Let's zoom in and see what a day of the week chart looks like:
Let's zoom in further and look at the hours of the day. I'm based in the UK so I've used UTC time:
It would seem there are good and bad times to promote.
12:00 am (midnight) UTC means: In the UK people may be reading in bed, In the US they are getting home from work, In Australia and New Zealand it's lunchtime
5:00 pm UTC means: many people in the UK will be commuting home; The US is just staring lunch; Australia is asleep and New Zealand just about to wake up.
At the other end 2:00 am UTC: The UK is asleep apart from the person who read all 9 of our A Vested Interest series (1.2 million words) in a 4 day marathon; The US is travelling home from work; Australia is busy working and New Zealand has just finished lunch.
It does seem that 2:00 am and 7:00 am UTC are not the best times to schedule Tweets but not promoting then will mean you miss some sales.
What about weekends - Do you get different results?
Pretty much the same apart from 3:00 pm UTC when the UK and US are at work weekdays, Australia is going to bed and New Zealand is fast asleep. It seems weekdays at 3:00 pm is the worst time to promote.
Where and how did I get this data? I subscribe to http://www.novelrank.com This is a free service which will track book sales. I registered all my books and subscribed to their RSS feed. Each time a book is sold I get an email message with the details something like:
"Dark Secrets (A Vested Interest)" by Shelia Chapman, John Chapman (Kindle Edition) has sold 1 book on Amazon.co.uk on Monday, Jul 1, 2013 at 3am (GMT) Western Europe Time, London, Lisbon, Casablanca.I took all these emails for the last 6 months and tweeked them to fit a spreadsheet.
It has jumped to a new Amazon SalesRank of 21,492 from a previous SalesRank of 41,322.
If you do this your results won't be exactly the same of course. We write a series and give away the first book free. I think it takes the average reader 10-20 days to read the first free book - at least, when we do a promotion, that's how long it takes for the second book's sales to start rising.
We sell more books than the average indie author but still not an astonishing number - not enough to live on. To get a more accurate picture we should compare our results with a few moderate sellers and a few top 100 indie authors. That would be possible since you can track other author's sales too. I just don't have the time though. Anyone want to help?