35 karma · joined June 6, 2014
For example, in 2012-2013 almost 40% of traffic on our Network was Youtube related, not just streaming, Youtube specifically! Obviously millions of these requests are going to be for the exact same video, say for example Gangnam Style. So instead of us pulling the data from Youtube every time a user requests that particular video we store that video on our Network, so when a user requests it (even for the first time), it actually comes from us. All my company needs to do is check for any changes to the specific page. The result is a better customer experience (faster streaming) but also far less network traffic (this literally saves millions).
Without the storing of IP addresses accessed we wouldn't have the information needed to determine what sites/data we should be cashing and what we don’t need to bother with and this sort of thing would be considerably harder to pull off.
The main reason for holding call data is for billing purposes, no e-mail information has ever been stored (other than in customer's own inboxes and sent items in their e-mail accounts). All browsing data is anonymised and only used for optimisation purposes.
Hundreds of thousands of pounds are spent annually on data storage, it’s welcoming that there is now a reason to reduce the amount of useless data we hold.