FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION
(See Section 10 and Rule 13)
Title of invention:
A SYSTEM AND METHOD FOR MODELING A CONSUMER-CENTRIC PLANOGRAM BASED ON LINEAR OR NONLINEAR SUBSTITUTION
Applicant:
TATA Consultancy Services Limited A company Incorporated in India under The Companies Act, 1956
Having address:
Nirmal Building. 9th Floor,
Nariman Point, Mumbai 400021.
Maharashtra, India
The following specification describes the invention and the manner in which it is to be performed.

FIELD OF THE INVENTION
The present invention relates to the field of data collection and statistical data analysis. More particularly, the invention relates to the field of simulating a virtual planogram on the basis consumer-centric retail data processing and statistical substitution based data analytics.
BACKGROUND OF THE INVENTION
Retail sector has developed leaps and bounds in the recent years in the global market. The leading enterprises in the retail sector are switching to modem technologies such as computer systems for enabling automation of various activities in the store to focus increase in sales for the consumer products. One such activity is effective arrangement of various products of different brands in the shelves and racks of the store using automation tools such as planogram.
Generally, planograming is done to meet the demand and thus choice of placement of Stock-keeping Units (SKUs) is governed by sales. A planogram is a simulation tool that simulates visual representation of arranging different products of various brands. Planogram models a road-map for placement of products in the retail store to target the potential consumers. Another objective sufficed by the planogram is to effectively manage the available space by considering the market share of each product. Though, there have been efforts made in the past for effective modeling of planogram, none of the known techniques in the art is capable of analyzing the probable nonlinear substitutes for the within category product and then implementing the computer-based planograms on the basis of analysis to arrange a substitute products to help customers with shopping ease.

In general, a consumer visiting a merchandising store will be interested in buying her preferred product from the store. In certain scenarios, the consumer preferred product may not be available in the store. In such cases, the consumer may try to substitute the product with other competing products in the market. However, if close substitute is also not available then repeated absence of the preferred product may force the consumers to switch the store.
Based on the availability, when customers substitute, they need not necessarily buy an equivalent product in terms size, quality, attributes or units. Further, such a phenomenon may lead to change in consumption or change in customers' number of store visits, leading to nonlinear substitution. Here, nonlinear substitution refers to any of the following possibilities: (1) substituting the stock out product by higher/lower price alternative (i.e. nonlinearity in price) (2) substitute product may be of higher/lower pack size (i.e. nonlinearity in size) (3) substitution by buying lesser/more number of units of the alternatives (i.e. nonlinearity in units/volume) (4) substitution approach which may result into higher/lower consumption (i.e. nonlinearity in consumption) (5) substitution approach that may also result into more/less store trips (i.e. nonlinearity in price). However, in the existing scenarios of stock outs, the store administration is unable to determine probable nonlinear substitutes for the stock-out products using the known computer-based planograming systems and methods.
Thus, the existing computer-implemented simulation methods and systems in the art lack in considering consumer nonlinear substitution as one of the primary factors of designing planogram. This may result in store switching by the consumer if the preferred product or its substitute (though dissimilar in terms of size, price, unit etc) is not available consecutively for more number of occasions. Another drawback of ignorance of consumer nonlinear substitution is the lowering the probable chance of

increase in sales due to substitution by large package size, higher price and /or higher unit purchases of the alternative product.
Therefore, the current planograming methods and systems are not effective as they lack benefits of nonlinear consumer preference substitution approach for modeling. Hence in view of this, there is a long-felt need to enable a computer-implemented system and method in the art that models a virtual planogram based on the statistical linear or non-linear probabilistic substitution analysis using the consumer past purchase patterns that derives a visual representation and also provides a list of probable substitutes in case of product stock-outs in the merchandising stores.
OBJECTS OF THE INVENTION
The primary object of a present invention is to enable a computer-implemented system and method for modeling a consumer-centric planogram on the basis of . linear/Non-linear statistical probabilistic analysis.
Another object of the invention is to calculate a Competitive Substitution Index (CSI) for various products depicting comparability of competitive intensity amongst said products.
Yet another object of the invention is to perform linear or non-linear component analysis on the computed CSI in order to generate a hierarchical structure clustering varied products into multi-tier groups at various levels.
Still another object of the invention is to model a virtual planogram based on the generated hierarchical structure and space optimization.

SUMMARY OF THE INVENTION:
Before the present systems and methods, enablement are described, it is to be understood that this application is not limited to the particular apparatus, systems, and methodologies described herein, as there can be multiple possible embodiments which are not expressly illustrated in the present disclosures. It is also to be understood that the terminology used in the description is for the purpose of describing the particular versions or embodiments only, and is not intended to limit the scope of the present application.
In one embodiment, the present invention discloses a computer-implemented system and method for modeling a virtual planogram for arranging a plurality of SKU in a merchandising store. In accordance with an embodiment of the present invention, an SKU means SKU/Product/Brand/SKU size combination." The system of the present invention comprises an input module that is adapted for acquiring a panel data for competing SKUs from the Point-of-Sale (POS) terminal, wherein said panel data correlates a consumer historical/sequential/ probable future transactions associated with the SKUs in the merchandising stores. In one embodiment, the system of the present invention further comprises a competitive substitution index (CSI) computation module configured for computing a competitive substitution index (CSI) value for SKUs by analyzing the acquired panel data, wherein said competitive substitution index signifies a competitive intensity between individual SKUs with all probable substitute SKUs.
In one embodiment, the system of the present invention comprises a data analysis module configured for generating a hierarchical structure of plurality of SKUs by performing a linear/ nonlinear principal component analysis on the SKUs classifying

said plurality of SKUs into plurality of categories in a manner such that each category comprise of one or more SKUs having CSI value in close proximity with that of the other. In one embodiment, a data modeler of the system is then adapted for modeling a virtual intelligent planogram wherein closest substitutes are placed in vicinity of each other in a shelf. Further, the system also implements space optimization algorithm while designing the virtual planogram in assistance with the generated hierarchical structure. Finally, the Graphical User Interface (GUI) module displays the virtual planogram modeled and designed by the system of the present invention.
BRIEF DESCRIPTION OF DRAWINGS
The foregoing summary, as well as the following detailed description of embodiments, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, there is shown in the present document example constructions of the invention; however, the invention is not limited to the specific methods and apparatus disclosed in the document and the drawings:
Figure 1 is a system diagram (100) illustrating the various modules configured for modeling a consumer-centric planogram on the basis of linear or non-linear statistical analysis according to an embodiment of the invention.
Figure 2 is a flow diagram (200) illustrating different steps implemented for computing a competitive substitution index (CSI) for multiple SKUs in accordance with an embodiment of the invention.

Figure 3 is a flow diagram (300) illustrating different steps implemented for designing a Hierarchical Tree Diagram based on linear or non-linear principal component analysis (PCA) in accordance with an embodiment of the invention.
Figure 4 is an exemplary Hierarchical Tree Diagram (400) illustrating an example market structure graph designed as a result of linear/Non-linear principal component analysis (PCA) in accordance with an embodiment of the invention.
Figure 5 is a flow diagram (500) illustrating different steps implemented for developing a virtual planogram on the basis of designed hierarchical Tree Diagram in accordance with an embodiment of the invention.
DETAILED DESCRIPTION
The description has been presented with reference to an exemplary embodiment of the invention. Persons skilled in the art and technology to which this invention pertains will appreciate that alterations and changes in the described method and system of operation can be practiced without meaningfully departing from the principle spirit and scope of this invention.
According to one embodiment of the present invention, a computer-implemented system and method to model a consumer-centric planogram on the basis of linear or non-linear probabilistic analysis is disclosed. In an embodiment, the method and system of the present invention facilitates computation of Competitive Substitution Index (CSI) for the SKUs to be arranged in the modeled planogram. In order to enable this, all the competing SKU are indexed based on their substitution by analysis of the pane] data captured through merchandising environment. Such substitution is developed based on customer purchase patterns and probability of

purchases/ sequential purchases. These probabilities are combined and logarithm based index is generated called as Competitive Substitution Index (CS1).
In an embodiment, the system of the present invention is further configured for developing Hierarchical Market Structure Analysis based on the computed Competitive Substitution Index (CSI). In order to enable this, for each SKU, CSI is developed across different households considering their purchase data at periodic intervals. This index recalculated after few more periods may change. Further, for CSI values obtained across different households, Principal Component Analysis (PCA) is applied. According to the embodiment, either linear/non-linear PCA is applied on the resultant CSI value for SKUs to develop a hierarchical market structure. The Hierarchical Market Structure enables grouping of SKUs/brands together that compete equally fiercely with the rest of the SKUs based on CSI values. Accordingly, the grouping is done in a manner such that those SKUs/Brands that compete similarly with the rest of the SKUs are deemed equivalent and hence are categorized under the same group. The generated Hierarchical Market Structure is a multi-tier group with each higher level group is further segregated into sub-groups by application of linear/non-linear PCA. Such grouping thus developed is illustratively presented in a tree and matrix format which represents market structure for a group of SKUs/brands.
In an embodiment, as a result of development of Hierarchical Market Structure of SKUs/Brands, the system is configured to model a Planogram based on analysis of such a market structure. To achieve this, the developed market structure is therefore combined with SKU/brand sizes and shelf size space optimization algorithm is applied to create a virtual planogram. The created virtual planogram than can be replicated in the merchandising stores for arrangement of SKUs/Brands in a manner such that the SKUs out-of-stock can be substituted with the closest competitive

SKU/Brand. Various embodiments of the invention are now described in detail with reference to the figures appended below:
Referring to figure 1 is a system diagram illustrating various modules configured for modeling a consumer-centric planogram on the basis of linear or non-linear statistical analysis according to an embodiment of the invention. As illustrated in figure 1, the system (100) comprehends a distributed merchandising environment comprising several Point-of-Sale terminals (108) communicatively coupled with a central repository panel database (106) and a computing machine (102). The merchandising environment is adapted to develop a virtual planogram using linear or non-linear statistical analysis performed by the computing machine (102). The computing machine (102) further comprises of a machine-readable medium (101) electronically coupled to a processing device (103) as shown in figure 1. The machine-readable medium (101) comprises a plurality of modules configured for processing of one or more tasks as directed by the processing device (103) in response to the instructions stored on the machine-readable medium (101). In one embodiment, the central repository panel database (106) is adapted to store panel data captured through one or more Point-of-Sale (POS) server terminals (108).
In one embodiment, the computing machine (102) acts as a central processing server electronically coupled with the panel database (106) which is further communicatively coupled with the POS terminals (108). In one embodiment, the computing machine (102) may act as a standalone device configured to instruct the modules installed to perform the tasks of modeling the planogram. In an embodiment, the computing machine (102) may comprise a server computer, a client user computer, a personal computer (PC), a tablet PC, a laptop computer, a desktop computer, or any machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine. The computing machine

includes an embedded processing device (103) (e.g., a central processing unit (CPU), a graphics processing unit (GPU, or both), a memory which communicate with each other via a bus. The machine-readable medium (101) stores programmed instructions that when executed by the embedded processing device (103) in the computing machine (102) configures each of the modules including an input module (105). a competitive substitution index (CSI) computation module (107), a data analysis module (109), a data modeler (111) and a Graphical User Interface Module (113) to implement the methodology of modeling the virtual planogram as disclosed herein in accordance with one embodiment of the proposed invention. Further, while a single machine is illustrated, the term "machine" shall also be taken to include any collection of machines that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies as discussed herein.
In one embodiment, the input module (105) is configured for acquiring a panel data stored in the central repository panel database (106) related to past purchase patterns of a plurality of consumers from the POS sever terminals (108). The panel data represents historical purchase of different brands/SKUs from the store and the probable purchases of a particular set of products/SKUs of a particular brand over a specified periodic interval of time in a particular geographical region covering plurality of merchandising stores selling the products for varied brands. In this embodiment, the processing device (103) in responsive to the instructions received from the machine readable medium (101) directs the input module (105) to acquire the panel data from the panel database (106) maintained in the form of past purchase patterns, future purchase probabilities and substitution of few of the SKU/brands in the stores with the similar kind of brands/SKUs based on sequential purchases in the stores by variety of consumers. In one embodiment, the panel database (106) electronically transmits request for panel data collection to the POS server terminals

(108) communicatively coupled with panel database (106) in the merchandising environment.
In one embodiment of the invention, the processing device (103) in responsive to the instructions received from the machine readable medium (101) is triggered to configure the competitive substitution index (CS1) computation module (107) to compute Competitive Substitution Index (CSI) for each of the SKUs/Brands by analyzing the panel data acquired by the input module (105). In one embodiment, all the competing SKUs from the panel data captured from panel database (106) are indexed based on their substitution data. The substitution data is inferred based on probabilistic statistical analysis of consumer historical purchase patterns, probability of future purchases and probability of sequential purchases. The computation module (107) combines these probabilities to generate a logarithm based index for each SKU/Brand known as Competitive Substitution Index (CSI). In one embodiment, the computation module (107) is configured to calculate substitution index between all the competing brands for a given focal brand and to generate a singular number that captures combined competitive intensity across all the brands and ensures the comparability of the captured competitive intensity by transforming it on a common logarithm index. Figure 2 illustrates a flow diagram illustrating process flow of computing a competitive substitution index (CSI) by the computation module (107) according to one embodiment of the invention.
In one embodiment, as illustrated in figure 2, at step (201), a count table for an SKU and switching combination is calculated. At step (203), a conditional probability of purchasing a particular product is calculated. Next, at step (205), market share from past purchase data for a particular product is calculated. Further, at step (207), a difference in market share of individual SKUs is calculated. Based on the calculated difference in market share, a table illustrating a relative position of each

product/SKU/Brand with respect to other products/SKUs/Brands is computed at step (209). As illustrated in figure 2, at step (211). the values of the computed table are transformed into logarithmic values. At step (213), as a result of transformation of values, a resultant table, "Competitive Substitution Index (CSI)" is generated.
In one embodiment of the invention, the processing device (103) in responsive to the instructions received from the machine readable medium (101) configures the data analysis module (109) illustrated in figure I to perform a linear or non-linear substitution analysis on the calculated CSI values for each of competing brands/SKUs. In one embodiment, such linear or non-linear analysis is performed by applying Principal Component Analysis (PCA) on the computed CSI for each of the SKU/Brand. The linear or non-linear Principal Component Analysis (PCA) is employed on the CSI as underlying variable to explore competitive similarity among the SKUs/brands and linear or nonlinear substitution for each product. Such analysis derives insights representing the linear/non-linear substitution of set of SKUs/brands in the merchandising stores. The Principal Component Analysis (PCA) enables to derive possible correlation between various SKUs to depict and illustrate which SKUs/products are competing similarly with all other SKUs. In an embodiment, such linear/non-linear substitution is captured over plots as shown below:


Referring the above figure (Plot I), it indicates that the plot of principal component 1 and 2 are diagonal to each other but all the points are not properly aligned to two principal axes. Both principal component 1 and 2 takes into account 47.26% variation of the data.
Similarly, in the following figure (Plot II). all three combinations of principal components are depicted and plotted against each other. Except the first figure of PCI vs. PC2, no other figure depict any sort of linear relationship since the data points are not aligned to form a straight line. So it indicates that there is probable chance some nonlinear relationship existing between competing brands/SKUs/products and hence the system also facilitates application of the nonlinear principal component analysis.


In one embodiment, the data analysis module (109) is therefore adapted to execute Nonlinear Principal Component Analysis (NLPCA) which is operated to capture the nonlinear substitution in the purchase pattern. The non-linear relationship between the two principal components (PCI and PC2) is depicted in the figure Plot TIT illustrated as below:


In the above figure (Plot III), as depicted, the PCI vs. PC2 accounts for 79.62% variation of the data which is more than what is observed in the figure Plot I (PCI vs. PC2 47.26%). Thus, it can be concluded from the observed results of Plot I and Plot III that the Nonlinear Principal Component Analysis (NLPCA) captures more variation in the data points. The following graphs (Plot IV, Plot V) show the relationship between all the three principal components as per NLPCA result.

As indicated in the graph Plot IV, all the points plotted does not depict a linear relationship and it is almost looking like a nest while Plot V represent the nonlinear relationship depicted in the reverse 'S! shaped curve format.
In one embodiment, the processing device (103) in responsive to the instructions received from the machine readable medium (101) configures the data analysis module (109) illustrated in figure 1 to derive a hierarchical structure of plurality of SKU/Brand/Product on the basis of linear/non-linear Principal Component Analysis. This represents market structure for a group of SKUs/brands. The proposed market structure combines both linear and nonlinear PCA or substitution methods to develop comprehensive market structure. The derived hierarchical structure enables classification of different SKUs/Products into multiple groups or categories. In an embodiment, the SKUs/brands are grouped together in a manner such that the SKUs/Brands those compete more fiercely with each other are categorized in the same group. For each group, again NLPCA is applied to develop second tier groups. Such grouping of SKUs/Brands/Products is iteratively done until no more groups can be broken. The developed hierarchical structure is presented in a tree and matrix format. Figure 3 illustrates a flow diagram illustrating process flow of generating a hierarchical structure by the data analysis module (109) as a result of linear/nonlinear substitution analysis according to one embodiment of the invention.
In one embodiment, as illustrated in figure 3, at step (301), a data input in the form of Competitive Substitution Index (CS1) table obtained in the previous stages is received. At step (303). Principal Component Analysis (PCA) is performed on the received CSI table. At step (305), it is validated whether the PCA performed at step (303) depicts non-linear relationship. If the PCA performed is non-linear, then at step (307), Non-linear PCA (NLPCA) is applied on the data input. As a result of linear/non-linear analysis at steps (303, 305). a market structure diagram is generated

at step (309). Further, according to one embodiment, at step (311), NLPCA is iteratively re-performed on selected combinations of SKUs resulting in generation of a hierarchical Tree Diagram based on NLPCA analysis at step (313). The market structure graph generated as a result of the process flow illustrated in figure 3 depicts which SKUs compete with each other more fiercely. One exemplary graph indicating the market structure is illustrated in figure 4.
In an exemplary embodiment, figure 4 illustrates a Hierarchical Tree Diagram generated for the product category "Soft Drinks" based on linear/Non-lmear PCA analysis executed by the data analysis module (109) as per the directions of the processing device (103) in response to the programmed instructions stored on the machine readable medium (101) as illustrated in figure 1. Referring to figure 4, Principal Component Analysis (PCA) is applied on the SKUs X1....X23 to further segregate them into multi-tier groups by comparing the competitive intensity amongst these products using the computed CSI table. In this exemplary embodiment, XI, X2 and X3 are categorized as Coca-Cola in the first level, followed by grouping these into Strong Category at second level, then followed by Cola at third level and finally into Soft Drinks Category, Similarly, by application of linear/Non-Iinear PCA analysis, the other products X4....X23 are categorized into multiple levels to form a multi-tier matrix tree diagram illustrated in figure 4. At each level, linear/Non-linear PCA analysis is applied on the products categorized in the preceding level to further categorize the sub-products into the subsequent levels.
In one embodiment, as a result of generation of the Hierarchical Tree Diagram, the embedded processing device (103) in response to the instructions received from the machine readable medium (101) configures the data modeler (111) illustrated in figure 1 to model and design a virtual planogram on the basis of information relating to competition between different SKU's captured through the generated Hierarchical

Tree Diagram. Figure 5 illustrates a process flow diagram depicting steps followed by the data modeler (111) in order to model and develop the virtual planogram on the basis of captured hierarchy of competing SKUs.
In one embodiment, as illustrated in figure 5, a data input in the form of the hierarchical market structure is received at step (501). At step (503), the dimension Information about different SKUs/Brands with respect to various attributes is collected. At step (505). the dimension Information of the shelves that are required for analyzing the space-constraints as a result of impact caused due to product substitution is collected. At step (507), based on Market Structure output, an Intelligent Planogram is modeled. At step (509), the product substitution based on modeled planogram is reflecting in arranging the shelves at the merchandising store. In one embodiment, as a result of generation of the virtual planogram, the embedded processing device (103) in response to the instructions received from the machine readable medium (101) configures the GUI module (113) illustrated in figure 1 to display the virtual planogram.
In one embodiment, space optimization can be done by using the mixed integer nonlinear programming model (MINLP) and the Mixed Integer program (MIP). In this embodiment, the mixed integer non-linear programming formulation helps in the multiple product assortment problems with shelf space allocation and display location effects. The Mixed Integer program (MJP) relaxes the non-convex optimization problem into a linear Mixed Integer Program (MIP). The MIP not only generates near-optimal solutions, but also provides a posteriori error bound to evaluate the quality of the solution.
Although the invention has been described in terms of specific embodiments and applications, persons skilled in the art can, in light of this teaching, generate

additional embodiments without exceeding the scope or departing from the spirit of the invention described herein.
ADVANTAGES OF THE INVENTION
The present invention has following advantages:
• The present invention enables a design of planogram that is modeled considering the past consumer purchases, sequential purchase and probable substitutes.
• The present invention enables calculating the customer substitution based on panel/store data which helps in developing market structure for varied categories of products.
• The present invention enables to create planograms that account for customer (linear or non-linear) substitution which method helps to reduce store switching under stock-out situations.
• The present invention enables creating planograms based on the market structure analysis wherein close substitutes are placed nearby irrespective of their sizes to promote higher consumption and thus higher profits to retailers.

CLAIMS:
1. A system (100) for modeling a planogram of at least one Stock-keeping
unit (SKU) in a merchandising environment characterized by estimation of
probability of substitution of one Stock-keeping unit (SKU) with the
closest proximity competitor SKU in scenarios of stock-outs using
probabilistic linear or non-linear substitution analysis performed by a
computer, the said system comprising:
i) a panel database (106);
ii) one or more Point-of-Sale (POS) terminal (108) and a computing
machine (102) communicatively coupled with said panel database (106);
iii) said computing machine (102) further comprising:
a machine-readable medium (101) and a processing device (103) electronically coupled with said machine-readable medium (101), said processing device capable of executing plurality of modules stored on the machine-readable medium (101) wherein, the modules are:
a) an input module (105);
b) a competitive substitution index (CSI) computation module (107);
c) a data analysis module (109);
d) a data modeler (111); and
e) a GUI module.
2. The system of claim 1. wherein said space optimization is achieved by
means of a mixed integer Nonlinear programming model (MINLP) and
Mixed Integer program (MIP).

3. A computer-implemented method for modeling a planogram of at least one
Stock-keeping unit (SKU) in a merchandising environment characterized
by estimation of probability of substitution of one SKU with the closest
proximity competitor SKU in scenarios of stock-outs using probabilistic
linear or non-linear substitution analysis, the method comprising steps of:
a) acquiring a panel data for at least two competing SKU from at least one Point-of-Sale (POS) terminal, wherein said panel data correlates a consumer historical/sequential/ probable future transactions associated with at least one SKU;
b) computing a competitive substitution index (CSI) value for at least one SKU by analyzing the acquired panel data, wherein said competitive substitution index signifies a competitive intensity between individual SKU and probable substitution of one SKU with the other;
c) generating a hierarchical structure of plurality of SKU by executing a principal component analysis on at least two SKU classifying said plurality of SKUs into plurality of categories such that each category comprise of one or more SKUs having CSI value in close proximity with that of the other; and
d) modeling a virtual intelligent planogram wherein at least one stock-out SKU is substituted with the closest competitive SKU identified through the generated hierarchical structure.
4. The method of claim 3, wherein said CSI is a logarithmic value computed
by deriving plurality of combinations for a SKU switch, probability of
purchase of each SKU and difference in market share of each SKU on the
basis of acquired panel data.

5. The method of claim 3, wherein said principal component analysis is executed to derive linear or non-linear relationships amongst individual competing SKUs to explore competitive similarity to enable relevant substitution.
6. The method of claim 3, wherein each category groups one or more SKUs that relatively compete fiercely with each other.
7. The method of claim 3, wherein said modeling step further comprise of consideration of space-constraint attributes such as SKU size, SKU brand, shelf dimensions or combinations thereof.
8. The method of claim 1. wherein said space optimization is enabled by using combination of Mixed integer Nonlinear programming model (M1NLP) and Mixed Integer program (MIP).
9. The method of claim 8. wherein said mixed integer nonlinear programming model (MINLP) enables in the multiple product assortment problems with shelf space allocation and display location effects.
10. The method of claim 8, wherein said Mixed Integer program (MIP) enables relaxation of non-convex optimization problem into a linear Mixed Integer program (MIP) that generates qualitative optimal solutions.

11. The method of claim 3, wherein said panel data illustrates plurality of configurations of SKU substitutes for a particular stock-out SKU prerecorded at plurality of merchandising stores.