Adding a control group to a delivery

In this video, I will explain how you - can define a control group for a delivery and how you - can then extract the profiles of that control group after - the delivery has been executed.

The control groups feature allows you to measure the impact of the - Campaign by excluding a portion of your audience from - receiving a delivery. We’ll then be able to compare the behavior of the target population, - which did receive the message with the behavior of contacts, - which were not targeted. In addition, you can then - also explicitly target the control group in future campaigns.

For the purpose of this demo, I’ve already created a - workflow with an audience and an email delivery.

To access the control group feature, you need to access the delivery, - click on the Audience card, and then so like the Control Group tab.

There are two main ways you - can define a control group. The first option is to - extract the number of profiles from the main target. The second option is to - exclude some profiles based on criteria defined in a query. You can of course also use the - combination of both methods, when you define a control group. All profiles that are part of the control group will be removed from the main targets during - the delivery preparation. So they will not receive - a message once it is sent. Let’s do it. First choose the target extraction option. Then choose to extract a - random sample or you can choose to keep only the first - records after sorting, which allows you to - define a limitation based on a sorting order.

If you choose to extract - the random sample, you have two options. The sample can either be a percentage of the initial population, or you can define a maximum - size, which is a fixed number. For example, if you select the percentage and set the threshold to - 10 in the limit section, the control group will be made above 10%, so they could randomly from - the targeted population.

If you select a second - extraction option that keep only the first records after sorting, you can - define a limitation based on what are more sorting orders. For example, you can select - the age as a sorting criteria. So let me show you how to do this. Click create element, select the field.

In this case, the age field, click confirm, check that the descending - order is selected.

Now select the limit. I would like to exclude 100 people.

This will result in - the control group being made up of 100 oldest - recipients from your topic.

In addition, you can also use the target exclusion method to - exclude some profiles based on criteria to find an aquarium. So let’s say I also want - to exclude profiles younger than 20 years.

So I will click the - define target exclusion from the left hand side. So like the field age, and then select less than, twice!

This is the query editor. So you can define multiple criteria and add to the ones we’ve defined. So once you’re done, click confirm.

As you can see on the Oddlins tile, we have defined a control - group with two criteria. The first being the 100 oldest profiles. So this is done through defining - the order as descending, and then I take it to 100. And then we also added - all profiles under 20. All profiles that meet these criteria, will be excluded from the delivery.

Don’t forget to save.

So now we’re all set. We can go back to the targeting - workflow and execute it. Let’s start the workflow.

Now the message has been sent and we can see all the - profiles have been removed from the target by - checking the delivery logs.

In our case, as there’s - a recurring delivery, just like the right - delivery, which is this one. It’s the only one we have. And from that specific delivery, I can access the logs by - clicking on the three dots. On the sending logs tab you can see there is a status called ignored, and a natural failure - called control group. So these are all the profiles - that have been excluded from the target audience - based on the control group that we defined earlier.

You can also check the exclusion causes and that table, you can see the number of the excluded profiles - by exclusion costs. We can see that 305 - profiles have been excluded, as they are part of the control group.

So what can we do with - this control group now that we’ve sent the delivery? There’s several options. First you can extract the - sending box to compare how the control group that did - not receive the communication, active compared to the effective target.

You can also use the delivery - logs to be define a new target or a new delivery.

Let me show you how to do this. Let’s go back to our workflow, and let me first disable these activities and activate the query - that I’ve already created.

What I want to do is - retrieve the profiles based on the control group - from the delivery box.

First, go to properties. As we went to retrieve all - the profiles that were part of the control group - from the delivery logs we’ll need to define that the - delivery box as the resource. And targeting dimension should be profile.

Once you defined this, you can set the target.

So let’s define the query. And this case I drag and - drop with delivery box here, then select the status field, and select that the value ignored. And as we have seen in the sending logs. Then I added another query - on that delivery logs. And so the nature of failure field, and select the value control group. So this query will retrieve - all profiles that were part of the control group. its safe.

Next, you can either build a new delivery, or you can export the log data using an extract file activity followed by a transfer - file activity for example. And this will allow you - to upload the information into your reporting tool to - analyze track user behavior of the effective targets - compared to the control group. And so I know the answer to - measure impact of your campaign. -