Methods and apparatus to generate consumer data
Methods and apparatus to generate consumer data are disclosed. An example method of selecting a sample of transaction data corresponding to a membership program includes defining a first type of member of the membership program; defining a second type of member of the membership program; calculating, via a processor, a target for the sample; selecting, via the processor, a first portion of the transaction data for the first type of member in accordance with the target; generating, via the processor, an updated target by recalculating the target with the first portion of the transaction data removed from consideration; and selecting, via the processor, a second portion of the transaction data for the second type of member in accordance with the updated target.
1. An apparatus to prevent transaction data skew, the apparatus comprising:
a consumer data generator to:
determine a target for a characteristic of a sample of transaction data based on panelist data and retailer data;
determine a first sample size requirement for a first type of member of a membership program and a second sample size requirement for a second type of member of the membership program;
select a first portion of the sample of transaction data for the first type of member, the first portion to satisfy the target and the first sample size requirement;
generate an updated target for the characteristic based on the first portion of the sample of transaction data removed from consideration; and
prevent skew due to differential sampling rates in the sample of transaction data by selecting (a) the first portion of the transaction data and (b) a second portion of the sample of transaction data for the second type of member that satisfies the updated target and the second sample size requirement.
2. The apparatus as defined in claim 1 , wherein the consumer data generator is to apply a purchase threshold filter to filter out portions of the transaction data corresponding to households that do not use a membership card at a threshold frequency.
3. The apparatus as defined in claim 1 , wherein the consumer data generator is to apply a geographic filter to the transaction data to filter out membership card data corresponding to card holders that have moved away from a known address associated with a membership card.
4. The apparatus as defined in claim 1 , wherein the consumer data generator is to define the first and second types of member of the membership program by stratifying the transaction data into first and second strata, the first type of member corresponding to single banner households and the second type of member corresponding to multi-banner households.
5. The apparatus as defined in claim 1 , wherein the consumer data generator is to determine the target based on at least one of a demographic profile or a purchasing behavior profile.
6. The apparatus as defined in claim 1 , wherein the characteristic of the sample of transaction data includes at least one of an average spend within a banner or a spend distribution within the banner.
7. A non-transitory tangible computer readable medium comprising instructions that, when executed, cause at least one processor to at least:
determine a target for a characteristic of a sample of transaction data based on panelist data and retailer data;
determine a first sample size requirement for a first type of member of a membership program and a second sample size requirement for a second type of member of the membership program;
select a first portion of the sample of transaction data for the first type of member, the first portion to satisfy the target and the first sample size requirement;
generate an updated target for the characteristic based on the first portion of the sample of transaction data removed from consideration; and
prevent skew due to differential sampling rates in the sample of transaction data by selecting (a) the first portion of the transaction data and (b) a second portion of the sample of transaction data for the second type of member that satisfies the updated target and the second sample size requirement.
8. The computer readable medium as defined in claim 7 , wherein the instructions, when executed, cause the at least one processor to apply a purchase threshold filter to the transaction data.
9. The computer readable medium as defined in claim 7 , wherein the instructions, when executed, cause the at least one processor to define the first and second types of member of the membership program by stratifying the transaction data into first and second strata, the first type of member corresponding to a single banner household and the second type of member corresponding to a multi-banner household.
10. The computer readable medium as defined in claim 7 , wherein the instructions, when executed, cause the at least one processor to determine the target based on a purchasing behavior profile.
11. The computer readable medium as defined in claim 7 , wherein the membership program includes an incentive program for registered users.
12. The computer readable medium as defined in claim 7 , wherein the instructions, when executed, cause the at least one processor to determine the target based on a demographic profile.
13. The computer readable medium as defined in claim 7 , wherein the instructions, when executed, cause the at least one processor to determine the target based on a geographic profile.
14. The computer readable medium as defined in claim 7 , wherein the instructions, when executed, cause the at least one processor to apply a geographic filter to the transaction data.
15. The computer readable medium as defined in claim 7 , wherein the characteristic of the sample of transaction data includes at least one of an average spend within a banner or a spend distribution within the banner.
16. A method to prevent transaction data skew, the method comprising:
deter mining, by executing an instruction with a processor, a target for a characteristic of a sample of transaction data based on panelist data and retailer data;
determining, by executing an instruction with the processor, a first sample size requirement for a first type of member of a membership program and a second sample size requirement for a second type of member of the membership program;
selecting, by executing an instruction with the processor, a first portion of the sample of transaction data for the first type of member, the first portion to satisfy the target and the first sample size requirement;
generating, by executing an instruction with the processor, an updated target for the characteristic based on the first portion of the sample of transaction data removed from consideration; and
preventing skew due to differential sampling rates in the sample of transaction data by selecting, by executing an instruction with the processor, (a) the first portion of the transaction data and (b) a second portion of the sample of transaction data for the second type of member that satisfies the updated target and the second sample size requirement.
17. The method as defined in claim 16 , further including applying a purchase threshold filter to filter out portions of the transaction data corresponding to households that do not use a membership card at a threshold frequency.
18. The method as defined in claim 16 , further including applying a geographic filter to the transaction data to filter out membership card data corresponding to card holders that have moved away from a known address associated with a membership card.
19. The method as defined in claim 16 , wherein the defining of the first and second types of member of the membership program includes stratifying the transaction data into first and second strata, the first type of member corresponding to a single banner household and the second type of member corresponding to a multi-banner household.
20. The method as defined in claim 16 , wherein the characteristic of the sample of transaction data includes at least one of an average spend within a banner or a spend distribution within the banner.