Methods and apparatus to generate consumer data

1. An apparatus to prevent transaction data skew, the apparatus comprising:

memory;

machine-readable instructions; and

a processor to execute the machine-readable instructions to:

identify a first linkage between membership card data associated with a membership program and transaction data associated with panel product data and a second linkage between the membership card data and the transaction data, the first linkage and the second linkage associated with a first time;

detect a change associated with the first linkage between the first time and a second time, the second time earlier than the first time;

responsive to the detection of the change, update the first linkage;

combine the updated first linkage with the second linkage to generate a final linkage file;

select a first portion of a sample of the transaction data for a first type of member of the membership program using the final linkage file, the first portion to satisfy a first target for a characteristic of the sample of transaction data, the first type of member corresponding to a first household type, the first household type associated with first retail purchasing behavior;

generate a second target for a characteristic of the transaction data based on the first portion of the sample of transaction data removed from consideration, the second target different from the first target; and

prevent skew caused by differential sampling rates in the sample of transaction data by selecting (a) the first portion of the sample of the transaction data and (b) a second portion of the sample of transaction data for a second type of member that satisfies the second target, the second type of member corresponding to a second household type, the second household type associated with second retail purchasing behavior.

2. The apparatus of claim 1 , wherein the processor is to determine an initial target for a characteristic of a sample of transaction data based on panelist data and retailer data.

3. The apparatus of claim 2 , 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.

4. The apparatus of claim 2 , wherein the first portion is to satisfy a first sample size requirement, the first sample size requirement based on point-of-sale data.

5. The apparatus of claim 1 , wherein preventing skew due to differential sampling rates includes selecting a second portion of the sample of transaction data for the second type of member that satisfies a second sample size requirement.

6. The apparatus of claim 1 , wherein the processor 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.

7. The apparatus as of claim 1 , wherein the processor 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.

8. A non-transitory tangible computer readable medium comprising instructions that, when executed, cause at least one processor to at least:

identify a first linkage between membership card data associated with a membership program and transaction data associated with panel product data and a second linkage between the membership card data and the transaction data, the first linkage and the second linkage associated with a first time;

detect a change associated with the first linkage between the first time and a second time, the second time earlier than the first time;

responsive to the detection of the change, update the first linkage;

combine the updated first linkage with the second linkage to generate a final linkage file;

select a first portion of a sample of the transaction data for a first type of member of the membership program using the final linkage file, the first portion to satisfy a first target for a characteristic of the sample of transaction data, the first type of member corresponding to a first household type, the first household type associated with first retail purchasing behavior;

generate a second target for a characteristic of the transaction data based on the first portion of the sample of transaction data removed from consideration, the second target different from the first target; and

prevent skew caused by differential sampling rates in the sample of transaction data by selecting (a) the first portion of the sample of the transaction data and (b) a second portion of the sample of transaction data for a second type of member that satisfies the second target, the second type of member corresponding to a second household type, the second household type associated with second retail purchasing behavior.

9. The computer readable medium as defined in claim 8 , wherein the instructions, when executed, cause the at least one processor to determine an initial target for a characteristic of a sample of transaction data based on panelist data and retailer data.

10. The computer readable medium as defined in claim 9 , 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.

11. The computer readable medium as defined in claim 9 , wherein the first portion is to satisfy a first sample size requirement based on point-of-sale data.

12. The computer readable medium as defined in claim 8 , wherein the instructions, when executed, cause the at least one processor to prevent skew due to differential sampling rates includes selecting a second portion of the sample of transaction data for the second type of member that satisfies a second sample size requirement.

13. The computer readable medium as defined in claim 8 , wherein the instructions, when executed, cause the at least one processor 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.

14. The computer readable medium as defined in claim 8 , wherein the instructions, when executed, cause the at least one processor 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.

15. A method to prevent transaction data skew, the method comprising:

identifying a first linkage between membership card data associated with a membership program and transaction data associated with panel product data and a second linkage between the membership card data and the transaction data, the first linkage and the second linkage associated with a first time;

detecting a change associated with the first linkage between the first time and a second time, the second time earlier than the first time;

responsive to the detection of the change, updating the first linkage;

combining the updated first linkage with the second linkage to generate a final linkage file;

selecting a first portion of a sample of the transaction data for a first type of member of the membership program using the final linkage file, the first portion to satisfy a first target for a characteristic of the sample of transaction data, the first type of member corresponding to a first household type, the first household type associated with first retail purchasing behavior;

generating a second target for a characteristic of the transaction data based on the first portion of the sample of transaction data removed from consideration, the second target different from the first target; and

preventing skew caused by differential sampling rates in the sample of transaction data by selecting (a) the first portion of the sample of the transaction data and (b) a second portion of the sample of transaction data for a second type of member that satisfies the second target, the second type of member corresponding to a second household type, the second household type associated with second retail purchasing behavior.

16. The method of claim 15 , further including determining an initial target for a characteristic of a sample of transaction data based on panelist data and retailer data.

17. The method of 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.

18. The method of claim 16 , wherein the first portion is to satisfy a first sample size requirement based on point-of-sale data.

19. The method of claim 15 , further including selecting a second portion of the sample of transaction data for the second type of member that satisfies a second sample size requirement.

20. The method of claim 15 , 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.