Methods and apparatus to predict new product performance metrics

• US 8,543,446 B2
• Filed: 02/27/2012
• Issued: 09/24/2013
• Est. Priority Date: 02/28/2011
• Status: Active Grant

• Current Assignee: Nielsen Consumer LLC (Nielsen Holdings plc)
• Original Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Inventors: Richardson, Bruce C., Quinn, Martin, Long, Jeffrey S.
• Primary Examiner(s): Jasmin, Lynda
• Assistant Examiner(s): Koneru, Sujay
• Application Number:
  US13/405,856
• Publication Number:
  US 20120296699A1
• Time in Patent Office: 575 Days
• Field of Search: 705 711- 742
• US Class Current: 705/7.31
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

Methods and apparatus are disclosed to predict new product performance metrics. An example disclosed method includes selecting a product hierarchy tree based on a distribution measure for a new product, identifying a segment location of the product hierarchy tree for the new product, retrieving a relationship matrix associated with the segment location of the product hierarchy tree, the relationship matrix comprising available product performance coefficients and attribute information, identifying a spatial location for the new product in the relationship matrix based on the attribute information, calculating, using a programmed processor, an impact parameter based on the spatial location within the relationship matrix, and calculating the performance metric for the new product based on the impact parameter and a channel sales value.

Claims

1. A method to predict performance metrics of a new product, comprising:

identifying, with a programmed processor, a segment location of a product hierarchy tree based on a projected distribution measure for the new product;

retrieving, with the programmed processor, a relationship matrix associated with the segment location of the product hierarchy tree, the relationship matrix comprising available product performance coefficients and attribute information;

calculating, using the programmed processor, an impact parameter based on a spatial location within the relationship matrix, the spatial location based on the attribute information;

calculating the performance metrics for the new product for first and second assortments of related products on a store shelf, the first and second assortments associated with first and second proportional volume shifting effects, respectively, the performance metric calculated based on the impact parameter and a sum of (a) a ratio of category sales in a target geography and overall category sales in a sales channel associated with the new product, and (b) a ratio of channel sales in the target geography and overall channel sales in the sales channel associated with the new product; and

selecting a candidate product to de-list based on sales effects with the first and second assortment of products on the store shelf.

2. A method as described in claim 1, further comprising identifying a plurality of product hierarchy trees related to at least one of a market channel or a market category associated with the new product.

3. A method as described in claim 1, further comprising retrieving a second relationship matrix associated with the projected distribution measure.

4. A method as described in claim 3, wherein the projected distribution measure includes a geographic indicator of at least one of a category or a channel associated with the new product.

5. A method as described in claim 4, further comprising identifying products in the relationship matrix that match sibling products in the product hierarchy tree.

6. A method as described in claim 5, further comprising adding a row and column associated with the new product to the relationship matrix adjacent to the matched sibling products.

7. A method as described in claim 6, further comprising calculating cell coefficient values for a plurality of cells in the row and column associated with the new product based on the matched sibling products.

8. A method as described in claim 1, further comprising tailoring the relationship matrix based on geographic attributes associated with a geography of interest.

9. A method as described in claim 1, further comprising de-listing the candidate related product when overall category sales for a store decrease and sales of the candidate related product decrease.

10. A method as described in claim 1, further comprising refraining from de-listing the candidate related product when sales of the candidate related product decrease while overall category sales for a store increase.

11. An apparatus to predict performance metrics of a new product, comprising:

a tree manager to select a product hierarchy tree based on a distribution measure for a new product, and to identify a segment location of the product hierarchy tree based on the distribution measure for the new product;

a matrix modifier to retrieve a relationship matrix associated with the segment location of the product hierarchy tree, the relationship matrix comprising available product performance coefficients and attribute information, the matrix modifier to identify a spatial location for the new product in the relationship matrix based on the attribute information; and

a proximity weighting engine to calculate an impact parameter based on the spatial location within the relationship matrix, and to calculate the performance metric for the new product for a first and second assortment of related products on a store shelf, the first and second assortments to be associated with first and second proportional volume shifting effects, respectively, the performance metric calculated based on the impact parameter and a sum of (a) a ratio of category sales in a target geography and overall category sales in a sales channel associated with the new product, and (b) a ratio of channel sales in the target geography and overall channel sales in the sales channel associated with the new product; and

a new product manager to select a candidate product to de-list based on a sales effect with the first and second assortment of products on the store shelf; and

wherein the tree manager, the matrix modifier, the proximity weighting engine and the new product manager comprising a logic circuit.

12. An apparatus as described in claim 11, wherein the tree manager is to select the distribution measure based on a channel and a category.

13. A tangible machine readable storage device or storage disk comprising machine-accessible instructions, that, when executed, cause a machine to, at least:

identify a segment location of a product hierarchy tree based on a projected distribution measure for the new product;

retrieve a relationship matrix associated with the segment location of the product hierarchy tree, the relationship matrix comprising available product performance coefficients and attribute information;

calculate an impact parameter based on a spatial location within the relationship matrix, the spatial location based on the attribute information;

calculate performance metrics for the new product for a first and second assortment of related products on a store shelf, the first and second assortments associated with first and second proportional volume shifting effects, respectively, the performance metrics calculated based on the impact parameter and a sum of (a) a ratio of category sales in a target geography and overall category sales in a sales channel associated with the new product, and (b) a ratio of channel sales in the target geography and overall channel sales in the sales channel associated with the new product; and

select a candidate product to de-list based on sales effects with the first and second assortment of products on the store shelf.

14. A tangible storage device or storage disk as described in claim 13, wherein the machine-accessible instructions, when executed, cause the machine to identify the projected distribution measure is indicative of a channel and a category.

15. A tangible storage device or storage disk as described in claim 13, wherein the machine-accessible instructions, when executed, cause the machine to receive a candidate segment location in response to identifying the segment location of the product hierarchy tree.

16. A tangible storage device or storage disk as described in claim 15, wherein the machine-accessible instructions, when executed, cause the machine to select the candidate segment location based on a product similarity input from a user.

17. A tangible storage device or storage disk as described in claim 13, wherein the machine-accessible instructions, when executed, cause the machine to retrieve a second relationship matrix associated with the projected distribution measure.

18. A tangible storage device or storage disk as described in claim 17, wherein the machine-accessible instructions, when executed, cause the machine to identify at least one of a category or a channel associated with the new product.

19. A tangible storage device or storage disk as described in claim 18, wherein the machine-accessible instructions, when executed, cause the machine to identify products in the relationship matrix that match sibling products in the product hierarchy tree.

20. A tangible storage device or storage disk as described in claim 19, wherein the machine-accessible instructions, when executed, cause the machine to add a row and column associated with the new product to the relationship matrix adjacent to the matched sibling products.

21. A tangible storage device or storage disk as described in claim 20, wherein the machine-accessible instructions, when executed, cause the machine to calculate cell coefficient values for a plurality of cells in the row and column associated with the new product based on the matched sibling products.