|Ei blot til Lyst (#5): Ibsen and Strindberg||Talking Heads – Life During Wartime (live, Stop Making Sense) [#postpunk]|
For some time I’ve been looking for a way to analyze email traffic on my gmail account. There used to be a nice Chrome extension, but since it relied on harvesting the data from an iGoogle app it was closed down. Now however there is finally a reasonable way to do this. Gmail Meter is a Google Drive script that extracts raw spreadsheet data from your account. You can then receive monthly graphs and numbers about sent and received emails, word count, top senders and receivers, reply time, and much more. Because you also have access to the documents themselves, there are endless possibilities for fine-tuning the output.
Here is a small selection of facts about my personal email use.
I received 1487 emails in the month of March, from 551 unique senders. I replied to 11.8% of them. I sent a total of 328 emails to 107 people. Twitter sent me the most emails (although they are filtered so as to not appear in my inbox). I sent the most emails to Andreas Stokke (55 emails). Peak receiving time is 7am to 3pm. I receive 19% of the weekly emails on Mondays, but only 12% on Thursdays. I send 18% on Mondays, and only 9% on Thursdays. On Sundays I send 14%. I label or filter 93.8% of my incoming emails. 16% of my replies are within less than 5 min. 17% take more than a day. Over 45% of my emails have a word count of less than 30 words.
That is just the first month (including Easter). Later on I’ll have more long term data.
Needless to say, the data contains a lot of noise traffic. Spam is not included, but all filtered list emails are. Since I receive email notifications from social media and subscribe to a lot of lists, the numbers swell significantly. Obviously I only read a fragment of these emails. Fortunately, in the future I can purge the data by ruling out certain labels, such as label:emaillist and label:social. That will give a more accurate picture of the email traffic that is reply-worthy, so to speak. The filters can do much more. Because I filter based on topic, say, teaching or admin, I can learn a lot about the volume of various tasks.
But do I really want to see those numbers?