A/B testing is a method of comparing two versions of an email to see which performs better on a specific metric, such as Open Rate, Click Rate, or Conversion Rate.
First of all, we have prepared a list with the most important features of
A/B testing in email marketing:
Compare target audience behavior
Validate the idea that converts best
Help with campaign optimization
Influence and facilitate decision making
Increase conversion
What can you test with A/B testing?
Email subjects
Commercial offers
The images
The content
Call-to-action links
The location of links
The date and time the campaigns were sent
Start by defining your goals
To begin with, it is very important to define the goal of the A/B testing campaign.
For example, goals can be of different forms:
Registrations for an event or a course
The acquisition of a product or service
Downloads of various documents
A reservation in the calendar
Landing Page Visualization
A/B testing – the example applied to product sales
Let's say you own an online store that sells clothing and uganda telegram you want to test two different offers to see which one converts better.
We suggest you create two variants, with different content to test.
Variant A
The offer for price reduction Variant B
The offer for free delivery
Subject : “20% off the entire fall collection” Subject : “Free shipping on all orders”
Content : Include an image of several items from the fall collection, highlight their quality and style, and offer a 20% off code for the first purchase. Content : Include eye-catching images of your most popular products and emphasize that anyone who orders will receive free shipping without the need for a discount code.
variant-a
variant-b
How to correctly set up the A/B testing campaign?
Choose a portion of your subscriber list to send variation A and another portion to send variation B.
Make sure that these two groups are comparable demographically or behaviorally.
Let’s say you have a subscriber list for your online clothing store, and that list includes people from different geographic regions and age groups.
To A/B test your offers, you need to make sure that group A and group B have a similar distribution of these demographic characteristics.
More concretely:
From a demographic perspective: If the majority of your subscribers are women aged 25-35, make sure that group A and group B have a similar proportion of women in this age category.
From a behavioral perspective: If you have subscribers who are frequent buyers and others who buy less often, try to make sure that both groups contain a similar proportion of each type of customer.
The main idea is to remove demographic or behavioral variables that might influence the results in an undesirable way.
This way, you can more accurately assess the impact of your offer on their behavior, knowing that any differences you see are more likely to be related to the offer and not other variable factors.
Send Variant A to the first party in the list and Variant B to the other party at the same time and on the same day to minimize differences in the timing of sending.How do you segment customer lists on the NewsMAN platform?
NewsMAN email marketing platform has advanced segmentation features, making it easy to define the audience for A/B testing.
Following the example below, you can segment the list according to several criteria:
gender
city
age
preferences produced
average order value
and much more
Also, the NewsMAN platform allows you to segment the list of subscribers, based on their actions in the newsletter or on the website.
pages viewed/product categories
click-type actions
“buy” or “add to cart” type actions
newsletter subscribers last month
and so on
Depending on the data you have about your subscriber list, you can start segmenting for A/B testing.
1. Fixed Subscriber Properties
Subscriber ID
Domain
Location by IP
Country by IP
Region by IP First
Name
Last Name
Stars 1-5
Last Activity
Last Click
Browser
Device
Operating System 2. Custom variables:
Name, first name, date of birth, gender, age, etc.
Subscription date
Subscription method
According to the percentage of the opening rate/total
According to the click rate of the newsletter/total percentage
He clicked on the link X
The subscriber is/is not in the segments
The subscriber received/did not receive the newsletter X
The subscriber opened/did not open the newsletter X
Remarketing
The subscriber completed the form X
To create a segment in the list, go to the section
→ Segments → Create → Name the segment (Segment Name) → Choose one or more rules in “Segment Rules” → Click “Add” to add the condition → Click “Save”.
→ After clicking “Save”, the subscriber download operation will start.
→ Once the segment is loaded, the wheel – indicating that the process is in progress – will disappear.
We have created several video tutorials for segmentation for you.
Determine how you will measure results
The next important step in running your test campaign properly is to define the metrics you will track and analyze based on your test results.
Below, we've identified the most important steps you could take.
email opens
clicks on certain links or buttons
email campaign traffic conversion
How do you determine which is A and which is B?
Which is which?
Well, if you fail to define from the beginning what you want to examine in variant A and vice versa, you will have problems defining the outcome.
You need to correctly determine, from the beginning, what each variant will be used for.
Version A is the control version, the one that keeps the original email.
Version B is the variation, that is, the email with one or more modifications based on the variable you want to test.
NewsMAN-variant-a
NewsMAN-variant-b
For example, if you want to see how your subject line is performing, you can adjust your subject line message with a sales-oriented call to action.
Version A Version B
[Valid today only] – 50% off your favorite products 50% off your favorite products
Make sure both versions are identical except for the variable you want to test.
ab testing example