How Blubrry Statistics are Calculated
Blubrry Podcast Statistics provide accurate measurements of media-file downloads. Other stats services (like web-hosting analytics) are not designed to specifically measure these kinds of downloads. Blubrry Statistics are first calculated by unique IP address, which is easy to identify and audit. Then, Blubrry Statistics further analyze each countable download based on user agent and geographical information. The final results exclude bots, crawlers, and multiple downloads from the same network location. The final calculation is then adjusted to compensate for known large-network locations, like universities and businesses, that have multiple users under the same IP address.
Blubrry Statistics are for podcasters and advertisers who seek an accurate picture of how podcast media is being consumed. By using a unique Redirect URL in front of a media-file URL, Blubrry is able to record each time a media file is requested. Through internal algorithms and processes, Blubrry will count valid requests as downloads and ignore erroneous requests from bots, web crawlers, and other illegitimate sources.
Let’s take a look at some common stats systems and how they compare to Blubrry Statistics.
FeedBurner/Feedly Statistics and Podcasting
FeedBurner’s statistics measure a different unit than Blubrry Statistics. Blubrry measure media-file downloads. FeedBurner measures feed subscribers based on how many devices were subscribed to a FeedBurner RSS feed in the last 24-48 hours. FeedBurner doesn’t count how many media-file downloads have occurred through your website or your podcast RSS feed.
Since about 2009, cloud-based services like Feedly began combining multiple subscribers into a unified system. This system will pull one feed from a single source for all of those users. This cloud-based system has led to poor statistics reporting from Feedly. Due to these limitations, there is no ideal way to measure feed subscriptions. Blubrry does not recommend using FeedBurner or Feedly stats, since those services will report much lower numbers than are actually accurate.
Web Server Statistics and Podcasting
Web stats packages, such as Webalizer or AWStats, process your website logs to make that data readable. These services record ALL requests for all files on your server, with no consideration for how those files are consumed. Web statistics will count bots, spiders, and multiple requests from the same device as “individual” requests. This leads to a RAW total of how many times a media file was requested, producing greatly inflated download numbers. Web statistics will give you an accurate measurement on how much bandwidth your server uses, but it will not accurately tell you who is consuming your media from what device.
Shortened URL Hit Counters and Podcasting
Services such as Bitly, TinyURL, and goo.gl provide a shortened URL intended only for web browsers. These services report “hit counts”, which refer to the number of times a web browser navigates to a unique URL. Hit counts are notoriously inflated when compared to Blubrry Statistics download totals since hit counts do not factor in the multiple hits that occur during file downloads.
Shortened URLs to media files should be avoided at all times. Such URLs remove the file extension from the URL, which is required for Apple Podcasts and most other podcast-consumption apps. Furthermore, such URLs will be crawled by internet search engines. These engines will send unnecessary web crawler traffic to media files behind shortened URLs.