Electrosmart Digital Publishing BMAPS Digital Publishing
 
HOME
ESMART BLOG
 
PRINT PUBLISHING
EBOOK PUBLISHING
BOOK MARKETING
ONLINE PUBLISHING
ONLINE MARKETING
ONLINE BUSINESS
BUSINESS TOOLS
NEWS
ARTICLES
EVENTS
RESOURCES
SUPPORT
Sitemap
Members Vault
ESL Login

Latest Updates

IM News
-
Blog
-
Resources
-
Free Tools

Authorship

5 Steps to Being a Successful Author

Guidelines for Editing Non-Fiction Books

Book Formatting

Selecting Font Size and Type

Getting Published

Realistic Odds of Getting Published

7 Steps to Writing A Book Publishing Query

Tips on Writing Book Proposals The Way Editors Want

Publishing Ettiquette




website security

Understanding Amazon Sales Ranking

 

The algorithm Amazon uses to generate its sales ranking is as well guarded as Googles page ranking algorithm, hence the information below is at best, a guess.

Amazon defines its sales ranking system:

“The Sales Ranking system exhibits how books are selling. The lower the number, the higher the sales. The calculation is based on sales and is updated each hour to reflect recent and historical sales of every item sold. "

From what we can gather from our research:

Only the top 10,000 books are updated every hour

The ranking is not based on the actual number of books sold, rather it is based on a comparison against the sales figures of the other 9,999 books within that same hour.

The sale curve is extrapolated from a trending calculation - a book that held a ranking of 1,000 at 3pm and 2,000 at 4pm, is 'expected' to have a ranking of 3,000 ranking at 5pm, regardless of how many books it actually sells between 4pm and 5pm.

Books with rankings between 10,000 and 100,000 are recalculated once a day. The calculations use both historic sales and current trend projections. Newly-released books tend to outrank older established titles, even though the actual sales figures for the latter far exceed the former. This similar characteristic occurs in Google Search Rankings.

Books with rankings over 100,000 are recalculated every day - based on historic sales and projections, however, historical data has less weighting. Sales projections and trending are seen as more indicative, hence a books ranking can jump from 750,000 to 225,000 within 24 hours. With typically low sales figures in this range, even a single sale can escalate a book above 525,000 others.

This change in weighting between current sales, sales predictions and historical sales between the ranking groups makes it very difficult for big sellers to maintain a high legitimate ranking. This can provide a very skewed perspective for those who do not understand the system used [that means all of us except Amazon].

It frustrates me that the algorithm is not published in more detail so those of us wanting to use the statistics for market research can rely on the information communicated. Instead, to get an average ranking we need to track and calculate:

  • Average Daily Ranking - For a top 10,000 book = Rnaking every hour for 24 hours divided by 24
  • Average Weekly Ranking - For a top 100,000 book = Ranking every day for 7 days divided by 7.

From the average ranking, an estimate of the sales volumes can be extracted:

An average ranking guide of sales:

  • 2,000,000 - a single copy has been ordered.
  • 1,000,000 - total sales under 40.
  • 100,00 - total sales under 200.
  • 10,000 - 1 to 10 copies sold per week.
  • 1,000 - 10 to 100 copies sold per week.
  • 100 - 100 to 200 copies sold per week.
  • 10 - 200 to 1000 copies sold per week.
  • Top 10 - over 1,000 copies sold per week

Next: PLR Tactics

Back To Top

Publishing Index | Market Research | Self Publishing | On Demand Publishing | Comparing POD Models | Book Components | Book Format | Completion Tips | Proofreading and Editing | Professional Services | Preparing For Printing | Marketing & Sales | On-Demand Publishers | Self Publishers | PLR Tactics | EPublishing Tools | RSS | Webmaster Tools | Article Directories | Website Publishing | Google Updates