In email marketing, everything is simple: either your letters sell, or they don’t. Thoughtful design, high-quality text and visual content, time and service of sending, various marketing tricks - all these are conditions that will ultimately influence your reader, or rather, their conversion action. But even their observance does not guarantee a positive result. The solution is as follows: carefully prepare the ground, i.e. think through all possible scenarios and then competently evaluate the results. And how to do this inemail marketing— let's look at the example of one of the popular domestic serviceseSputnik.
A/B testing and automation of mailings: check everything! - image
Educational program: understanding the concepts and highlighting the main points
The title of the article uses the term “A/B testing” (English version of split testing), which is a cliche for all marketers, which in a broad sense can be explained as quality control based on comparison. The main rule for conducting anytesting— these are the same conditions (time, respondent base and their mood), which is quite difficult to usa student mobile number lead achieve in email marketing. Even A/B testing of one-time email subjects is in fact ineffective: the results obtained on Wednesday morning are no longer suitable for sending on Thursday evening — the conditions have changed, which means it is impossible to predict the outcome of new series of emails.
If you want to conduct A/B testing and use the results now, analyze the data after sending automated mailings with trigger letters that come after target actions (greetings, thanks, abandoned carts, reviews and other autoresponder letters). A completed target action is a specific condition that allows you to conduct research on different formats:
arrangement of text and graphic elements (button color, banner location, etc.);
testing the sending time and measuring the effect (for example, a letter following a review left or two days/week after a purchase, etc.);
use of multiple communication channels:mailing list+SMS/ letter+push-message etc.
A/B Testing and Email Automation: Test Everything! - 2 - image
The more conditions you take into account when creating a script, the less opportunity the subscriber will have to “jump off”. Work out all the details — you will set the “rhythm” for other factors. To test the effectiveness of automated emails, you can experiment with the following conditions:
topic and length
time
message - its structure and format
This is a base, a minimum, which allows you to get to know your audience's preferences better. But today, this is not enough for effective email marketing - you need to change the logic of readers' thinking and "lead them" at every step. To do this, you need to take a global approach to your mailings and think through any user behavior scenario, emphasizing an individual approach in each letter with the help of dynamic content. Specifying a name, city or phone number is not enough, write down all possible conditions (registration in the database, geography, previously completed actions - viewing products or leaving a review, etc.). The more consistent you are in the details, the more useful the data obtained will be. The main thing is not to be afraid of experiments - they will definitely lead you to effective mailings! Dmitry Kudrenko, CEO of the e-marketing service eSputnik, talks about this in more detail:Here.
1 email marketing
This is research first and foremost, so test every aspect of your automated email script, change the placement of elements, and their effectiveness.
2 test results
This is not a guide to action, but a food for thought for further improvements.
3 write letters for people
Write in such a way that you yourself would not send them to Spam, but would read them and even buy something
A/B Testing and Email Automation: Test Everything! - 3 - image
Publication date: 10/22/2018
