This Monday, Digital Book World announced that they will be publishing weekly e-book bestseller lists every Monday on their site and in their daily newsletter. When it comes to e-books we’re all quite hungry know just what exactly is going on so it’s great to see DBW injecting unique insight into sales trends. I say unique because these lists rank titles within price bands, not subjects or formats like most print lists, making it easier to define correlations between title rank and price. Have a look at the price-banded lists to see what I mean. They also have an overall Top 25 across all price bands.
Because the DBW lists are not based on retailer-supplied raw sales data, I thought it would be a good exercise to understand how the lists are generated, what they can and can’t show us, and what questions remain open.
Methodology
In a post detailing their methodology, Dan Lubyte, founder of Iobyte Solutions, who compiles these lists for DBW, talks about how each title with a bestseller rank on a retailer site is scored and ranked. It’s fantastic to see this sort of transparency and I encourage you to read it. Titles are ranked by aggregating their daily rank as seen across five retailer websites, and are given additional weight based on retailer market share and prior rankings. Approximate retailer market share is determined by the editors of DBW and Iobyte Solutions. I do think we need more insight into how retailer market share is determined to solidify the usefulness of this type of rankings analysis.
What the DBW E-Book Bestsellers can show you
- A combination of daily Amazon, Nook, Google, Kobo and Sony top seller rankings weighted according to retailer market share and prior rank.
- Title rankings differentiated by price bands so that $0.99 books appear separately from books that are $12+.
- Publisher for each title.
- The lowest price for any given title seen during the past week (titles are placed in a price band based on this lowest price).
- Correlations between price changes and rank changes for ranked titles.
- Ranks that take into account seven days of rankings.
What the DBW E-Book Bestsellers cannot show you
- Rankings based solely on volume of sales.
- Rankings inclusive of Apple’s top seller rankings.
Open questions
- Are the rankings USA-specific? As those in Canada know, Amazon’s American Kindle store is the only way we can purchase Kindle books. Do these sales affect their Amazon rank?
- Does the $0 - $2.99 price band really include free books or should it be read as $0.01 - $2.99?
- How should differences in retailer ranking algorithms be accounted for? OK, this isn’t answerable without knowing the secret sauce recipe but I wanted to make sure everyone remembers why the fight to obtain retailer unit sales data is a good one – so we don’t have to ask this question to begin with.
In the absence of sales volume data, the DBW e-book bestsellers lists provide rankings that are based on rational analysis of the information we do have. That’s commendable and could bear interesting fruit as rankings change over time once the prior-rank weightings come into play. This week’s lists already make it clear that the Big Six don’t always price high. They make up five of the top ten bestsellers in the $0 - $2.99 range. There are only two self-published authors seen and they too fall in the $0 - $2.99 range. The average price in the Top 25 is $10.17.
