|
A/B List
Splitting A/B List
Splitting is the act of sending half of your
subscribers one newsletter and half of them
another newsletter then comparing if one
performed better than the other. For
example, you might vary the subject line, you
might vary the content, or you might simply move
the content around within the same newsletter.
You might also vary the time of day or day of
week on which you send the email marketing
campaign.
Here are
some of the more common things you might vary in
your A/B list
splitting testing:
- Different
email subject line
- Varying your
landing pages
- Modifying
the creative including moving things around
-
Personalizing your messages with name and
other personal information
- Varying the
email message "From Name"
- Comparing
HTML versus TEXT newsletter results
- Testing
short copy versus lengthy copy
- Sending on a
different day or time of day
Below are some
instructions on implementing your own A/B List
Splitting in eNewsletter Pro...
|
Step |
Suggestions |
|
Establish a campaign to use as a benchmark
(Campaign A) |
Testing
without any benchmark is a waste of time.
We suggest you find your best performing
campaign to date and use it as your
benchmark. You will simply want to
copy the newsletter and update it with your
most recent content. |
|
Determine what changes you will make and use
as Campaign B |
The most
important thing to remember in A/B list
splitting is to make your changes
one-by-one. Changing multiple
variables at once will not allow you
determine which change affected the
performance. Instead of changing your
logo, changing the subject and moving things
all around in just one campaign, space these
changes over separate email campaigns.
That way each change will be isolated and
traceable. A list of common changes is
shown at the top of this email. |
|
Split
your list |
Now that
you have a benchmark newsletter and have
created a modified version, you are ready to
split your list and send both Campaign A and
B. eNewsletter Pro 5 and higher offers
a simple way to split your lists. This
option is found during the send process.
The option is titled "List Splitting".
This option allows you to maintain your
subscribers on just one list but
consistently split the delivery into two
groups.
Although multiple options are offered, we
recommend using the Odd/Even subscribers
option. This option selects
subscribers based on their SubscriberID
which is an incremental number created
automatically by eNewsletter Pro. This
option produces a more random set of results
by essentially choosing every other
subscriber.
The
other option is to select the first or
second half of subscribers. When
sending eNewsletter Pro always sorts by
SubscriberID ASC. That means the first
half of the list always represents those
subscribers who have been members the
longest and the second half represents newer
subscribers. This option is therefore
less random then the first option.
So,
you will now send your benchmark newsletter
(Campaign A) to one half of the list such as
the "Odd Subscribers" and the modified
version to the "Even Subscribers" (Campaign
B). |
|
Analyze the Results |
Now that
you have delivered two campaigns, you will
want to compare the results and determine if
your changes have improved performance.
Pay particular attention to any changes in
click rates, views, and if you are measuring
ROI if there have been any changes in sales
or leads. Any changes in bounce rates
will also be of interest especially if
changing the "From Name". |
|
Make
additional gradual changes and repeat the
process |
If the
campaign changes were successful, you might
now make Campaign B your new benchmark
campaign. Now when you send your next
newsletter make an additional change to send
to your B list and repeat the whole process
to see if these changes enhance the
performance. |
|
