JH Various real-time graphs are available during the sending of a broadcast and when completed.
To access those graphs, choose Reports from the Choose action… drop-down menu on the Broadcasts list screen.
The available graphs are:
Depending on the settings of your email systems as defined in Setup / Email Systems, broadcasts may be sent over more than one run of the scheduled task.
The Broadcast Log screen shows the progress of the broadcast as it goes out across those sessions.
To access the broadcast log, choose Show log from the Choose action… drop-down menu on the Broadcasts list screen.
You can send yourself a test of a broadcast (and any of its variants) to check its content and layout prior to setting its final scheduling.
Choose Test from the Choose action… drop-down menu on the Broadcasts (or Broadcast Variants) list screen.
“Send yourself a test while the broadcast is still in draft to check content, layout and formatting”
Every broadcast can have any number of variants so that you can test out different aspects of a broadcast to compare their success. When a variant is first added to a broadcast, it is an exact copy of the original broadcast. It can then be tweaked to test items such as:
#V2 Exclusive features – A/B Statistical Significance testing and improved reporting. See below.
To manage the variants of a broadcast, choose the Variants option from the Choose action… drop-down menu or click the number under the Variants column.
“Use variants to test various aspects of broadcasts against each other”
Once you have done so, this screen will appear:
To create your first split-test variant click on Add a variant to this message from the I want to… drop-down menu. This creates a copy of the original message as your first variant and opens up the message editor for the variant you have just created.
In the example image below, we changed the Subject line so we can split test the variant against the original.
You can change any aspect of the message. Once you have made your changes, click the Save Variant button.
You will then be redirected back to the Broadcast Variants list screen, which now shows the variant you just created.
In the image below, take a look at the Frequency column.
You can change the percentage of contacts to be sent each message using the slider in the Frequency column, as pointed out below.
The system decides at random during the sending process which message will be sent to each contact.
What does ‘Statistically significant’ mean?
‘Statistically significant‘ means that the result is unlikely to be explained solely by chance or random factors. i.e. you can rely on the result more than if it is not statistically significant. There are factors and considerations where this is not the case which are outlined below.
Set up your split tests in the same way as outlined above. The benefit of using #V2 will become apparent once your test has completed and you are looking at the results.
#V2 reporting will also display the words xx.x% Significant or put another way the level of confidence you can have in the winning result.
The winning test variant will be highlighted in green only when the statistical significance of the test result is over 90%.
In the image above, the level of confidence that the ‘winner’ will improve conversions, compared to the original message, is 94.6%.
A/B Testing
Below we can see two images that relate to simple A/B testing.
In the image above we see that the second option has the higher open rate (81.8%). While the open rate of option 2 is better than the original, the statistical significance of the result is less than 90% (87.7%). Compare that to the image below…
In the image above we can see that again option 2 beat the control (option 1) with an open rate of 62.5% compared to option 1 (12.5%). In this case the statistical significance was greater than 90% and #V2 highlighted the winner in green.
Note of caution: Any statistician will point out that there needs to be enough data (sends) for the test to be 100% reliable. Generally speaking the more you send the more reliable the result and the results of split tests sent to less than 250 contacts should be considered ‘indicators’.
Ultimately, a winning subject line is a winning subject line and providing you send another email with that subject line to like-for-like contacts, you can reasonably expect a winning subject line to be more effective. That expectation increases once you know that it is statistically significant.
The options for creating and editing a broadcast are identical. Note that you cannot edit a running broadcast or one that has already been sent.
To create a new broadcast, choose Broadcasts / Schedule a Broadcast from the menu bar:
… or choose Create and Schedule a broadcast from the I want to… menu on the broadcast queue screen:
To edit an existing broadcast, choose Edit from the Choose action… drop-down menu.
“Use segments to precisely target the contacts to receive your broadcasts”
Note that the date and time to send a broadcast is based on the server date and time with the timezone applied as set in Setup / System Settings / System / General.
“If you want to send your broadcast only to contacts who have confirmed a subscription, check that box in the Omit These Contacts section”
The Omit these contacts section in #V2 has additional options that greatly improve your capability to send precisely targeted messages based on your contact’s engagement with emails you have previously sent.
The additional options are:
Placing a checkmark in the box of the chosen option opens up a new section allowing you to select from the available choices.
In the example above, the broadcast will be sent to all active contacts who did NOT open the selected previous broadcasts (e.g., sending to unopeners).
“Use the Layout section to add pre- created headers, footers and links”
In this section, you can opt to share your broadcast by using the system’s Broadcast Archives functionality.
All sent and unsent broadcasts are held in the broadcast queue which is a central place where you manage all your broadcasts.
The above image shows the broadcast queue with three (3) broadcasts, one of which is not yet approved for sending.
From the menu bar, choose Broadcasts / Show Queue to view the broadcast queue.
The columns of data in the broadcast queue are:
Note that the Target count doesn’t exclude contacts that are not sent the broadcast at the moment of sending for reasons such as:
There is a setting in a broadcast that omits contacts whose subscriptions are all cancelled and whose subscriptions have all failed. Broadcasts will not be sent to these contacts.
Using the Order by drop-down menu, the queue can be sorted by:
Using the With selected… menu, you can approve/disapprove or delete the selected broadcasts.
The Filter drop-down menu allows you to filter your list by these criteria:
Using the I want to… menu, you can:
“The options in the actions column of the broadcast queue change depending on the status of the broadcast”
The Choose action… menu has the following options:
Different options may appear at other times, such as when a broadcast is being sent.
arpReach has a powerful broadcast feature that allows you to send “one-off” messages to your contacts independently of any follow-up messages they might be receiving.
For example, you might use the broadcast feature to send out a weekly offer to everyone in your sales sequence. Or you might choose to send a monthly newsletter to all your contacts.
Broadcast messages can be plain text, HTML, or multi-part (both plain text and HTML). You can choose to send a specific format or the format preferred by each contact.
For each broadcast, you can create variants of the main message to use for split-testing purposes. Changes that can be made to variants include subject, message copy, message format, layout, and attachments.
With so many options available, it is a good idea to practice sending broadcast messages to a small number of your own email addresses to familiarize yourself with the process.
In this section, you will discover how to create a broadcast message.
To view the Broadcast Action Map, click Broadcasts on the menu bar.
You can create as many subscription forms for each autoresponder as you need. Many form settings are held within arpReach instead of in fields in the HTML of the form. This allows you greater central control of forms, meaning you can change form settings without having to alter the form’s HTML on your websites.
“Many form settings are held within arpReach instead of the HTML of the form code to allow easier changes without editing code”
To manage an autoresponder’s subscription forms, go to the Autoresponders list screen and choose Subscription forms from the Choose action… drop-down menu or click the number in the Forms column.
Once you have done so, this screen will appear:
Note that live forms must remain on the list. Do not delete forms that are currently accepting subscribers as they will become inoperable.
You can create a new subscription form by choosing Create a form for this autoresponder from the I want to… menu on the Subscription Forms list screen.
The options are the same for both creating a new form and editing an existing form.
This is where you define the content of the form and the way it operates.
For every field you want to include on the form, check the box in the Include on form column. Enter a label for each field you have chosen. Check the Mandatory and Capitalize columns if required.
Custom fields that you have defined in Content / Custom Fields will also appear in the field list and can be included on the form.
Here you can specify other autoresponders that this form will add a subscription to when a contact complete the form.
“Use captcha to ensure your forms are submitted by people and not by machines”
The image above is an example of an arpReach subscription form with the reCAPTCHA V2 Checkbox (styled and formatted using an HTML editor).
Custom pages are displayed to your contacts when they perform one of eight actions:
The following items in arpReach have custom page settings and the software will use them in the following precedence order:
You can also choose the method of passing form data to the subscribe success and failure pages:
#V2-exclusive: Pass form data on custom unsubscribe pages
#V2 has the capability to also pass contact data on custom unsubscribe success and failure pages using the GET or POST methods.
When you have created your form, you can generate its HTML by choosing Get form code from the Choose action… drop-down menu on the Subscription Forms list screen.
Click the form code to copy it to your clipboard, then paste the code into your preferred HTML editing software for styling.
As part of the styling process, the order of the fields can be changed and this will not cause a problem.
When you have finished styling the form in your preferred HTML editing software, copy the resultant HTML code and paste it into your web page.
Various real-time graphs are available for autoresponders and follow-up messages.
To view the available autoresponder graphs, choose Reports from the Choose action… drop-down menu on the Autoresponders list screen:
The default graph is Subscription overview, which shows the percentages of contacts based on their subscription status.
Click on the graph drop-down menu to change the graph to any of the other predefined options.
To view the available graphs for follow-up messages, choose Reports from the Choose action… drop-down menu on the Follow-Up Messages list screen:
The available graphs are:
The default graph is Overview. Click on the graph drop-down menu to change the graph to Variant success.
You can send yourself a test of a single follow-up message (and/or any of its variants) or send yourself test copies of all the main follow-up messages of the autoresponder in one go.
This way you can check if their content and layout are as you intended before setting their final scheduling.
You can send yourself a test of a single follow-up message and any of its variants by choosing Test from the Choose action… drop-down menu on the Follow-Up Messages list screen:
The information needed to initiate the test are the following:
To test a variant separately, go to the Message Variants list screen and choose Test from the Choose action… drop-down menu:
All you need is to input the email address you want to send the test message to and choose the format to send:
To send yourself copies of all the main (non-variant) messages in the follow-up sequence, choose Test from the Choose action… drop-down menu on the Autoresponders list screen:
The information needed to send the test messages are:
When testing the entire follow-up sequence, the messages may arrive in the wrong order depending on the route they take during delivery. This is normal and not a cause for concern.