries of the present takes us into the past (bottom left quadrants dealing with time) when he comes to believe he is looking at the  outline of the walls of an old Anglo-Saxon meeting place, a wintan. The assignation of the place to the past puts it in the sub-quadrant which is bounded and safe. It is the present which is unbounded and unsafe.

The final act of the book is where certainty vanishes. It takes place largely in the activity quadrant and is dominated by the contradiction between the discrete and the diffused.

New and different stories

So, it could be done. The question is why would anyone do it? Well, as a display of virtuosity perhaps. This is arguably the motivation behind David Mitchell’s Cloud Atlas. But, there are two more important reasons. The first is to give the reader the pleasurable experience of pattern and rhythm. The second, and most important, is that it carries the message of the book. The message is that there are no bounded spaces which are inherently safe for Sol. If the fractal pattern is self-similar all the way down to the smallest level of magnification, then no space is without peril. The advantage of considering fractals as a basis for story-telling is that it may open the door to new and different kinds of stories.

A final point. I’m not suggesting this as a grand scheme for the construction of stories. Grand schemes tend to lead to mechanical cloned stories. This is just the schema that I developed to help me tell this particular story.

We’re a long way from robots writing compelling stories, but software can make an author’s life simpler. There are three broad classes of software intended to help writers.

• Planners help you plan a story,
• Organisers help you write it,
• Editors help you edit it.

I list some of the main alternative here.

As I’ve said before, though I love software, I’m not a big fan of the Organisers for authors. I’m just as happy using an ordinary word-processor and a spreadsheet. So none of the programs in the Organiser category is in my toolkit and I haven’t attempted to test and review them. Remember, a computer program won’t write a single word of your novel for you. I do however recommend Beemgee from the Planner category and ProWritingAId from the Editor group. I’ve also tried Contour from the Planner group and it’s pretty good. I was less persuaded by the usefulness of Contour’s companion program, Persona, compared with Beemgee.

Planners

Most of the software in the organiser group below includes some planning functions. However, there are more dedicated planners

• Contour is aimed at screenwriters supporting them in developing an outline. It’s based on four key question: Who is the main character? What is the main character trying to accomplish? Who is trying to stop the main character? What happens if the main character fails? It costs US$39.95
• Persona is character-creation companion software to Contour. It’s based on archetypes and costs US$39.95
• Beemgee offers a very detailed series of prompts for defining character and plot outlines. I like the way Beemgee links plot to character. The basic version is free. The premium version with more functionality costs €59per year.
• Dramatica, like Contour, is based on a writing theory. In this case, the central idea is that “every complete story is a model of the mind’s problem solving process”. It costs US$99.95.
• Storyweaver is a simplified version of Dramatica. It costs US$29.95
• Snowflake Pro.  This is Randy Ingermanson’s famous method turned into software. It will cost you US$100. Or you can just read the ten steps of the Snowflake on his blog for free.

 

Organisers

This type of software acts as a word-processor and filing cabinet. They conveniently store your plot outline, timeline, notes on background research, and your stray snippets. Some include prompts for generating characters. All include a stripped-down and uncluttered word-processor. The particular advantage of word-processors for writers is that they’re designed for you to move chapters and chunks around easily. Scrivener is the best known of these.

• Scrivener
• PowerStructure
• Storyist
• Shaxpir

In addition some include templates or other devices to help you build your story arc.  I’d recommend treating this facility with some caution. Following templates may lead to mechanical stories.

• New Novelist contains pre-built templates for major genres
• Wavemaker includes a version of the Snowflake method
• Novel Factory
• WriteItNow includes eight templates for novels, screenplays, stage plays etc

For ease of presentation, I’ve listed all the features and the costs of the main products in a table.

Editors

In a previous blogpost I reviewed Fictionary (which implausibly claims to give your novel a structural assessment), and copy editors Autocrit and ProWritingAid, as well as some free alternatives.  I recommended ProWritingAid as best value for money.

 

The bottom line

Many of the story Planners can be copied out as question templates. So you could try the demo version, copy the questions, and create your own tool. There’s pretty much nothing in the Organiser group you can’t do with a word-processor and a spreadsheet, or indeed paper and pencil. Only the Editing tools depend on algorithms and databases you can’t copy.

I’m a geek. I love tech. But I’m not an easy sell. The world is full of artificial intelligence that isn’t as intelligent as you’d think. I previously put the literary analysis website Who do you write like?   to the test and found it could not accurately identify James Joyce as himself.

This month, I put two more literary tools under the spotlight—Fictionary and Autocrit. They’ve reinforced my view that, while algorithms are pretty good at copy-edit-level text analysis, they aren’t yet up to the job of structural analysis. I’m sticking with wetware for that task.

Fictionary

Fictionary is a web-based program that claims to provide a structural assessment of your book. According to its website it:

• Automates visualisation of your story arc
• Evaluates your story, scene-by-scene, against 38 story elements
• Guides you through an edit of plot, character, and scene
• Offers tips for rewrites

The service costs US$20 a month or $200 a year. I used the 14-day free trial.

Fictionary runs only on Google Chrome or Safari browsers and requires the Word docx format.

In my test, it felt buggy. The upload cut-off the first two chapters of my novel. This happened even on a second attempt. Manual corrections of some lists (such as characters) didn’t take.

The programme provided this visual of the arc for my novel.

For comparison, this is the story arc Fictionary generated for Dickens’ A Christmas Carol.

Pretty much the same. The story arc is a template, rather than an analysis of the text.

My own schematic of the chapters in my book looks nothing like this template arc. My cumulative diagram (below) is more like a roller coaster. It shows the number of stimulus/response pairs in each chapter (a proxy for dramatic intensity) and whether the protagonist’s emotion is positive or negative (a proxy for advances and setbacks).

This didn’t give me confidence that the algorithm was smart enough to parse my book. And it isn’t. The part indicated as the inciting incident isn’t the inciting incident and the climax isn’t the climax. You have to feed it with lots of coding. As you do so, the programme offers editing hints. These are all generic rules of thumb, rather than deep analysis of the text.  Examples are:

• What is the purpose of this scene?
• What type of scene is this (dialogue, thought, description, action)? Variety is important
• Anchor the beginning of a new scene so the reader doesn’t get lost
• Provide a hook for the scene

You have to divide your manuscript into scenes, code each scene, pick out your characters from a list of proper nouns, and provide lots of other information. I wasn’t convinced that this was any advance on doing the analysis myself.

 

Autocrit

AutoCrit analyses your work to identify areas for improvement, including pacing and momentum, dialogue, strong writing, word choice and repetition. You can also compare your composition to that of popular authors.

This screenshot shows the summary for my book.

This score of 80.52 is described as being in the 75-85 territory of best sellers, perhaps a little hard to believe since I haven’t finished editing. George R R Martin’s Game of Thrones gets a tally of 79.99. However, Autocrit is measuring something, since a much earlier draft of my novel achieved lower at 71.39.  Another novel, the one I’m most proud of, scored 84.96.

The indicators the program is using to create the scores are those in the bottom diagram: repetition, pacing, dialogue, word choice and “strong writing”. This latter category includes overuse of adverbs, consistency of tense, showing versus telling, clichés, redundancies, and filler words. Which means this isn’t really a structural analysis. It’s a copy editor with a beguiling summary screen.

 

Copy editors

If AI is not yet smart enough to perform a structural edit, it’s invaluable for the more mechanical copy editing process.

I’ve been using ProWritingAid since 2015 and I swear by it. ProWritingAid analyses your text and produces reports on areas such as overused words, writing style, sentence length, grammar and repeated words and phrases.

So, how does ProWritingAid compare with Autocrit?

• Autocrit ran significantly faster than ProWritingAid. The latter took two-and-a-half minutes to analyse 75,000 words, compared with around 30 seconds for Autocrit.
• Autocrit was not as effective at finding repeated phrases.
• Autocrit highlighted overuse of passive voice, whereas ProWritingAid found this to be well within target. On closer examination, Autocrit is using frequency of the verbs to be and to have as proxies for passive voice, and is thus less accurate here.
• Both programs flagged overuse of “filler” words. Autocrit found the main culprits to be “that”, “very”, “seem” and “really”. The list of overused words made editing easy and allowed me to bump up my style score in Autocrit from “too much” to “average”. But again, ProWritingAId is doing something more complex and my filler score dropped only slightly, from 47.1% to 46.8%. ProWritingAid suggests that no more than 40% of words should be “fillers”. In my defence, Shakespeare’s “to be or not to be” speech from Hamlet scores 53.4%.
• Autocrit gave me a pass for frequency of adverbs (which it measured at 11), whereas ProWritingAid flagged the 12 it found as borderline. Neither program states what an acceptable rate is, though the Hemingway program uses a threshold of under 1% of words.

• Both programs okayed my readability, though they produced different values for the Flesch reading ease score (83 in ProWriting Aid and 78 in Autocrit).
• Neither was very accurate at detecting show-versus-tell issues, which is unsurprising because word or sentence analysis is unlikely to be very sensitive on this problem.

Overall, the two programs have broadly similar features, though Autocrit is much more expensive. You get a year’s subscription to ProWriting Aid for the cost of two months with Autocrit.

The bottom-line verdict

• No algorithms are yet sophisticated enough, despite overblown marketing claims, to replace a human editor for structural edits.
• Copy editing of spelling, punctuation, grammar, and word-use can now be automated.
• Autocrit and ProWritingAId do a pretty good job of copy editing. Auotcrit’s uniqueness is the comparison with published fiction. ProWritingAid offers a more complex analysis and better value for money

Free Alternatives

Some of the features of the copy editing programs are available in free software. Grammarly and After the Deadline will review spelling and grammar. Hemingway will assess readability scores, and detect overuse of adverbs and passive voice.

Who can resist a personality test or a fortune teller? Writers are no exception. The online tool I Write Like promises to tell you which famous author your style most resembles. The tool works by Bayesian analysis, much like a spam filter on your e-mail. I tried it and was rewarded with the answer that I wrote like Vladimir Nabakov.

Then I tried it again. And again. Seventeen times with seventeen different stories. I got thirteen answers ranging from the flattering Tolstoy to the surprising Stephanie Meyer. Five times I got James Joyce. So I started to wonder what the tool was really responding to and analysed the five “Joycean” stories in more detail.

The stories didn’t have genre in common. Two were literary, one was humour, one a thriller and the final one a psychological flash story. Did they then have some stylistic similarity? I used the Hemingway app which measures lexical complexity and assigns a readability score. They averaged grade 4.2, but varied widely from grade 7 (more complex) to grade 2 (less complex).

They also had a higher average “lexical density” than is typical for fiction or than the non-“Joycean” stories. Lexical density is a measure of how many words in a text carry information (nouns, adjectives, verbs and adverbs) compared with non-informative grammatical words (such as articles, conjunctions, prepositions). Fiction typically has a lexical density of between 49% and 51%. Only 20% of the five “Joycean” stories fell within this range, but only 25% of the non-“Joycean” stories did either.

To check that the I Write Like website wasn’t just throwing out random names, I fed it the same texts on two different days. It gave the same answers, so it is measuring something. Then I ran the obvious test – I fed the tool text from James Joyce’s Portrait of the Artist as a Young Man. It identified this as being like Agatha Christie! To be fair, it identified Tolstoy’s Anna Karenina as being like Tolstoy.

In the course of this chase, I also looked at another literary analysis tool, the Online Authorship Attribution Tool. This is similar to the tools used by Universities to detect plagiarism in student essays. It compares three features of an unidentified text with known samples: the use of function words, such as “and” and “the” which are independent of content; punctuation; and lexical structure such as sentence length, word length and complexity of vocabulary. This tool failed to identify any of my “Joycean” stories as being like James Joyce. However, it also failed to identify chapter five of Portrait of the Artist as being by Joyce either.

I looked more closely at my five “Joycean” stories. They did have one thing in common – they all contained an extended monologue or internal dialogue. I have no way of knowing for sure whether this was what the tool was detecting. But it made sense, more sense than the idea that I have thirteen different styles. The story that was like Stephanie Meyer (author of the vampire romance Twilight series) was sci-fi and contained a hunt.

The moral of the investigation is: use these toys for fun by all means, but don’t take their readings any more seriously than you would a personality test or a horoscope in a magazine.

How do you find an agent or publishers who will accept manuscripts without an agent? How do you know which ones are scammers, or vanity publishers? When I was around fourteen or fifteen, I got a Writers and Artists Yearbook for Christmas. It was a hefty book even then, listing agents and publishers.

In the era of online databases, you can find much of this information on the Internet, and more. A lot of it is free. Authors Publish, which will e-mail you a useful update every week, is a resource I use a lot. Last week, they published a useful article on all aspects of submitting a manuscript, full of links to other resources.

There are guidelines on things like writing query letters and constructing pitches. There are also databases for finding agents and publishers and checking their credentials. I test-drove a couple of these. The results were disappointing.

AgentQuery is billed as a reputable search engine for agents. Also listed were Querytracker, and Publishers Marketplace. All are free, though Querytracker requires you to join. Querytracker seemed the most useful for finding agents, because it lists country. Both QueryTracker and Publishers Marketplace contain listings of who represents who – a useful feature if you decide your writing is like someone who is already published, though they failed to find the agents for the two authors I tried.

AgentQuery is the only one that allows you to search by genre, but not so helpful if you’re outside the US, since it seems to list only US agents. I tested it on agents for my current book, and it returned precisely none of the agents I had selected to pitch to, all of whom are in the UK. Query Tracker found two of five agents I selected for testing, while Publishers Marketplace found three. I wasn’t convinced any of these sites would replace my own diligent research.

Then there are some websites against which you can check the credentials of agents and publishers. Anyone can set up and agency or a publisher. Some are not very good, and some are scams. Anyone who charges a reading fee should probably be avoided. And some publishers are vanity publishers where you pay them to publish your work rather than the other way round.

Preditors and Editors identifies those who are not recommended, as requiring fees or offering vanity publishing. I tested the site on four small publishers, and it had no listing for any of them. Index to Agents, Publishers and Others is a community resource, driven by postings from users. It listed three of the four test publishers, though the information on one was out of date. You should bear in mind that user postings may or may not be accurate and dispassionate. Again. I felt it was probably better to rely on my own research.

With both agents and publishers, you can take a look at their websites. Ask yourself, do they look professional? Check out their authors – are they people with whom you’d feel in good company? Are they interested in your genre? Do they offer editorial and, in the case of the publishers, marketing and distribution? What are the royalty arrangements? Databases can help, but remember they are neither complete nor fully accurate.

Michael Crichton says “books aren’t written, they’re rewritten. Including your own. It’s one of the hardest things to accept, especially after the seventh rewrite hasn’t quite done it.”

As promised last week, I’m devoting this post to the feared and hated topic of revision and a great approach to revision that I learnt from the University of Iowa course that has now finished. The take-home messages were:

• Writing is only the beginning; where the real work and the real fun of it comes is in the rewriting
• Let time pass between edits – part of writing is not writing
• Build the rewrite in “layers” – focussing on one thing at a time for each edit
• Enjoy the problem-solving – failure leads to mastery – the more problems you solve, the better you get at it

I got really excited about a method for structuring the revision process, which I’m going to call the spine and ribs method. It was proposed by one of the teaching assistants, Christa Fraser. But this is my own adaptation of it, for whose faults no blame attaches to Christa. The method involves two passes, and each is structured as a set of questions. I’ll lay out the questions here. Then I’ll illustrate its use with the revision I made of a story about a shoe shine guy who has an encounter with a mysterious customer.
Pass One

1. What is the “soul” of the piece?
   What are you trying to convey and what concerns were working there of which you were not fully conscious?
2. Who are the main characters and what is the setting here?
3. What are the primary obsessions, preoccupations, desires, fears, obligations, etc. of our characters?

Once you’ve answered these questions, in what ways are these things driving or pulling the narrative forward? These are the long lines or spine of the story

4. What are some of the short lines of narrative tension that are already there?

Now that you have identified the spine of the narrative, are some of the existing short lines of narrative tension now incongruous with the re-aligned story? Are there new short lines, or ribs, that are opening up and wanting to align themselves along the spine of the story?

You revise the piece using these questions. After the story has been realigned so that you feel that you can see the true shape of the story as it wants to and ought to be, you return for Pass Two.

Pass Two
1. Record and adjust the timeline
The story’s timeline may extend earlier than the earliest point of present moment action and later than the last moment of present action. Can you explore references to the backstory, or hopes and plans that extend beyond the action?

2. Adjust the arc of understanding.
Are you explaining too much? Too little?

3. Polish the mirrors in the mirrored hall of infinity: moments that endlessly reflect
What are the motifs? Where do they repeat? Which parts should mirror each other? Which parts should be prefigured? This is where you play with structure.

4. Polish the burrs off
Read the piece aloud and listen to the rhythm. You’ll hear the false notes, unnatural syntax, incorrect words or phrases, repetitive elements, and anything else that will disrupt your reader’s experience of the narrative at a sound and language level.

5. Share the work with friends whose feedback you trust

6. Repeat all these steps until you think the story is ready

7. Put the story away for months.

8. Repeat all of the above until you think the story or longer work of fiction is ready for submission or publication.
The spine and ribs revision method applied

I used this method for the last assignment of the course, which was to revise a story we had already written. I revised a story about Horacio, a shoe-shine guy. Horacio has a simple moral worldview that good people are recognisable by the care they lavish on their shoes. He has an encounter with an enigmatic drifter, who he sees initially as a devil and then as an angel. An apparent miracle occurs which forces Horacio to question his morality.

I was really pleased and surprised by the result. In pass one, I came to recognise that in addition to Horacio’s worldview there was another idea lurking: an exploration of the way we try to fit events to our worldview. I considered whether I needed to give Horacio more of a backstory, to explain his morality. I also considered whether I needed to give a clearer explanation of the drifter, and whether he really was a messianic figure, or whether he was manipulating Horacio. The main conclusion was that I needed to give the story more room to breathe – to draw out the conflict between Horacio and the drifter. I also moved the devil/angel definition of the drifter from the narrator’s voice to Horacio’s voice.

In pass two, I had a lot of fun. At this point I layered in a lot more complexity that had been lurking in the story as subtext without me being aware of it. In particular, I saw that the events which challenge our worldview with good outcomes need not themselves be good.
I gave Horacio a backstory, tracing his morality to his flinty mother who scoured him of religious doubt. I considered, and rejected, extending to story to what happened after Horacio’s revelation.

The work on structure and motif in the second pass was really the glory of the thing for me, perhaps because playing with structure is the bit of revision I enjoy most. I recognised motifs not only of devils and angels, and good shoes/good men, but also of burnishing/scouring and aridity that I hadn’t been conscious of. I saw how I could connect them more clearly. I saw that I could prefigure the climax (hopefully, without it being obvious). I gave Horacio a more explicit revelation that good shoes did not necessarily imply a good person. I altered the ending so that it echoed the beginning. Interestingly, this changed the story but not the subtext. In the original version, Horacio leaves his shoe-shine stand and follows the drifter. In the revised version, he refuses. But this didn’t matter because it still underscored Horacio’s revelation.

I shared the rewrite with two other writers on the course before submitting it. One wanted a tighter structure and deletions, and the other a greater opening up. I tried two different approaches to structure and I now had two different versions of the piece! One was narratively more straightforward but eliminated some of the backstory. The other was more convoluted but denser. My head opted for the first version, my heart for the second. In the end, I went for the first version when I realised that my heart was attached to a piece of descriptive writing that didn’t really move the story forward.

I’m really pleased with the result. One reader said “you have deepened and expanded this into a lovely parable”, and another described it as a “richly intelligent read, full of a kind of tenderness”. Now I just need to leave it to infuse and mature in a deep recess of my hard drive before I return to it.