When you define interaction points for your visitors, you can further specialize those
interaction points by adding them to zones. Each zone can have its own set of rules that determine
how offers are presented to visitors. Use the Add zone icon on the
Interaction points tab of the Interactive channel window
to add a zone.
Procedure
-
Click the Add zone icon on the Interaction points
tab.
The Add/edit Zone dialog is displayed.
- Enter a Name for the zone.
- Enter a Description for the zone.
This description that you enter here is displayed on the Interaction
points tab of the Interactive channel window after you save the
zone.
- Optional:
Click the Advanced features link to display more settings that you can
apply to the zone.
If you display the advanced features, you can determine the following settings for the zone you
are adding or editing:
The Learning mode section specifies the rules by which learning is
applied to the zone you are creating or editing. You can choose any of the following settings:
-
Inherit from interactive channel is the default setting, and tells
Unica Interact to use the learning mode that is
specified for the interactive channel.
-
Use marketer's scores only indicates that Unica Interact uses only the marketer's scores for offer
weighting.
-
Use custom learning model uses a custom learning model that you can
specify from the drop-down by name. If no custom learning models are defined for this interactive
channel, this option is dimmed and cannot be selected.
For rule groups in this zone, resolve non-uniform learning rules by
handles a situation where there is a conflict between how offers are sorted in a mixed-learning
scenario, such as when some offers are suggested by sources in which learning is enabled, and other
offers are suggested from sources where learning is disabled. For example, because each zone and
each treatment rule group can define its own learning rule, and the rule that is defined for a
treatment rule group can be different from the rule that is defined for the zone, the choices here
determine how Unica Interact resolves that
conflict.
There are several different levels in Unica Interact at which learning can be enabled or disabled:
globally, for each interactive channel, for each treatment rule group, and for each zone. The
choices in the For rule groups in this zone, resolve non-uniform learning rules
by section provide the following solutions to the conflict:
-
Intermixing - Use Learning Mode for this zone (Default) indicates that the
sequence of offers is selected based on the learning mode that is defined for this zone. This
overrides the settings on individual rules.
-
Prioritizing non-learning offers indicates that the sequence of offers
selects non-learning offers first, followed by learning-based offers.
-
Prioritizing learning offers indicates that the sequence of offers selects
learning offers first, followed by non-learning-based offers.
-
Prioritizing non-learning offers: dedupe provides the same results as the
Prioritizing Non-learning Offers option, but with duplicate offers removed
when the same offer is provided by learning and non-learning sources.
-
Prioritizing learning offers: dedupe provides the same results as the
Prioritizing Learning Offers option, but with duplicate offers removed when
the same offer is provided by learning and non-learning sources.
% Random specifies the chance that the returned offer is randomly
selected, without considering scores. You can also use percentRandomSelection
in
Campaign | partitions | partition[n] | Interact | learning
to globally configure
the percent of the time the learning module presents a random offer. If you set %
Random here, it overrides the global configuration.
-
Click Save and return to go back to the Interaction Points tab or
Save and add another to continue adding interaction points.
Results
To edit the name or description of a zone, click the name of the zone on the
Interaction points tab.
To delete a zone, select the check box next to the zone on the Interaction Points tab and click
the Delete icon. You cannot delete a zone while it is used in a treatment
rule.
Example
For example, suppose that you have a zone
with four rule groups:
Rule group Group1 has a learning module that is named LM1.
LM1 contains offer1 with a score of 50%, and offer2 with a score 60%.
Rule group Group2 has no learning module.
Rule group Group3 is inheriting for the zone.
Rule group Group4 contains offer3 with a score of 65%, and
offer4 with a score of 45%. No learning model is assigned to this
rule group. The Use marketer score option is
selected for this rule group.
In this example, offers for each particular zone have a mix
of offers presented after evaluating the score, using learning or
no learning, depending on the configuration. The following list describes
some scenarios:
If you select Intermixing - use learning mode for this zone (default) for
this zone, the sequence of offers to be selected are based on the highest score first, regardless of
the built-in learning score. Using the rule groups above, the sequence would be Offer3 weighted at
65%, Offer2 weighted at 60%, Offer1 weighted at 50%, and Offer4 weighted at 40%.
If you select Prioritizing non-learning offers for the zone, the sequence
of offers is based on non-learning offers being selected first, so the sequence that uses the
example rule groups would be offer3 non-learning, offer4 non-learning, offer1 with learning, and
finally offer2.
If you select Prioritizing learning offers for this zone, the sequence of
offers to be selected are based on selecting the learning offers first, so the sequence that uses
the example rule groups would be offer1 Learning, offer2 Learning, offer3 non-learning, and finally
offer4.
If you select Prioritizing non-learning offers: dedupe for the zone, the
results would be the same as when you select Prioritizing non-learning
offers; however, the results would remove duplicate offers.
If you select Prioritizing learning offers: dedupe for the zone, the
results would be the same as when you select Prioritizing learning offers;
however, the results would remove duplicate offers.
If you specify a value for the Related to % Random field, offers other than the high priority would be prioritized
according to the specified percentage.