Saturday, December 31, 2005

Titles Revisited

LuLu, a service for digital do-it-yourselfers who want to publish their own books, music, etc, recently issued this press release describing a study they commissioned to determine the common elements of bestselling novel titles—the implication being that as clothes make the man, titles make the novel. To quote from the release:

Of the 11 variables studied, three were found to be key ‘differentiators’ between bestsellers and non-bestsellers:

  • Whether the title is literal or figurative
  • The word type of the first word
  • The title’s grammar pattern
A scoring system based on the study was developed to predict how likely a particular title would be to produce a bestseller, with .83 (i.e., 83% probability) being the highest available score. Again quoting from the study, here are the ten top-scoring novels that have been NYT number one bestsellers:

  1. Sleeping Murder by Agatha Christie 0.83
  2. Something of Value by Robert Ruark 0.80
  3. Looking for Mr. Goodbar by Judith Rossner 0.80
  4. Presumed Innocent by Scott Turow 0.80
  5. Everything’s Eventual by Stephen King 0.80
  6. Rising Sun by Michael Crichton 0.80
  7. Smiley’s People by John le Carré 0.77
  8. Three Fates by Nora Roberts 0.77
  9. Four Blind Mice by James Patterson 0.77
  10. Valhalla Rising by Clive Cussler 0.72
The good folks at LuLu have taken things a step further by providing a scoring utility on their web site for prospective authors to test their titles. With Candy from Strangers coming out in September, of course I couldn’t pass up an opportunity to validate what I’d always assumed was a great title.

Guess what? Turns out it’s not so great after all. In fact, it’s pretty damn bad. Candy from Strangers is a little tricky to categorize on the key multiple choice question the utility has for “Title grammar type,” but after convincing myself the correct selection was “phrase with no verb,” I discovered that the utility thinks that there’s only a 26.3% chance that the book will be a bestseller!

Not that that seems so very bad out of context. If you had told me a week ago there was a better than one chance in four that Candy would be a bestseller, I would have been ecstatic. But now that I understand how much rides on the title selection, I figured I owed it to myself to run some alternatives through the utility to maximize my chances.

The Challengers

I decided to stick with titles that alluded to the idea of receiving gifts, possibly from strangers, as a result of putting a wish list on a web site. I also threw a few that were variants of the old saws, “Don’t look a gift horse in the mouth” and “Be careful what you wish for” since they had a gift or a wish theme and the implication that either could be less than perfect.

Here is the list of new, improved titles I came up with and their associated scores on the LuLu title-o-meter:

  • Ride a Gift Horse – 44.2
  • Be Careful What You Wish For – 44.2
  • Wish List – 59.3
  • Gift Horses – 59.3
  • To Ride a Gift Horse – 63.7
  • Strangers’ Candy – 69
  • A Gift Horse’s Mouth – 69
  • List of Wishes – 69
  • Bite of the Gift Horse – 69
  • Caroline’s Wish List – 76.9
  • Her Gift Horse – 83.1
  • Her Wish List – 83.1
You’ll see that the very best titles apparently begin with a possessive pronoun followed by a noun modifying a noun. Turns out you can also have an adjective modifying the final noun, so My Golden Booger also scores 83.1, assuming it’s a figurative reference. (Feel free to pick that one up—so to speak—for your own novel.)

I was a little surprised To Ride a Gift Horse didn’t do better because the press release mentioned that To Kill a Mockingbird had scored 80 and the construction seemed exactly the same, but either the utility has a bug or I didn’t understand how to properly categorize it to achieve the same score.

Am I going to change my title? It’s not entirely my call of course because those decisions are made in conjunction with (or sometimes exclusively by) your publisher, but maybe it’s worth another discussion with Ben since Candy from Strangers did score so poorly, and as I mentioned in an earlier post, there is an issue with other books sharing the same title. (Who would have thought such a bad title would be so popular!) However, I don’t think anyone—even the LuLu folks—really believe a title by itself can make or break a book. To paraphrase some of my pedantic software developer friends, a good title is a necessary but not sufficient condition for a bestselling book.

And, as an aside, the one thing that strikes me as fundamentally flawed with the rating system (apart from the whole idea that title by itself can have so much influence) is the emphasis the system seems to place on the question of whether the title is figurative or literal.

Since the reader, in many cases, cannot know whether a title is figurative or literal until after they’ve read the book, how can that factor exert so much (apparently subliminal) influence on their decision to buy?

LuLu Bestsellers Scored

The LuLu site sells books by LuLu authors and can be made to display those books in bestselling order. Just for grins, I’m going to leave you with the top ten list of bestselling LuLu mysteries (1-10) with their associated title-o-meter scoring. (And, yes, I’ve had to make some assumptions about whether the titles are figurative or not.)

  1. Freaknic! – 35.9
  2. COLLU$ION – 31.7
  3. Consequences – 31.7
  4. Cancer in the Family – 26.3
  5. Clash of the Figments – 69
  6. A Brand You Can Count On – 63.7
  7. A Life of Crime – 10.2
  8. A Sister's Secret – 41.4
  9. Crooks and Creatures – 35.9


Post a Comment

<< Home