Control group
Why do you require Control groups
The control groups mechanism helps to evaluate the influence of communication on customers behavior in a segment by comparing indicators (conversion into purchases, average receipt, average revenue per customer) for customers who received and who did not receive communication.
The mechanism allows you to select a small group (usually 5-10% of the segment size) from the target communication segment. This control group does NOT receive communication, and the platform calculates indicators for it in the same way as for the main group (that receives communication).
In the post-analysis, the platform calculates what the revenue would have been in the main group if the indicators in it were the same as in the control group (that is, if the main segment had not received communication).
Creating control group
Control groups can be used in all kinds of mass mailings (e-mail, SMS, push) and campaigns.
In mass mailings
Use the Control group size slider to create a control group in the Mailing volume block. The size of the control group can vary from 0 to 50% of the target segment (0% by default). The number of customers included in the control group is displayed.
In campaigns
To create a control group, use the Control group size slider on the campaign settings page. The size of the control group can range from 0% to 50% of the target segment (0% by default).
Reports
The control group report is available on the page where all other mailing reports or campaign reports.
The table displays the following information for each group:
Number of recipients: how many messages were sent to this group;
Messages delivered: how many messages were delivered to this group;
Number of customers who made a purchase;
Conversion into buyers: how many percent of customers, who received the email, made a purchase under the influence of the mailing;
Number of receipts in the group;
Average number of receipts per customer;
Amount of purchases;
Average revenue per customer;
Anticipated revenue at conversion and average receipts of the control group: how much money customers would have spent on purchases if there would be no mailing;
Calculated for the main group only;
Equals to (Messages delivered)*(Conversion into buyers in control group)*(Average revenue per customer in control group).
For example, the report shows:
99,048 people received the mailing.
If you wouldn't make the mailing, conversion into buyers would be 8.39%, as in the control group.
With such a conversion 99,048 * 0.0839 = 8,310 buyers would have made a purchase.
The control group's average revenue per customer: 17 USD. So, if you wouldn’t make the mailing, purchases will be made for the amount of 8,310 x 17 = 141,270.
Additional revenue is a profit received via mailing.
Equals (Total amount of purchases in the main group) - (Anticipated revenue at conversion and average receipt of the control group).
For example, the sum of all main group purchases is 176,029 and anticipated revenue is 148,368 USD. The difference makes up 27 662 USD.
Values in some columns are highlighted in color: green - the best value, red - the worst value.
For example, in Group A the average revenue per customer is 2,317 rubles, which is the best value among all groups (highlighted in green). In group B - 995 rubles, this is the worst value (highlighted in red).
The report shows the attributed receipts associated with the mailing. You can read more about attribution in the article Receipt attribution.
Segment builder
In the segment builder, you can use filters associated with control groups:
Participated in Control group “E-mail mailing”;
Participated in Control group “SMS mailing”;
Participated in Control group “Push mailing”;
Participated in Control group “Campaign”.
For mailings/campaigns with a control group the Participated in e-mail mailing, Participated in SMS mailing, Participated in push mailing, Participated in Control group “Campaign filters are not included in the sampling of the control group participants.
If you want to select the entire mailing/campaign segment (control group + main group), you should apply two filters via the logical operator OR. Example:
Last updated
