I've rounded up ten infographics on email marketing, covering delivery times, email landing pages, mobile email and more.
Where possible, I've added the infographics to this post in a readable size, but for others you can click on the image to see a larger version...
The best and worst times to send emails (pure360)
Mobile email (eDialog)
Anatomy of an Apple email (Flowtown)
Email landing page checklist (unboounce)
Email delivery checklist (pure360)
2010 email statistics (metia)
Email irritations (ccloop)
What spam is worth (New Scientist)
An old one, but interesting. This spam campaign had a conversion rate of 0.000008%...
Evolution of email (Microsoft)
The science of social timing (KISSmetrics)
Reader comments (12)
9:20AM on 22nd August 2011
Can't read the Metia one, no link! :)
Handy collection though, thanks.
2:21PM on 23rd August 2011
That was a comprehensive and thoroughly informative visual post! I have shared the link on my LinkedIn stream. Plenty of takeaways. Thanks!
Regards
2:53PM on 23rd August 2011
Excellent round-up, though slightly amusing that the Pure360 graphic is missing the word 'to' from step 1. Perhaps step 0 should have been 'Have you check all the content for typos?'. :-)
2:56PM on 23rd August 2011
LOL... despite re-reading it twice before posting I knew I'd have a typo in that comment if I dared to suggest step 0 should be check for typos!
6:10PM on 23rd August 2011
test test
Founder at Zettasphere
6:12PM on 24th August 2011
Nice collection of good looking infographics.
Though I wonder whether the trend for pretty marketing purposed graphics has stopped authors thinking about quality of information.
The Pure360 'best and worst time to send' says 10pm to 9am is like throwing emails into a black hole. Whereas this is when Groupon send their emails. Their emails drive massive revenue.
5pm to 7pm is listed as the time for holiday emails. However in a structured time test for a travel company there was no such effect for this company (more details see http://blog.emailvision.com/eng/campaign-send-time-isnt-important).
Such a summarised infographic doesn't give enough insight and information that people need, it encourages a rule based approach to marketing that is not sensible.
I was pleased to see the eDialog graphic lists the source of information and size of survey. The joke goes 63% of statistics are made up. If sources/samples size and methods are not given how can readers trust the information?
Head of Marketing at Pure360
12:12PM on 25th August 2011
Glad you pointed out the pitfalls of testing Ben, we're all guilty of slipping now and then (now updated) ;)
Tim you make an interesting point and I'm glad you've raised it - one thing we are constantly telling people is that although there can be generic trends taken from campaigns, people should always test their own emails to see how their particular readership reacts. As we did not write this blog there was no reference to our sources - however you can check out the full guide it was based on at http://bit.ly/timings_guide_Pure360).
We of course appreciate the constructive feedback from a fellow ESP and will bear it in mind for our next infographic, always learning...
Founder at Zettasphere
8:10PM on 25th August 2011
Hello Abi, thank you for the source reference URL, showing the infographic was based on an average just under 20K emails from each of 34 companies (660,000 emails total).
I believe what's interesting on the timing issue is there is no rule, no consistency and little general guideance of value. As you state on your original post "Contrary to popular assumptions, the volume of marketing emails opened drops markedly during the lunch hour", whereas an analysis of 15 million B2C emails I undertook showed the opposite, in fact an open rise during lunch.
My recommendation is this, send time may not be at all important, if you want to know the true answer you must test your campaign. Without testing, picking a random time will be as good, statistically, as following timing based on the campaigns of others.
As an idea for you on the subject of infographics, given how the images are often pasted and re-published, then put a text reference to the *source post* in the graphic along with your home page URL (as you did)
3:07PM on 26th August 2011
Tim, that final comment is exactly right. All "image infographics" should include the URL of a HTML or PDF version.
This other version can repeat the data and have links to all the data sources and methodology used.
"Fact checking" is fundamental to journalism, so I think no reputable Website should publish an infographic which does not allow its facts to be checked easily.
Founder at Zettasphere
10:12PM on 1st September 2011
Another report, another view about send time.
"The findings suggest that email marketers who schedule their email campaigns to be delivered between 1 am and 5 am can expect higher volumes of email opens and clicks" taken from:
http://www.mediapost.com/publications/?fa=Articles.showArticle&art_aid=157560&mlfe=1
12:51PM on 4th January 2012
Tim, reference your "timing" email suggestions: What companies like Pure 360 and Emailvision (indeed almost all email providers) fail to recognise is that your lists are made up of hundreds of thousands of individuals all doing different things at different times. So rather than test your audience as a group, you should be sending emails to the recipients individually, at their best time to receive it. I believe that Silverpop and Lyris both have send time optimisation and have released many case studies about "the best time to send an email"
Founder at Zettasphere
2:32PM on 4th January 2012
Carl, I do agree. With the same concept as targeting content at an individual level then why not target timing at an individual level. I've seen the Silverpop case studies on this. Especially true if working across international timezones. Have you results from using this type of approach?
Part of what I'm saying is that in fact timing to the 'hour' is not as important as often considered, even at an individual level. So yes, try it, test it, be it at individual level, segment or whole database. From what I've seen it will not always be a significant factor. I've had info of a case using an individual level timed optimisation that was not significant, sadly only second hand information and I don't like to talk factually unless I've dived into the numbers myself.
Timing according to lifecycle stage, or triggered emails (the ultimate in timing) should be equally considered rather than just hour/day.
So I would encourage anyone to not use timing according to what worked for someone else but test it themselves and then move on to other things - re-testing every six months to check its still right.
Log in to post a comment