WO2000072176A2

WO2000072176A2 - Approximation of weekly sales data from audit data for use in analytical marketing models

Info

Publication number: WO2000072176A2
Application number: PCT/US2000/013429
Authority: WO
Inventors: William Todd Kirk; Henry George Bright
Original assignee: The Coca-Cola Company
Priority date: 1999-05-21
Filing date: 2000-05-16
Publication date: 2000-11-30
Also published as: IL146177A0; CA2369073A1; JP2003511747A; BR0010732A; WO2000072176A8; KR20010113835A; AU4714200A; EP1190338A2; CN1375088A; MXPA01011609A

Abstract

Audit data collected at a retail establishment pertains to a plurality of purchases. For each purchase, an associated purchase date and an associated number of purchased units is determined. For each purchase date, a depletion rate is determined that will cause the number of purchased units to be depleted upon the occurrence of the next successive purchase date. The depletion rate is an estimation of daily sales occurring between successive purchase dates. Causal data may also be collected at the retail establishment, which is considered to be valid during a causal audit period. Total estimated sales occurring during the causal audit period are determined by summing the estimated daily sales for each purchase date within the causal audit period. Relationships may be identified between the causal data and the total estimated sales during the causal audit period. If a relationship between the causal data and the total estimated sales is identified, one or more marketing strategies may be developed for taking advantage of the identified relationship. Marketing strategies may be analyzed using one or more analytical marketing models to determine whether any of the marketing strategies are viable. Any marketing strategies that are determined to be viable may be implemented in the retail environment. At a later time, updated audit data and causal data may be collected and analyzed in the above-described manner for the purpose of monitoring the performance of the one or more implemented marketing strategies.

Description

Approximation of Weekly Sales Data From Audit Data For Use in Analytical Marketing Models

Technical Field

The present invention relates to identifying relationships between sales data and causal data in a retail environment. More particularly, the present invention relates to approximating scanner-like sales data from purchase data in order to identify relationships between sales data and causal data in an audit-based retail environment.

Background Of The Invention

Scanner technology has revolutionized sales tracking in the retail environment. Scanners, such as bar code scanners and the like, are used in conjunction with electronic cash registers, point of sale terminals, and other checkout devices in order to collect and record data relating to sales by a retailer to a consumer. Such sales data may be used to track the precise date and time of the sale of an inventory item. Accordingly, scanners allow retailers and/or suppliers to determine the amount of sales in a given period of time without the need for traditional auditing. As is known in the art, traditional auditing involves the time consuming process of counting of the number of items in invemtory, a repeating the inventory count at a late time, and recording the number of purchases made by the retailer between the two inventory counts.

The use of scanners to efficiently and accurately collect sales data frees traditional audit personnel to perform causal audits. Causal audits involve the collection of causal data. As is well known in the art, a causal may be any factor, event or condition that causes increased or decreased sales of an inventory item. For example, any expenditure of money for the purpose of enticing consumers to purchase a product, e.g. print advertisements or displays, may be a potential causal. In addition, causals may take the form of environmental factors and market conditions surrounding competitive and/or complementary products. Causals may be either permanent (i.e., constant) or temporary (i.e.. variable). Temporary causal are likely to fluctuate on a daily or weekly basis. Identification of relationships between sales data and causal data often proves useful in the development of new marketing strategies for increased sales. Identifying relationships between sales data and causal data requires a time alignment of the sales data and the causal data. Time alignment of sales data and temporary causal data has only been heretofore practical in a scanner-based system, where the precise times and dates of sales are known. Without the luxury of scanners, sales data is collected by traditional inventory auditing techniques. Due to the fact that traditional inventory audits are time consuming, they are typically performed no more frequently than on a monthly or bi-monthly basis. Therefore, audit data is used to estimate sales over a relatively long time duration and does not directly provide the level of detail required for time alignment with temporary causal data.

Accordingly, there exists a need for a system and method for approximating scanner-like sales data from audit data in order to identify relationships between the sales data and causal data in an audit-based retail environment.

Summary Of The Invention

The present invention meets the above described needs by providing a system and method for approximating scanner- like sales data from audit data. Audit data is collected at a retail establishment. Audit data pertains to a plurality of purchases by a retailer. For each purchase, an associated purchase date and an associated number of purchased units is determined. For each for each purchase date, a depletion rate is determined that will cause the number of purchased units to be depleted upon the occurrence of the next successive purchase date. The depletion rate is assumed to be equal to the approximated daily sales between successive purchase dates. Causal data may also be collected at the retail establishment. A causal audit period is determined comprising a period of time in which the causal data is considered to be valid. Then, the total estimated sales during the causal audit period are determined by summing the estimated daily sales for each purchase date within the causal audit period. Given the causal data and the total estimated sales during the causal audit period, an attempt may be made to identify a relationship between the causal data and the total estimated sales. Attempting to identify a relationship between the causal data and the total estimated sales may comprise analyzing historical sales data. If a relationship between the causal data and the total estimated sales is identified, one or more marketing strategies may be developed for taking advantage of the identified relationship. Marketing strategies may then be analyzed using one or more analytical marketing models to determine whether any of the marketing strategies are viable. Any marketing strategies that are determined to be viable may be implemented in the retail environment.

The present invention provides computer-readable media having stored thereon computer-executable instruction for performing the above-described process. In addition, the present invention provides a system operable to perform the above-described method comprising one or more memory storage devices, at least one processor and one or more communications links for receiving and transmitting data to a remote device.

Brief Description Of The Drawings

FIG. 1 is a flow chart illustrating an exemplary method for approximating scanner-like data from audit data so as to identify effects of causal data on estimated sales data.

FIG. 2 is a graph of dated purchases by a retailer that plots purchased units versus days of the week.

FIG. 3 is a graphical illustration of the process of calculating estimated daily sales based on dated purchases by a retailer.

FIG. 4 is a graphical representation of an exemplary process involved in time aligning causal data and estimated weekly sales data. FIG. 5 illustrates an exemplary method for assisting marketing managers in making better and more actionable marketing decisions, in accordance with an exemplary embodiment of the present invention.

FIG. 6 is a functional block diagram illustrating an exemplary operating environment for an exemplary embodiment of the present invention.

Detailed Description Of Exemplary Embodiments

Exemplary embodiments of the present invention will hereinafter be described with reference to the drawings, in which like numerals indicate like elements throughout the several figures. As mentioned, the present invention provides a system and method for approximating scanner-like sales data from audit data, so as to identify relationships between estimated weekly sales data and causal data in an audit-based retail environment. For clarity, the exemplary embodiments are described herein from the perspective of a retailer. Thus, the term "purchase^*" is herein used to refer to a purchase of items by the retailer from a supplier or manufacturer, etc. Similarly, a ^"'sale" is meant herein to refer to a sale of an item from a retailer to a consumer. In addition, the term "scanner-like sales data" herein refers to dated sales data. Accordingly, the present invention seeks to approximate dated sales data from audit data that relates to purchases by a retailer over a time duration.

FIG. 1 is a flow chart illustrating an exemplary method 100 for approximating scanner-like data from audit data so as to identify effects of causal data on estimated sales data, in accordance with an exemplary embodiment of the present invention. As shown, the exemplary method 100 progresses from starting block 101 to step 102, where an inventory count is performed at a retail establishment. As should be apparent, an inventory count is meant to describe the process of counting the number of items or units that are in inventory. It is assumed at this point that a prior inventory count has previously been performed. Next, purchase data is collected at step 103 for purchases occurring between the inventory count and the prior inventory count. At step 104, the purchase date of each purchase occurring between the inventory count and the prior inventory count is determined. Purchase dates may be ascertained through examination of purchase receipts, if available. If purchase receipts are not available, retailers or suppliers may be instructed in advance to record dates of purchases in a journal or other type of log.

At step 105, causal data is collected at the given retail establishment. As previously described, causal data may relate to any factor, event or condition that may cause increased or decreased sales of an inventory item. Causal data may be collected internal to or external from the retail establishment. Exemplary causals may comprise display racks, posters, advertisements, placement or price of competitive or complementary products, weather conditions, economic conditions, and the like. Other types of causal data will be apparent to those of ordinary skill in the art. For more information regarding causals, see Hardie. Bruce et. al (1998), "Attribute-based Market Share Models: Methodological Development and Managerial Applications.^" University of Pennsylvania: Wharton School Working Paper 98-009, which is herein incorporated by reference in its entirety. Conventionally, a causal auditor maintains a list of identified potential causals and inspects a retail environment to determine whether such causals are present. It should be noted from FIG. 1 that the collection of causal data may occur before, after or concurrently with the performance of the inventory count, the collection of purchase data and the determination of purchase dates.

At step 106 an assumption is made that the number of units purchased on a given purchase date combined with a safety stock is equal to the starting inventory level on that purchase date. This assumption is based on the notion that a retailer will attempt to maintain a safety stock level of inventory and that all inventory units in excess of the safety stock should be sold before the next purchase. Therefore, upon the next purchase, all purchased units will exceed the safety stock. Accordingly, the assumption of step 106 treats the safety stock as a constant. Next, at step 107 each starting inventory level (i.e. number of units purchased plus safety stock) is decayed to the safety stock level at a depletion rate, in order to calculate average daily sales between purchase dates. In other words, it is assumed that inventory units in excess of the safety stock level will be exhausted upon the occurrence of the next successive inventory purchase date. To illustrate the principles described with reference to steps 106-107, consider the following example: if 21 units are purchased on a Tuesday and the next successive purchase date occurs on the following Friday, it is assumed that the starting inventory level on Tuesday is 21 inventory units plus a safety stock. Then, it is assumed that the 21 inventory units in excess of the safety stock will be depleted by way of sales occurring on Tuesday, Wednesday and Thursday, inclusive. The depletion rate for the present example is 7 units per day, which means that on Wednesday the starting inventory level will be 14 units plus the safety stock, on Thursday the starting inventory level will be 7 units plus the safety stock and on Friday the starting inventory level will be equal to the safety stock (until the next purchase is completed).

Upon completion of step 107, estimated sales will have been calculated for every day within the period between the inventory count of step 102 and the prior inventory count, i.e., the audit period. At step 108 the time period during which the causal data is considered to be valid is determined. As presiously stated, a causal audit may occur on a single day. one week per month. However, the causal data may be considered to be valid for the entire week surrounding the causal audit. For example, if a causal comprises a particular display, it may be common practice to utilize the display in the retail establishment for at least one week. At step 110. the estimated number of sales occurring during the time period in which the causal data is valid is calculated. Next, at step 112 an attepmt is made to identify effects that the causal data has had on the estimated sales. Step 112 may be performed, for example, by way of comparison to historical data. The exemplary method 100 ends at step 114. FIGS. 2-4 graphically illustrate the exemplary method 100 for approximating scanner-like data from audit data so as to identify effects of causal data on estimated weekly sales data. Beginning with FIG. 2, a graph is shown that plots inventory units 202 versus days of the week 204. Audit data comprises records of purchases 206 by the retailer. In accordance with an exemplary embodiment of the present invention, the dates of purchases 206 must be known or otherwise determined. As shown in FIG. 2, the illustrative retailer has made six purchases within a sample four week period: a first purchase of 21 units 206A was made on the first Tuesday 204A, a second purchase of 72 units 206B was made on the first Friday 204B, a third purchase of 54 units 206C was made on the second Tuesday 204C, a fourth purchase of 55 units 206D was made on the second Friday 204D, a fifth purchase of 75 units 206E was made on the third Wednesday 204E, and a sixth purchase of 40 units 206F was made on the fourth Monday 204F.

FIG. 3 is a graphical illustration of an exemplary process of calculating estimated daily sales 310 based on the dated purchases 206. The graph of FIG. 3 plots inventory units 302 versus days of the week 204. As shown, the first purchase 206A was made on the first Tuesday 204A and the next purchase 206B was made on the first Friday 204B. As described with reference to FIG. 1 , the purchase amount on a particular day plus a safety stock level is considered to be the starting inventory level on that day. A starting inventory amount is decayed to the safety stock level at a depletion rate between purchase dates. Thus, 21 inventory units in excess of the safety stock is decayed at a depletion rate between the first Tuesday 204A and the first Friday 204B. The depletion rate is equal to the slope of the line connecting the points defined by the coordinates (Tuesday. 21 ) and (Friday, Safety Stock). As shown, the depletion rate for the 21 inventory units between the first Tuesday 204A and the first Friday 204B is equal to 7 inventory units per day. Therefore, there are estimated daily sales 310A1-3 of 7 units per day between the first purchase 206A and the second purchase 206B. The second purchase 206B of 72 units on the first Friday 204B is similarly decayed to the safety stock level between the second purchase 206B and the third purchase 206C. As shown, there are estimated daily sales 310B1-3 of 18 units per day between the second purchase 206B and the third purchase 206C. Likewise, the third purchase 206C of 54 units on the second Tuesday 204C is decayed to yield estimated daily sales 310C1-3 of 18 units per day between the third purchase 206C and the fourth purchase 206D. Continuing on, the fourth purchase 206D of 55 units on the second Friday 204D is decayed to yield estimated daily sales 310D1-3 of 11 units per day between the fourth purchase 206D and the fifth purchase 206E. Lastly, the fifth purchase 206E of 75 units on the third Wednesday 204E is decayed to yield estimated daily sales 310E1-3 of 15 units per day between the fifth purchase 206E and the sixth purchase 206F. The sixth and last purchase 206F in the audit data is not decayed due to the fact that the date of the next purchase date is unknown. From FIG. 3, it may be seen that estimated daily sales may be calculated for each day within the audit data period, namely each day between the first Tuesday 204A and the fourth Monday 204F shown on the graph. Accordingly, scanner-like sales data has been approximated from audit data.

FIG. 4 is a graphical representation of the process involved in time aligning causal data and estimated daily sales data 210. The graph of FIG. 4 plots estimated units sold 403 versus days of the week 204. As indicated, the day of the causal audit 402 occurred on the second Thursday 204J shown on the graph. In the example of FIG. 4. it is assumed that the time period in which causal data is valid is one week. Therefore, the week of the causal audit 404 runs from the first Sunday 204G to the second Saturday 204K shown on the graph. The total estimated sales during the week of the causal audit 404 may be determined by summing the estimated daily sales 310 for each day in the week of the causal audit 404. Specifically, the total estimated sales of 112 units is equal to the sum of the estimated daily sales 310B3 of 18 units on Sunday 204G, the estimated daily sales 310B4 of 18 units on Monday 204H. the estimated daily sales 310C1 of 18 units on Tuesday 204C, the estimated daily sales 310C2 of 18 units on Wednesday 2041, the estimated daily sales 310C3 of 18 units on Thursday 204 J, the estimated daily sales 310D1 of 1 1 units on Friday 204D, and the estimated daily sales 310D2 of 1 1 units on Saturday 204K. Given the fact that an estimated 1 12 units were sold during the week of the causal audit, determinations may be made as to whether particular causals had a positive or negative impact on sales. FIG. 5 illustrates an exemplary method 500 for assisting marketing managers in making better and more actionable marketing decisions, in accordance with the present invention. As shown, the exemplary method 500 begins at starting block 501 and proceeds to step 502, where performance of a particular marketing strategy is monitored. As is known in the art, a marketing strategy may comprise one or more causals that are created for the purpose of increasing sales of a particular inventory item. Monitoring the performance of a marketing strategy in a scanner-based environment involves collecting and analyzing scanner sales data and causal data in order to determine the effect of the marketing strategy (i.e., the one or more causals) on the sales. However, in an audit based system, scanner sales data is not available. Therefore, audit data must be collected and manipulated so as to approximate scanner-like sales data. The manipulated audit data and the causal data are then analyzed as they would be in a scanner-based environment. For example, estimated weekly sales data may be time aligned to correspond to weekly causal data. It may then be determined if such causal data has a positive impact on sales, compared to historical sales data.

Once audit data and causal data are collected and analyzed, the method advances to step 504, where new opportunities for increased sales are identified. As an example, it may be determined that a particular causal has had a positive impact on sales and that introduction of a similar or complementary causal may also have a positive impact on sales. In addition, it may be determined that a particular causal has had a negative impact on sales and that removal of that causal may provide an opportunity for increased sales. At step 506 new marketing strategies are created for taking advantage of the identified opportunities for increased sales. Such new marketing strategies may comprise methods for introducing and/or removing causals from a retail environment.

The potential of each of the new marketing strategies created at step 506 is assessed at step 508. Assessing the potential of a marketing idea may involve creating and studying analytical marketing models using software and/or statistical tools. An exemplary method for analyzing sales data and causal data is the Multiplicative Competitive Interaction (MCI) model, described in Cooper. Lee and Nagkanishi, Masao (1988) Market Share Analysis. Boston: Kluwer Academic Publishers, which is herein incorporated by reference in its entirety. Using manipulated audit data, a marketing model based on the MCI model demonstrates the opportunity to estimate consumer response to anticipated changes in non- scanner marketing environments. Alternatives ranked by predicted increase in brand sales due to fluctuation in share as well as expansion of the category are then easily identifiable. As is well known in the art, there are many theories and models for analyzing sales data and/or causal data. The following reference, which is also herein incorporated by reference in their entirety, discusses other examples of analytical marketing models: Neslin, Scott et. al (1994), "A Research Agenda for Making Scanner Data More Useful to Managers," Marketing Letters 5 (4). pp. 395- 412.

At step 510, those new marketing strategies that are deemed likely to have a positive impact on sales are implemented in the retail environment. After new marketing strategies are implemented in the retail environment, the exemplary method 500 returns to step 502 for monitoring of the marketing strategies. As shown, the exemplary method 500 is a continuous process aimed at constantly attempting to increase the rate of sales. FIG. 6 and the following discussion are intended to provide a brief and general description of a suitable computing environment for implementing the present invention. Although the system shown in FIG. 6 comprises a conventional personal computer 600, those skilled in the art will recognize that the invention also may be implemented using other types of computer system configurations. The computer 600 includes a central processing unit 622, a system memory 620. and an

Input/Output ("I/O") bus 626. A system bus 621 couples the central processing unit

622 to the system memory 620. A bus controller 623 controls the flow of data on the I/O bus 626 and between the central processing unit 622 and a variety of internal and external I/O devices. The I/O devices connected to the I O bus 626 may have direct access to the system memory 620 using a Direct Memory Access

("DMA") controller 624.

The I/O devices are connected to the I/O bus 626 via a set of device interfaces. The device interfaces may include both hardware components and software components. For instance, a hard disk drive 630 and a floppy disk drive 632 for reading or writing removable media 650 may be connected to the I/O bus 626 through disk drive controllers 640. An optical disk drive 634 for reading or writing optical media 652 may be connected to the I O bus 626 using a Small Computer System Interface ("SCSI") 641. Alternatively, an IDE (ATAPI) or EIDE 5 interface may be associated with an optical drive such as a may be the case with a CD-ROM drive. The drives and their associated computer-readable media provide nonvolatile storage for the computer 600. In addition to the computer-readable media described above, other types of computer-readable media may also be used, such as ZIP drives, or the like. o A display device 653, such as a monitor, is connected to the I/O bus

626 via another interface, such as a video adapter 642. A parallel interface 643 connects synchronous peripheral devices, such as a laser printer 656, to the I/O bus 626. A serial interface 644 connects communication devices to the I/O bus 626. A user may enter commands and information into the computer 600 via the serial 5 interface 644 or by using an input device, such as a keyboard 638, a mouse 636 or a modem 657. Other peripheral devices (not shown) may also be connected to the computer 600, such as audio input/output devices or image capture devices.

A number of program modules may be stored on the drives and in the system memory 620. The system memory 620 can include both Random Access Memory ("RAM") and Read Only Memory ("ROM"). The program modules control how the computer 600 functions and interacts with the user, with I/O devices or with other computers. Program modules include routines, operating systems 665, application programs, data structures, and other software or firmware components. In an illustrative embodiment, the present invention may comprise one or more audit data manipulation program modules 670. one or more data analysis program modules 675, and one or more analytical marketing model program modules 677 stored on the drives or in the system memory 620 of the computer 600. Specifically, audit data manipulation program modules 670 and data analysis program modules 675 may comprise computer-executable instructions for extracting audit data and causal data from one or more databases, manipulating the audit data so as to approximate scanner-like data, time aligning the approximated scanner-like data and the causal data, and identifying relationships between the time- aligned approximated scanner-like data and causal data. Furthermore, analytical marketing model program modules 677 may comprise computer-executable

1 o instructions for processing approximated scanner-like data and causal data to assess the viability of new marketing strategies.

The computer 600 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer 5 660. The remote computer 660 may be a server, a router, a peer device or other common network node, and typically includes many or all of the elements described in connection with the computer 600. In a networked environment, program modules and data may be stored on the remote computer 660. The logical connections depicted in FIG. 6 include a local area network ("LAN") 654 and a o wide area network ("WAN") 655. In a LAN environment, a network interface 645, such as an Ethernet adapter card, can be used to connect the computer 600 to the remote computer 660. In a WAN environment, the remote computer 660 may use a telecommunications device, such as a modem 657, to establish a connection. It will be appreciated that the network connections shown are illustrative and other devices of establishing a communications link between the computers may be used. The remote computer 660 may comprise, or may be in communication with, a hand-held device for collecting audit data or causal audit data in a retail establishment, such as a scanner, a laptop computer, or a palmtop computer. Audit data and causal data collected at a retail establishment may be stored in one or more memory storage devices 620 coupled to a remote computer and/or a local computer 600.

Alternative embodiments of the present invention will become apparent to those having ordinary skill in the art to which the present invention pertains. For example, although preferred time periods were recited pertaining to monthly audit data, estimated daily sales data, and estimated weekly sales data, other time periods for such data will occur to those of skill in the art. In addition other system architectures for implementing the methods of the present invention will be apparent to those skilled in the art. Such alternate embodiments are considered to be encompassed within the spirit and scope of the present invention. Accordingly, the scope of the present invention is described by the appended claims and is supported by the foregoing description.

Claims

CLAIMSWhat is claimed is:

1. A method for approximating daily sales data from audit data comprising the steps of: receiving audit data pertaining to a plurahty of purchases, each purchase having an associated purchase date and an associated number of purchased units; and for each purchase date, determining a depletion rate for the number of purchased units that will cause the number of purchased units to be depleted upon the occurrence of the next successive purchase date, whereby the depletion rate is equal to the approximated daily sales between successive purchase dates.

2. A computer-readable medium having stored thereon computer- executable instructions for performing the method of claim 1.

3. The method of claim 1 , wherein the audit data is collected at a retail establishment and is stored in a database; and wherein receiving the audit data comprises retrieving the audit data from the database.

4. A computer-readable medium having stored thereon computer- executable instructions for performing the method of claim 3.

5. A method for analyzing audit data and causal data compring the steps of: receving the audit data and the causal data, the audit data pertaining to a plurality of purchases, each purchase having an associated purchase date and an associated number of purchased units; manipulating the audit data so as to estimate daily sales between each successive purchase date by performing the step of. for each purchase date, determining a depletion rate for the number of purchased units that will cause the number of purchased units to be depleted upon the occurrence of the next successive purchase date; determining a causal audit period comprising a period of time in which the causal data is considered to be valid; determing a total estimated sales during the causal audit period by summing the estimated daily sales for each purchase date within the causal audit period; and attempting to identify a relationship between the causal data and the total estimated sales.

6. A computer-readable medium having stored thereon computer- executable instructions for performing the method of claim 5.

7. The method of claim 5, wherein the audit data is collected at a retail establishment and is stored in a database; and wherein receiving the audit data comprises retrieving the audit data from the database.

8. The method of claim 5, wherein attempting to attempting to identify a relationship between the causal data and the total estimated sales comprises analyzing historical sales data.

9. The method of claim 5, further comprising the steps of: identifying a relationship between the causal data and the total estimated sales; developing a marketing strategy for taking advantage of the identified relationship; and implementing the marketing strategy in a retail environment.

10. The method of claim 9, further comprising the step of analyzing the marketing strategy with an analytical marketing model to determine that the marketing strategy is viable prior to implementing the marketing strategy in the retail environment.

11. A computer-readable medium having stored thereon computer- executable instructions for performing the method of claim 10.

12. A system for approximating daily sales data from audit data 5 comprising: a memory storage device for storing audit data pertaining to a plurality of purchases, each purchase having an associated purchase date and an associated number of purchased units; and a processor for receiving the audit data from the memory storage device and o for each purchase date, determining a depletion rate for the number of purchased units that will cause the number of purchased units to be depleted upon the occurrence of the next successive purchase date, whereby the depletion rate is equal to the approximated daily sales between successive purchase dates. 5

13. The system of claim 12, wherein causal data is also stored in the memory storage device; and wherein the processor is further operable for: receiving the causal data from the memory storage device; determining a causal audit period comprising a time period in which the causal data is considered to be valid; determing a total estimated sales during the causal audit period by summing the estimated daily sales for each purchase date within the causal audit period; and processing the total estimated sales and the causal data to attempt to identify a relationship between the causal data and the total estimated sales.

14. The system of claim 13, wherein the audit data and the causal data is collected at a retail establishment and is transmitted to the memory storage device via a network.

15. The system of claim 13, wherein the processor is further operable for: processing the total estimated sales and the causal data to identifying a relationship between the causal data and the total estimated sales; executing an analytical marketing model for analyzing a marketing strategy for taking advantage of the identified relationship; and in response to execution of the analytical marketing model, communicating to a user that the marketin *gfc strategy is viable.

16. The system of claim 12, wherein causal data is stored in a second memory storage device; and wherein the processor is further operable to: receiving the causal data from the second memory storage device; o determine a causal audit period comprising a time period in which the causal data is considered to be valid; determing a total estimated sales during the causal audit period by summing the estimated daily sales for each purchase date within the causal audit period; and processing the total estimated sales and the causal data to attempt to identify a 5 relationship between the causal data and the total estimated sales.

17. The system of claim 16, wherein the processor is further operable for: processing the total estimated sales and the causal data to identifying a relationship between the causal data and the total estimated sales; executing an analytical marketing model for analyzing a marketing strategy for taking advantage of the identified relationship; and in response to execution of the analytical marketing model, communicating to a user that the marketing strategy is viable.

18. The system of claim 12, wherein the audit data is collected at a retail estabhshment and is transmitted to the memory storage device via a network.

19. A method for analyzing audit data and causal data, the audit data pertaining to a plurahty of purchases, each purchase having an associated purchase date and an associated number of purchased units compring the steps of: manipulating the audit data so as to estimate daily sales between each successive purchase date by performing the step of, for each purchase date, determining a depletion rate for the number of purchased units that will cause the number of purchased units to be depleted upon the occurrence of the next successive purchase date; determining a causal audit period comprising a period of time in which the causal data is considered to be valid; 5 determing a total estimated sales during the causal audit period by summing the estimated daily sales for each purchase date within the causal audit period; and identifying a relationship between the causal data and the total estimated sales; developing a marketing strategy for taking advantage of the identified relationship; and o implementing the marketing strategy in a retail environment.

20. A computer-readable medium having stored thereon computer- executable instructions for performing the method of claim 19.

5 21. The method of claim 19, wherein attempting to attempting to identify a relationship between the causal data and the total estimated sales comprises analyzing historical sales data.

22. The method of claim 19, further comprising the step of analyzing the marketing strategy with an analytical marketing model to determine that the marketing strategy is viable prior to implementing the marketing strategy in the retail environment.

23. The method of claim 19, further comprising developing a plurahty of marketing strategies for taking advantage of the identified relationship; analyzing the plurahty of marketing strategies with at least one analytical marketing model to determine whether at least one of the marketing strategies is viable; and implementing any of the marketing strategies that are determined to be viable in the retail environment.