METHODS AND APPARATUS TO DETERMINE INFORMED HOLDOUTS FOR AN ADVERTISEMENT CAMPAIGN

1 . An apparatus to reduce iterative computation efforts for an advertisement campaign, the apparatus comprising:

a buyer type determiner to determine a first group type and a second group type, the first and second group types associated with user identifiers corresponding to purchase instances, the user identifiers of the first group type indicative of a first threshold of purchase behaviors, and the user identifiers of the second group type indicative of a second threshold of purchase behaviors;

a holdout group identifier to identify (a) a first holdout group of the user identifiers of the first group type and (b) a second holdout group of the user identifiers of the second group type, the first and second holdout groups indicative of candidate user identifiers to be prevented from exposure to the advertisement campaign; and

a ratio constrainer to reduce computational lift calculation resource consumption for the advertising campaign by constraining the first holdout group to a first percentage of the first group type.

2 . The apparatus as defined in claim 1 , wherein the ratio constrainer is to constrain the second holdout group to a second percentage of the second group type.

3 . The apparatus as defined in claim 2 , wherein the first percentage is equal to the second percentage.

4 . The apparatus as defined in claim 1 , further including a publisher data retriever to retrieve, from a publisher, the user identifiers.

5 . The apparatus as defined in claim 4 , wherein the publisher data retriever is to retrieve, from the publisher, at least one of control group user identifiers, test group user identifiers, or exposed group user identifiers.

6 . The apparatus as defined in claim 1 , further including a household determiner to determine households that correspond to user identifiers associated with the purchase instances.

7 . The apparatus as defined in claim 1 , further including a lift calculator to calculate a lift value for the advertisement campaign based on the first and the second holdout groups that are not exposed to the advertisement campaign.

8 . The apparatus as defined in claim 7 , wherein the lift calculator is to determine an All Outlet Adjustment factor by extrapolating panelist data from an audience measurement entity.

9 . The apparatus as defined in claim 8 , wherein the lift calculator is to apply the All Outlet Adjustment factor to the lift value for the advertisement campaign.

10 . A method to reduce iterative computation efforts for an advertisement campaign, the method comprising:

determining a first group type and a second group type, the first and second group types associated with user identifiers corresponding to purchase instances, the user identifiers of the first group type indicative of a first threshold of purchase behaviors, and the user identifiers of the second group type indicative of a second threshold of purchase behaviors;

identifying (a) a first holdout group of the user identifiers of the first group type and (b) a second holdout group of the user identifiers of the second group type, the first and second holdout groups indicative of candidate user identifiers to be prevented from exposure to the advertisement campaign; and

reducing computational lift calculation resource consumption for the advertising campaign by constraining the first holdout group to a first percentage of the first group type.

11 . The method as defined in claim 10 , further including constraining the second holdout group to a second percentage of the second group type.

12 . The method as defined in claim 10 , further including retrieving, from a publisher, the user identifiers.

13 . The method as defined in claim 12 , further including retrieving, from the publisher, at least one of control group user identifiers, test group user identifiers, or exposed group user identifiers.

14 . The method as defined in claim 10 , further including determining households that correspond to user identifiers associated with the purchase instances.

15 . The method as defined in claim 10 , further including calculating a lift value for the advertisement campaign based on the first and the second holdout groups that are not exposed to the advertisement campaign.

16 . The method as defined in claim 15 , further including determining an All Outlet Adjustment factor by extrapolating panelist data from an audience measurement entity and applying the All Outlet Adjustment factor to the lift value for the advertisement campaign.

17 . A non-transitory computer readable storage medium comprising instructions that, when executed, cause a processor to at least:

determine a first group type and a second group type, the first and second group types associated with user identifiers corresponding to purchase instances, the user identifiers of the first group type indicative of a first threshold of purchase behaviors, and the user identifiers of the second group type indicative of a second threshold of purchase behaviors;

identify (a) a first holdout group of the user identifiers of the first group type and (b) a second holdout group of the user identifiers of the second group type, the first and second holdout groups indicative of candidate user identifiers to be prevented from exposure to an advertisement campaign; and

reduce computational lift calculation resource consumption for the advertising campaign by constraining the first holdout group to a first percentage of the first group type.

18 . The computer readable storage medium as defined in claim 17 , wherein the instructions, when executed, cause the processor to constrain the second holdout group to a second percentage of the second group type.

19 . The computer readable storage medium as defined in claim 17 , wherein the instructions, when executed, cause the processor to retrieve, from a publisher, the user identifiers.

20 . The computer readable storage medium as defined in claim 17 , wherein the instructions, when executed, cause the processor to determine households that correspond to user identifiers associated with the purchase instances.