For kicks i’ve been tracking the amazon ranking # for the art of pm. For most of this week it’s been hovering below 1000. Right now it’s in the 400s. Which I understand to be fantastic. 2o minutes into the marathon, I’m out of breath, but still on camera and waving.

I’m trying to figure out if there’s some way to map rankings to sales figures. I mean, even the NYT bestseller list is entirely misleading. If Book X is in the top twenty for two weeks, but drops quickly after a month, its total sales will be less than a book that never makes the NYT bestseller, but hovers at the equivalent of a top 50 or 100 ranking for years.

So the same goes for the amazon ranking. It’s hard to map that number to actual sales, and it’s hard to connect the ranking number to book quality. The best book in a category is unlikely to stay highly ranked in hourly sales figures, but would score very well in lifetime sales ranking for its category.

For example, Mythical Man month is currently ranked #1386, but is the best known book on software project management ever. Its lifetime sales per category is probably #1.

Now amazon does track per category top 20 lists – but what i can’t figure out is its correlation to the master amazon rankings. I’ve seen the #5 book in a category have a higher amazon sales ranking than the #2. Not sure if this is a timing issue – those stats are updated at different times, or if there’s some funky math in what constitutes a sale in a given category.

And of course, the bigger question that renders all this number watching useless: what percentage of total book sales are through Amazon? I have no idea. I suspect it’s higher for tech-sector books, but what is the baseline? Do the amazon trends jive with other online sellers? physical in store purchases? I haven’t found anyone who’s written about or researched this. All these questions are a big motivator to not spend much time watching little numbers go up and down.