Comps -- sales figures

AFAIK, you need Nielsen BookScan to see sales figures, but it's expensive, as well as incomplete (library sales aren't included, for instance, nor probably academic sales, which would really matter for you).

I don't know of any other route to actual sales data, unless you can access something through academic channels (how many programs nationwide adopted textbook X, for instance). Perhaps you could speak to a few professors in that field, people who are responsible for choosing which textbook to adopt. Or find some articles about overall trends in foreign language instruction.

I've heard there are ways of translating Amazon rankings into sales, but I don't know how reliable they are, given that a ranking is one snapshot from the whole sales history. Reviews/adds on Goodreads and Amazon are also sometimes used as a way of guesstimating sales, but that's very rough guesstimating, because not everyone who buys reviews.

You don't really need to know sales figures for your comparison titles: just that they've done well, or that the agent you're submitting to liked them.

I've found - using a few data points from various authors - that the number of reviews on Goodreads can be a general proxy - multiply by 10 the number of ratings and you get something resembling copies sold. Caveats that young adult and recent books skew higher - more reader engagement online - and older books published from before the internet skew lower.

Versailles said:

I've found - using a few data points from various authors - that the number of reviews on Goodreads can be a general proxy - multiply by 10 the number of ratings and you get something resembling copies sold. Caveats that young adult and recent books skew higher - more reader engagement online - and older books published from before the internet skew lower.

Click to expand...

The only true numbers I can offer: A good friend of mine has 30 ratings/10 reviews on Goodreads, 55 reviews on Amazon. Her sales are around 4000. The book has been out for four months. With this as a template, each Goodread rating=133 sales...each Amazon review=72 sales.

gem1122 said:

The only true numbers I can offer: A good friend of mine has 30 ratings/10 reviews on Goodreads, 55 reviews on Amazon. Her sales are around 4000. The book has been out for four months. With this as a template, each Goodread rating=133 sales...each Amazon review=72 sales.

Click to expand...

That's interesting, because if I were to use my numbers, I'd say it was the opposite way around: each Amazon review represents more sales than each GR review or rating. This stuff definitely varies; I don't think there's a clear correlation. Also, if the book is trade published, you don't have a solid sales figure until you factor in bookstore returns, which takes a while.

I would go with Old Hack's advice: you don't need sales figures to figure out good comps. Better to get a general sense of what people are reading in this category/genre. If nothing else, you could look through Amazon's chart of the 100 current best sellers in your book's sub-category. (I'm obsessed with sales numbers myself, but for completely other reasons, and I wish I were less obsessed with them. )

Versailles said:

I've found - using a few data points from various authors - that the number of reviews on Goodreads can be a general proxy - multiply by 10 the number of ratings and you get something resembling copies sold. Caveats that young adult and recent books skew higher - more reader engagement online - and older books published from before the internet skew lower.

Click to expand...

I've just looked up a couple of the books I've written, which I have detailed sales figures for.

One of them has sold over 50,000 copies. It has 60 or so reviews on Goodreads, so using your formula it should have sold 600 copies. Another has sold over 30,000 copies, and has over 500 reviews on Goodreads, leading to a projected sales figure of 5,000 copies.

gem1122 said:

The only true numbers I can offer: A good friend of mine has 30 ratings/10 reviews on Goodreads, 55 reviews on Amazon. Her sales are around 4000. The book has been out for four months. With this as a template, each Goodread rating=133 sales...each Amazon review=72 sales.

Click to expand...

Another formula which works for the book you've looked at, but not for any other title.

Please, everyone. You can't use Goodreads or Amazon reviews, or Amazon sales rankings, to extrapolate sales figures. Each book is unique. If you magic up sales figures based on these things you're going to be wildly inaccurate, and are going to make yourself look like a banana if anyone checks your figures against Bookscan. If you can't get your hands on accurate figures, don't use them.

No, I meant to say RATINGS, not reviews. Multiply ratings by 10 and you'll find you are in the ballpark. Of course it's not completely accurate but its a good proxy.

Versailles said:

No, I meant to say RATINGS, not reviews. Multiply ratings by 10 and you'll find you are in the ballpark. Of course it's not completely accurate but its a good proxy.

Click to expand...

Sorry, that doesn't make any sense to me.

If nothing else, if you look at OH's example, and if we assume the rating on a book was 5 stars, then she'd have sold 50 copies, not, you know, 50,000.

Versailles said:

No, I meant to say RATINGS, not reviews. Multiply ratings by 10 and you'll find you are in the ballpark. Of course it's not completely accurate but its a good proxy.

Click to expand...

Your method is still not remotely accurate, I'm afraid.

I'm still curious why sales seem to matter to the OP and others. Every time I've provided comps, it's in the form of "This book is similar to book X and book Y but reaches an audience more along the line of book Z" or "This book would appeal to readers of book A and book B." Agents and publishers want to know what your book compares to and the audience it would serve, not what the competition sold. You need to use valid comps, but books with better popularity or sales make better comps. That still doesn't mean you need to know actual sales figures.

Heck, my publishers even have trouble figuring out actual sales figures for their own published books.

Jeff

If a book has 2,000 ratings on Goodreads, its a proxy for having sold 20,000 copies. i'm not talking about the rating on a scale out of 5, I'm talking about the NUMBER of ratings that readers have bothered to submit. Based on numerous data points, from various authors, published commercially in the last 10 years, with a skew towards young adult who are more active on social media, it actually is remotely accurate. Thanks.

Yeah and I still think that's not remotely accurate. It might be for your books but it is not, for example, for mine. My textbook sold 10x the estimate, my worst selling romance sold 1/8 of the estimate. Anything I put on Netgalley is under because Netgalley directly causes Goodreads ratings. In the absence of some kind of objective confirmation using a representative sample, it's a somewhat informed guess and not something to share with a publishing professional. It's like saying 1/100 Amazon readers leave a review, vaguely ballparkish in aggregate, useless for individual cases.

Who is sharing what with a publishing professional? All I did was share a ballpark method that I have found works in some cases with plenty of caveats. Vaguely ballparkish - exactly. But if it makes you feel better, please continue to pick it apart as though I presented it as a bullet-proof, scientifically accurate methodology for pinpointing sales. Sheesh.

Versailles said:

Who is sharing what with a publishing professional? All I did was share a ballpark method that I have found works in some cases with plenty of caveats. Vaguely ballparkish - exactly. But if it makes you feel better, please continue to pick it apart as though I presented it as a bullet-proof, scientifically accurate methodology for pinpointing sales. Sheesh.

Click to expand...

The original poster was asking about sales figures for books which were comparable to theirs--I assume to use in a query or pitch. So the figures were almost certainly to be presented to a publishing professional.

Your "ballpark method" is horribly inaccurate, and shouldn't be used.

And don't you dare snark at our members like that. It's not on. Go and read The Newbie Guide right now (there's a link at the top of the page) and remind yourself of our One Rule before you post here again.