Hey,
My client wants to create control groups for campaigns. In Customer Intelligence Studio 6.6 there's a checkbox, in cells, called "cell represents a control group". Is there any written explanation regarding the behavior of this option? especially if the created control groups can be monitored through several campaigns
Hi Eitan123,
The easiest way to create control groups within a campaign is to use the feature called "Automatic holdout control groups" within the communication node. This feature essentially takes a random sample of the marketing cell and holds it back from the export file. If this sampling doesn't meet your needs and you want to be more explicit in the allocation of subjects to the control group then you can do this in the campaign diagram and use the "cell represents a control group" feature to do that by connecting it to the communication node. This will then set an indicator the Common Data Model (CI_CELL_PACKAGE.CONTROL_GROUP_TYPE_CD) allowing you to historically know that cell was a control group. The advantage of the "automatic holdout" method is that it creates the association between the target and control groups in the CI_CTL_GRP_CELL_X_TEST_CELL table, and many other features within the UI are based around the use of automatic holdouts.
More information can be found in the Marketing Automation User Guide section called "Control Groups"
Regards
James
Hi Eitan123,
The easiest way to create control groups within a campaign is to use the feature called "Automatic holdout control groups" within the communication node. This feature essentially takes a random sample of the marketing cell and holds it back from the export file. If this sampling doesn't meet your needs and you want to be more explicit in the allocation of subjects to the control group then you can do this in the campaign diagram and use the "cell represents a control group" feature to do that by connecting it to the communication node. This will then set an indicator the Common Data Model (CI_CELL_PACKAGE.CONTROL_GROUP_TYPE_CD) allowing you to historically know that cell was a control group. The advantage of the "automatic holdout" method is that it creates the association between the target and control groups in the CI_CTL_GRP_CELL_X_TEST_CELL table, and many other features within the UI are based around the use of automatic holdouts.
More information can be found in the Marketing Automation User Guide section called "Control Groups"
Regards
James
Cell_Package_SK is your friend for this.
The list of IDs selected for the campaign is in the Contact History table and each time a campaign is executed, each cell gets a new unique value for Cell_Package_SK. So by joining Contact History to CI_CELL_PACKAGE & CI_CTL_GRP_CELL_X_TEST_CELL via Cell_Package_SK and Marketing_Cell_SK you will get each target cell and control group membership for each execution of the campaign.
There are exactly three implications from this checkbox.
1) It is reflected in the CDM data model, under the CI_MARKETING_CELL table there is a column corresponding to this flag. And If my memory doesn't fail me, in CI_CONTACT_HISTORY too there is a similar column
2) In the Communication node, under "Export" tab, there is a checkbox - "Include holdout control group members". If it's checked, members of the control group (any kind - one defined via this checkbox on a cell node, or an "automated holdout" which is a different kind of a control group) will still be exported to the export file/table; If it's unchecked, they won't be.
3) If you ARE exporting control group members (see #2), then it's possible to differentiate between who's a member of a control group vs who's not in the output file, by adding a column to be exported: Marketing Cell => Control Group Flag
Want to review SAS CI360? G2 is offering a gift card or charitable donation for each accepted review. Use this link to opt out of receiving anything of value for your review.
Listen to the Reimagine Marketing podcast
Assess your marketing efforts with a free tool
SAS Customer Intelligence Learning Subscription (login required)
Compatibility notice re: SAS 9.4M8 (TS1M8) or later
SAS' Peter Ansbacher shows you how to use the dashboard in SAS Customer Intelligence 360 for better results.
Find more tutorials on the SAS Users YouTube channel.
Want to review SAS CI360? G2 is offering a gift card or charitable donation for each accepted review. Use this link to opt out of receiving anything of value for your review.
Listen to the Reimagine Marketing podcast
Assess your marketing efforts with a free tool
SAS Customer Intelligence Learning Subscription (login required)
Compatibility notice re: SAS 9.4M8 (TS1M8) or later