A study recently came out that went almost completely unnoticed. Nevertheless, it raises a crucial point. The practice of personalizing email with names/places may be detrimental: decreased reading rates and inbox placement and increased spam complaints. Seems rather counter-intuitive, doesn’t it? Let’s try to figure out what’s going on here.

[STUDY] Email Personalization- How Good Is It For Your Deliverability?

The terms of the study: the evolution of email marketing over 5 years

This study was conducted by Return Path, the world leader in the field of email intelligence. It involved more than 60 brands (Disney, Expedia, Nike, etc.) and was conducted in two stages, in 2008 and then in summer 2013. The initial aim was to compare the evolution of practices in email marketing.

The most striking change between 2008 and 2013 regards the collection of customer data: in 2008, more than half of the brands asked for very extensive information, such as a complete mailing address. In 2013, a third of brands simply request an email address with a zip code, or maybe a just name. Marketers require less information than before.

Among those that collect additional data, only 22% take advantage of it by personalizing messages with the name and/or location. Hence the question: what are we seeing here? Negligence or a clever tactic? To answer this question, Return Path compared the performances with or without personalization.

Personalization may be bad for your emails

Surprise: the more an email is customized with a name or place, the more frequently it lands in spam or the recipient marks it as spam, and the less they actually read the message. Catastrophic. Besides the time spent to manipulate custom fields, the marketer would also lose out on the final results.

[STUDY] Email Personalization- How Good Is It For Your Deliverability?2

 

(Source Return Path, The Email Subscriber Experience, 2008-2013)

As you know, in the future, Mailjet will offer the possibility to personalize your emails with merge tags. We have no intention of overriding this feature: here’s why.

1- A study is not an absolute truth

This study focuses on tens of millions of emails and a lot of brands, but that does not mean it reflects an absolute truth. This kind of work helps to make decisions, but your particular situation may be different. Return Path also explains this quite well. On the other hand, you need to ask yourself the right questions: if you spend time and energy to send emails with custom name or location fields, did you make sure you’re gaining performance?

2- There’s personalization, and then there’s personalization…

If we attempt to explain these figures, it is quite easy to image the following psychological reaction: when a recipient sees his first or last name or residence, he might feel attacked. He forgets that he supplied this information and has a negative reaction. But we shouldn’t confuse personalization of names/places with …personalized content.“

One thing we strongly recommend at Mailjet is always to send more and more relevant emails. Our interface and our API allow you to know exactly what links have been clicked. In consequence, our users can create segments based on their contacts’ areas of interest. From there, it is possible to send messages that better meet the needs of your recipients. Or even more simple: you can just personalize the subject and increase your opening rates in this way.

Basic example: an online sporting goods shop might notice that one segment reacts to all things related to …cycling,“ while another reacts to all things related to …skiing.“ Very simply, it’s possible to send them the same newsletter except for the subject line, to better capture the attention of each segment: …Flash sale: discount skis“ and …Flash sale: discount bikes.“ The content of the email must obviously integrate both of these objects. This customization is as simple as it is powerful.

Conclusion

The key to success is really just to send messages that are legitimate and desired by the recipient. Personalization of the content seems to be what matters, so run your own tests!

P.S.

This study also dealt with other practices: double opt-in, unsubscribes, etc. We recommend you give it a read; you can download it for free here.

[ Posted Fri, 04 Oct 2013 12:48:21 ]