EP1679649A1

EP1679649A1 - Methods of collecting and processing data releted to the consumer behaviour of customers of retail stores

Info

Publication number: EP1679649A1
Application number: EP05026498A
Authority: EP
Inventors: Wilfried Wauters; Jacques Chahine
Original assignee: AC Nielsen Belgium
Current assignee: AC Nielsen Belgium
Priority date: 2005-11-03
Filing date: 2005-12-05
Publication date: 2006-07-12

Abstract

Methods of collecting and processing data related to consumer buying behaviour of customers of retail stores, comprising (a) receiving, at a sample of retail stores during at least one collection session within which data is collected for a predetermined period of time, several data carriers on each of which is stored cash slip information regarding the products and related prices of a customer's individual purchase and each of which is output by a point-of-sale terminal of a store; (b) reading the data carrier in order to obtain the cash slip information from the data carrier; (c) recording the obtained cash slip information in a storage device; and (d) calculating in a computing device estimated product-specific sales numbers based on the recorded data.

Description

FIELD OF THE DISCLOSURE

This patent relates to the collection and processing of business- and market-sensitive data, and in particular to methods for collecting data related to the consumer behaviour of customers of retail stores.

BACKGROUND

Understanding market dynamics is an important requirement for the successful conduction of business. Typically, business relevant information is evaluated and predicted on the basis of knowledge about product movement, market-share, distribution, price and other market-sensitive information. Retail measurement services, which measure and track, e.g., sales volume and selling price help to gauge product penetration, overall product performance, distribution, promotion effectiveness, and price sensitivity.
Known technologies for surveying product and market relevant data are based on in-store scanning of product codes. For example, the codes of the products bought by a customer are scanned at the point of sale terminal. The data gathered by product scanners is cleaned, enriched and processed in order to evaluate business relevant information from the collected raw data. However, the method of scanning product codes has the disadvantage that active cooperation of the participating stores is required concerning the product scanner and cash register technology involved in the data collection process.

SUMMARY OF THE INVENTION

The invention is directed to a method of collecting and processing data related to consumer buying behaviour of customers of retail stores, the method comprising: (a) receiving, at a sample of retail stores during at least one collection session within which data is collected for a predetermined period of time, several data carriers on each of which is stored cash slip information regarding the products and related prices of a customer's individual purchase and each of which is output by a point-of-sale terminal of a store; (b) reading the data carrier in order to obtain the cash slip information from the data carrier; (c) recording the obtained cash slip information in a storage device; and (d) calculating in a computing device estimated product-specific sales numbers based on the recorded data.
According to another aspect, a method of collecting and processing data related to the consumer buying behaviour of customers of retail stores, comprising: (a) intercepting, at a sample of retail stores during at least one collection session within which data is collected for a predetermined period of time, a number of customers leaving the stores; (b) addressing each of these customers in order to obtain information contained on the cash slip regarding the products and related prices of their purchase in the store; (c) recording the cash slip information of these customers; and (d) calculating in a computing device estimated product-specific sales numbers based on the recorded data.
Other features are inherent in the methods and products disclosed or will become apparent to those skilled in the art from the following detailed description of embodiments and its accompanying drawings.

DESCRIPTION OF THE DRAWINGS

Fig. 1 is a schematic overview of a preferred example hand scanner for reading cash slips;
Fig. 2 is a schematic overview of an alternative example input device located at a remote central site for entering cash slip information into a processing device;
Fig. 3 is a schematic overview of an alternative example, in which cash slip information is entered in a computing system located at a remote central site;
Fig. 4 is a flow diagram showing method performed for an individual customer;
Fig. 5 is a flow diagram showing method concerning the evaluation of data collected from individual customers;
Fig. 6 is a flow diagram showing method concerning the evaluation of estimated product-identification information and collection session parameters;
Fig. 7 is a flow diagram showing method concerning the evaluation of estimated product-identification information, collection session parameters and additional market-relevant data;
Fig. 8 is a table showing an example distribution of collecting sessions over a reporting period and over regions of interest;
Fig. 9 is a pie chart showing an example result of a retail measurement analysis; and
Fig. 10 is a schematic overview of a known method of collecting sales information.

DETAILED DESCRIPTION

Before discussing preferred embodiments in detail with reference to the accompanying drawings, some general aspects of these examples are addressed beforehand.
Collecting retail information as used in this patent means gathering any data which is helpful to understand sales, product performance, consumer buying behaviour and market dynamics. This is achieved by identifying products and product dependent information based on information obtained from individual purchases of customer's in a store. Thus, the information obtained is not restricted to sales information, but covers also information about the consumers' buying behaviour.
The term 'collecting data' should be understood as also comprising evaluation processes performed on collected raw data in order to obtain meaningful conclusions, e.g. based on statistical considerations, from the collected raw data. This patent thus relates to both, the technical collection process for elevating the raw data, and to the evaluation of the collected raw data with the help of technical means.
The terms 'store', and 'retail store' as used in the following detailed description should be understand broadly as including any facilities and retail channels for which the disclosed method might be of interest. This comprises, e.g., supermarkets, gas/petrol stations, convenience stores, pharmacies, supercenters, food stores, mass merchandisers, warehouse clubs, and other local channels.
The collection process is performed for a sample of stores and for representative periods of time. This means, that not necessarily every store participating in the data collection process must be continuously surveyed in order to obtain statistically meaningful data. Instead, it is sufficient to determine, e.g. in a planning stage, a sample of stores out of the collection of stores and particular time periods, for which and during which the collection portions of the process are performed.
'Representative time periods' means here, that the selected periods of time should be appropriately designed and distributed according to statistical requirements and sociological considerations in order to obtain raw data which allows, e.g. a research company, to draw significant conclusions concerning the issues of interest described above. The 'sample of stores' should not be understood as a fixed and closed set of stores. Instead, the process is also compatible with changes in the sample's constitution, which might be foreseen in a planning stage, or, later, during execution of the process.
Some portions of the collection method are repeated in front of some or, preferably, all sample stores during the time periods foreseen for each of these stores. A thus defined segment of the collection process performed during a predetermined period of time is named 'collection session'.
During such a collection session, individual purchases of customers of the surveyed stores are collected and evaluated. The term 'individual purchase' used here indicates a specific purchase of one or more items (or shopping trip) of a customer leaving the store within the predetermined time period.
The data gathered with respect to an individual purchase thus comprises detailed product-related data, e.g. the names or identifiers of the purchased products, the quantity of each purchased product (pieces, weight, etc.), the price of each product (either piece price or weight price), as well as further individual parameters, e.g. the total price of the purchase, a time-stamp, a store identifier, etc.
Information reflecting such an individual purchase is collected, according to the preferred example described below, by receiving a data carrier output of a point-of-sale terminal of the store. The data carrier output stores or otherwise contains information about products and related prices of the customer's individual purchase, as well as additional parameters related to the individual purchase. The collection of information stored on the data carrier is denoted by the term 'cash slip information' in order to indicate that all kind of information, which relates to the customer's individual purchase is included (such information is typically found on a cash slip received by a customer at a point-of-sale terminal).
The data carrier is designed for receiving information about an individual purchase of the customer. Preferably, the data carrier is output from the point-of-sale terminal at the time of the purchase and handed over to the customer before he leaves the store.
The term 'data carrier' comprises any kind of medium which is capable of transporting or otherwise storing information, e.g. a magnetic storage device (e.g. a magnetic card), an optical storage device, a memory device (e.g. an USB stick), a print medium (e.g., a paper receipt), etc. The term 'data carrier' also comprises non-physical carriers like a short transmission signal (e.g. radio transmission, or infrared transmission) which transports the cash slip information from the point-of-sale terminal to the customer over air. In this case, the customer is equipped with an adequate device for receiving the transmission emitted from the point-of-sale terminal (e.g. a Bluetooth-capable mobile phone, handheld-PC, etc.).
Another process which is repeated for some, or preferably every individual purchase comprises reading the data carrier to obtain the cash slip information stored on the data carrier. Reading the data carrier may be realized with an appropriate reading device, e.g. a card reader in case of a magnetic care, a scanning device in case of a print medium, or a Bluetooth-capable data receiver (e.g. a handheld-PC) in case of a Bluetooth signal emitted by the point-of-sale terminal or by a Bluetooth-device of the customer. However, the reading process should not be understood as being restricted to the use of devices or to the particular devices named above.
A further process which is repeated for some or, preferably every individual purchase comprises recording the obtained cash slip information in a storage device. The 'storage device' addressed here should be distinguished from the 'data carrier' used to transport the individual cash slip information to the customer. The storage device is construed as a container for the information of a set of individual cash slips collected during a collecting session. As such, it stores in principal the same type of information as stored in one data carrier (individual cash slip information). However, it preferably stores this data as a sum of individual cash slip information (this is also called 'overall cash slip information' in the following). In addition, the storage may also store additional data related to the collection sessions during which the individual data sets were collected (e.g. region, type of store, session time parameters, number of intercepted customers, number of collected data carriers, etc.). In the preferred examples, this storage device is preferably located at a remote central site at which the collected data is evaluated and analysed. However, the storage device might as well be a transportable storage device used close to or inside a store (e.g. a memory unit inside a hand scanner used in front of a store to read the data carriers). When, in the following description the term 'remote' is used, it should indicate that recording the data is performed not immediately in front of or inside a store, but at a distinct central site at which the collected data is analysed. This, however, should not exclude the case that there exist more than one 'remote site' for evaluating the collected data.
The portions of the method described above are typically performed for some or, preferably every individual purchase of a customer. In contrast thereto, the portions of the method which follow refer to a post analysis of the collected raw data, which is preferably performed at a remote central analysing site.
In general, the process which follows concerns the evaluation of the collected raw data (respectively, overall cash slip information), which comprises calculating from the recorded data, estimated product-specific sales and/or other market-relevant data. This calculation process may comprise any kind of evaluations of the recorded data including statistical estimations based on the data sample.
The description which follows relates to some general aspects of the data collection process performed in or in front of a store during a collection session.
According to preferred examples which are described in more detail below, the collection process involves intercepting a number of customers leaving a store in order to gain cash slip information related to their individual purchase or shopping trip. Preferably, the customers are intercepted shortly after or shortly before leaving a store. Most preferably, the customers are intercepted after having left the store, in which case, the least interaction with the surveyed store becomes necessary. Typically, all customers leaving the store are observed and assistants try to intercept all or as many of the customers as possible. In order to guarantee good results of the final statistical evaluation of the recorded data, the total number of customers leaving the store should be recognised, even when some of the customers are not directly addressed by the assistants. Hence, 'intercepting a number of customers' is not restricted to a direct interaction between a customer and an assistant, but also comprises passive interactions necessary for registering customers which are not directly addressed. The interception of the customer(s) is performed by one assistant or a group of assistants familiar with the particularities of the data collection process.
Preferably, as many of the intercepted customers as possible are directly addressed in order to obtain from them the cash slip information related to their individual purchase. Gaining the cash slip information may be accomplished by taking the data carrier on which the cash slip information is stored from the customer(s) for a later analysis, or by reading the information in a short time period, during which the customer hands over the cash slip to the assistant.
The following description is directed to preferred examples, wherein reference is made to the accompanying drawings. In all Figures, the same reference signs indicate the same or substantially the same parts,
Fig. 1 is a schematic overview of a first example.
The dashed rectangles 1 and 2 in Fig. 1 indicate two sites, a local site at which a store 1 is located, and a central site 2, at which the evaluation of the collected data takes place. One or more check-out counters 3 are situated inside the store. The check-out counter 4 comprises a conveyer 5 on which consumers place their products 6. Each product on the conveyer 5 is guided over an optical product-code scanner 7 (e.g. a bar-code reader) which converts the code into one or more machine-readable signal(s). The signal is then transmitted to a point-of-sale terminal 8 which identifies the product 6 from the product-code. The point-of-sale terminal 8 has access to a database (not shown) which associates product related information to a product-code. Such product related information comprises product name, product type, piece price, weight price, etc. The point-of-sale terminal 8 identifies all products belonging to an individual purchase of a customer 9 and obtains the corresponding product information from the database.
Having completed the scanning process, the cash register displays the total sum of the purchase and outputs a data carrier 10, preferably a cash slip (or magnetic card of the customer), on which the identified product and price information is stored. Preferably, the cash slip is handed over (arrow 11) to the customer 9 who receives it and leaves the store (arrow 12) through one of the store entrances 13. In front of the store, one or more assistants 14 observe the customer 9 coming out of the store. Preferably, the assistants 14 immediately note the fact, that the customer has left the store in a notebook. If one of the assistants 14 is not occupied with another person, he addresses the customer 9 (arrow 15) in order to request the information stored on his cash slip 10. He asks the customer 9 if he would hand him his cash slip 10 and gathers the cash slip if the consumer 9 cooperates. Otherwise, he asks the customer 9 if he would let him know the contents of the cash slip and/or the total price noted on the cash slip. If the customer agrees, the assistant notes the total price of the customer's purchase. If the customer does not cooperate at all, the assistant notes one additional non-cooperating customer.
In the case that the customer 9 handed over his cash slip 10, the assistant 14 operates a hand scanner 17 in order to scan the cash slip information recorded on the cash slip. The cash slip information (31 in Fig. 5) is recorded in a storage device (21') of the hand scanner for later processing. In the case that the customer did not hand over his cash slip but communicated the contents of the cash slip and/or the total price of the cash slip, the assistant records the data in a portable data processing device. In the example shown in Fig. 1 this data processing device forms one common device with hand scanner 17 so that the assistant only needs a single technical device to do his job. Of course, the device 17 may also be used to count and record the number of non-cooperating customers.
In the case that none of the assistants 14 is free to address the customer 9, the customer is noted as one missed customer. Concerning data evaluation, such a missed customer can be treated similar to a non-cooperating customer.
In a preferred example there is always one assistant responsible for observing all customers leaving the store and noting a missed customer in the case that none of the other assistants is available for addressing this customer.
The above described processes are repeated for all customers 9 leaving the store during the collection session.
After completion of the collection session, the collected data (cash slip information of all observed customers, session parameters, etc.) is transmitted, via a communication network 18 such as the Internet, to a processing system 19 at the remote central site 2. The processing system comprises an input unit 20 for receiving the collected data and a second storage device 21 for recording the data. The processing system 19 is arranged for further evaluations of the recorded data which are described in more detail in Figs. 5, 6 and 7.
An alternative example is depicted in Fig. 2. Here, the data carrier 10 (or cash slip) is scanned (arrow 22) by a hand scanner 17 before being handed out (arrow 11) to the customer 9. The scanned cash slip information is then transmitted to the processing system 19 at the remote central site 2 where it is received by a receiving unit 20 and recorded in a storage device 21.
A further alternative example is shown in Fig. 3. Here, the process is basically the same as in Fig. 1 up to the point where the assistant 14 has intercepted and addressed the customer 9. The difference compared with the example shown in Fig. 1 is that the assistant 14, here, does not use a scanning device for reading the cash slip information. Instead, he transports (arrow 23) the data carrier 10 (or cash slip, in this particular embodiment) to the central site where a further assistant 24 enters (arrow 25), by, for example, an input unit, the cash slip information contained on the cash slip 10 into the data processing system 19 where it is stored in a storage device 21.
Besides, assistant 24 may also receive additional information such as total price information and collecting session parameters from the assistant 14 in order to enter them into the processing system 19. Preferably, each cash slip 10 is assigned a code when being entered into the processing system 19.
In contrast to these preferred examples, in a prior art method of collecting retail information, as depicted in Fig. 10, the point-of-sale terminal 8 is equipped with a special unit (not shown) which collects the information evaluated by the terminal 8 and transmits this information to a remote site for a further analysis. This prior art method requires specially adapted point-of-sale terminals making necessary the cooperation of the store holder.
The portion(s) of the method relating to the communication 15 between customer 9 and assistant 14, which have been already described above with respect to Fig. 1 are depicted in Fig. 4 in the form of a flow-diagram. Starting from the top of Fig. 4, the assistant 14 intercepts customer 9 when the customer leaves the store 2 (block 100) (the reference signs used here relate to Fig. 1). The fact that not every customer will be successfully intercepted (some will be missed when all assistants are occupied) is indicated by the decision 'successful?' (block 102) which follows. In the case that the interception is not successful, an assistant records one additional missed customer (block 104). In the case that the interception is successful, the assistant who intercepted the customer asks this customer if he would hand out his cash slip (block 106). In the case that the customer cooperates (block 108), the assistant collects the cash slip (block 110) and the assistant attempts to intercept the next customer (block 100). Otherwise, he asks the customer for the information (e.g., the total price) on the cash slip (block 112). In the case that the customer cooperates (block 114), the assistant records this information (block 118). If not, he records one addition non-cooperating customer (block 116).
Fig. 5 shows in more detail, how the data resulting from the collection process depicted in the example of Fig. 4 is evaluated. According to the description of Fig. 4, the example collection process results in a set of cash slip information comprising product-specific cash slip information of each of the individual purchases performed by the fully cooperating customers (28 in Fig. 5) which accepted to hand out their cash slip 10. One individual cash slip information is shown in Fig. 5 with reference sign 31. The individual cash slip information 31 obtained from one cash slip comprises detailed product-specific price information 33 and product-identification information 34. The set of individual cash slip information 31, when cumulated (indicated in Fig. 5 by the sum-sign), results in overall cash slip information 31' which is the collection of all individual cash slip information 31. In a preferred example implementation of this cumulation process, for each product type the number of pieces purchased by each customer is summed up. Alternatively, the amount of money spent by each customer on a particular product may be summed up, resulting in the total amount of money spent by all cooperating customers on a particular product type.
From the partly cooperating customers 29, which, for example, only communicated the total price of their cash slip, emerges a set of individual cash slip information 32 all of which comprises total price information 35 about an individual purchase. Likewise, the collection of this information results in overall cash slip information 32' comprising the summation of the total price information 35'.
From this overall total price information 35', a statistical estimation 36 (e.g. a ratio estimation) based on the detailed cash slip information 31' allows an estimation of the product-identification information 27 of all cooperating customers (all fully cooperating customers 28 and all partly cooperating customers 29).

An example of this first estimation process is given in the following:

Assuming that the overall cash slip information 32' of the partly cooperating customers indicates that they have spent together 50.000 € during a collection session.
Assuming further, that the overall cash slip information 31' of the fully cooperating customers 28 includes the following information:

overall total price of all purchases: 20.000 €
overall number of products purchased:
- Yoghurts: 2000 (a 2,00 €)
- Milk: 1000 (à 1,50 €)

It follows that the overall money spent by the fully cooperating customers 28 for particular products was:

Yoghurts: 4.000 €
Milk: 1.500 €

A ratio estimation now allows the research company to conclude on the overall money spent by the partly cooperating customers 29 on the particular products: $Yoghurts : (4.000 € / 20.000 €) \times 50.000 € = 10.000 €$
$Milk : (1.500 € / 20.000 €) \times 50.000 € = 3.750 €$
From this, the overall number of pieces per product purchased by the partly cooperating customers 28 is obtained, when taking into account the known piece price of the products:

Yoghurts: 5000
Milk: 2500

Therefore, the overall number of pieces per product purchased by the partly cooperating customers 29 and fully cooperating customers 28 together is:

Yoghurts: 7000
Milk: 3500

Thus, the ratio estimation 36 allows the research company to estimate product-identification information 37 of all cooperating customers (28, 29).
Next, the non-cooperating/missed customers 30 are included in the estimation process. The total number N of these customers is known from the data collection process. A second numerical estimation 38 allows the research company to conclude on the products bought by this group based on the estimated product-identification information of all cooperating customers 27.
Assuming, the total number N of non-cooperating/missed customers 30 has been counted to 5000, the number of partly cooperating customers 29 has been counted to 1500, and the number of fully cooperating customers was 1000, the following estimation is gained for the overall number of pieces per products bought by all customers in the collection session : $Yoghurts : (5000 + 1500 + 1000) / (1500 + 1000) \times 7000 Yoghurts = 21000 Yoghurts$
$Milk : (5000 + 1500 + 1000) / (1500 + 1000) \times 3500 Milk = 10500 Milk$
This result is indicated in Fig. 5 by the estimated product-identification information 39 of all customers in a collection session for a particular store observed.
Fig. 6 shows in a flow-diagram another processing mehod, in which the set of estimated product-identification information 39 obtained from the collection sessions for the surveyed stores is further evaluated in order to obtain information 43 for a complete reporting period. Usually the collection sessions do not cover the complete reporting period and not all stores of the respective type in the market. However, if the collection sessions are well balanced over the complete reporting period (days of a week, hours of a day, weeks in the reporting period), the collected and the estimated information can be projected such that meaningful conclusions are obtained for the whole reporting period for the particular store surveyed. This is indicated in Fig. 6 by a ratio estimation 41 which evaluates from information for all collection session time periods 42 and based on the set 40 of estimated product-identification information of all customers in a collection session for a particular store, estimated product-identification information of a complete reporting period for all surveyed stores. Finally, a numerical estimation 48 is performed. The result from the ratio estimation 41 is multiplied by the total number 49 of stores of the respective store type to estimate the product identification information of a complete reporting period for the complete market.
The basic information of the collection session time periods is their respective beginning and ending. But the projection process may also include further information about the reporting periods. It might, for example, be interesting to treat the last working days before Christmas holidays in a particular way since purchasing activities are typically increased in these days.
Other aspects of planning collection sessions, which are addressed here only very briefly, is the spatial and sociological distribution of the collection sessions. In order to yield meaningful results, balancing of the sessions with respect to such demands is also preferred.
Fig. 8 shows a table with an example distribution of collection sessions over the time of a day, the days of a week, and over regions covered by a retail report. In the properly balanced sample design of Fig. 8, the complete reporting period covers four weeks (1-4 in the first column of the table) and five regions (1-5 in the first row of the table). Regions 1 and 2 each decompose into two subregions 1 a, 1 b and, respectively, 2a, 2b ( subregions 1 a and 2a might be, e.g. of urban type, whereas subregions 2a and 2b might be, e.g. of rural type). The characters A, B and C listed in the table indicate particular times during a day. Here, character A indicates the time span from store opening at 9:00 to 12:15, character B indicates the time span from 12:15 to 15:15, and character B indicates the time span from 15:15 to store closing at 18:30.
In the case that imbalances occurred in the sample design or during the actual execution of data collection, these may be corrected by using a ratio-based projection of the sessions (e.g. according to a estimated turnover contribution of the sessions to the total turnover). The total turnover may, for example, be obtained as external information. Alternatively, imbalances in the collected data may be corrected on the basis of collected backdata from preceding months, if available.
A final evaluation 44 is shown in Fig. 7 in which product-specific sales information is gained based on the estimated product-identification information of a complete reporting period 43. This final data analysis may also comprise additional market information 46 (e.g., total number of stores of the respective types, regional information, social information, business information, etc.) and it may also comprise statistical considerations and conclusions based, e.g., on the number 47 of customers in the relevant market.
The additional market data indicated by reference sign 46 in Fig. 7 might also include information received from in-store visits in order to identify characteristics of particular products, and in order to record stocks which allow splitting and assigning sales of items where the cash slip label and the price is identical, such that a product-identification alone from the information on the cash slip is impossible.
The result of the method shown in Fig. 7 might, for example, be weekly sales for a virtual average store, for every store group included in the sample. From the sales information about such virtual average stores, sales information about the complete market can be obtained based on external market information such as the total number of existing stores in a particular store group.
The information obtained from the cash slips can be used to calculate, e.g.,

the average price in consumer units,
the average price equivalized (e.g. in liters or kilograms),
promotional sales (where store/week/item sales are supported by, e.g., a feature, a display or a price cut), and/or
the numerical distribution (the % of category turnover realized by the numerical distribution indicating if the selling stores are rather big stores or small stores).

Typical results obtained from the analysis might include market diagnostic information as follows:

U.S. households spent nearly 49 million dollars on Brand A products.
Brand A held a 64.8% share of Widget dollar sales.
1.79 million households purchased Brand C at least once.
2.0% of U.S. households made a Brand D purchase.
The average Brand D buying household spent $2.95 on Brand D.
Brand B buyers averaged 3.4 purchase occasions.
Brand A and B attract shoppers with small family sizes.

product type

In some examples the calculated data is smoothed by taking the average sales of the last four weeks in order increase the statistical quality of the data.
In other examples, the quality of the collected data is controlled by evaluating the time stamps on the cash slips and checking with them the recorder session time parameters.
It should be noted that even though in the description of the examples of Figs. 5 and 6, the information resulting from the estimation 36, 38 and 41 are indicated as product-identification information, this should not mean that this estimated information may not comprise other information such as price information or estimated price information. In general, the estimations 36, 38 and 41 in Figs. 5 and 6 should be understood as processing all available data and resulting in as much information as possible, obtainable from the collected data.
With the preferred examples, data related to the buying behaviour of consumers is collected without the necessity of intruding into existing point-of-sale terminals of participating stores and, furthermore, the collected data elevated is based on individual purchases of customers which allow conclusions on the household-level. As a consequence, the preferred examples enable reporting all purchases; even the ones from collectivities, impulse purchases, etc. Further, all households (even lower socio-economic households which are otherwise hard to report) participate in the study so that a fully representative sample of buyers is addressed. There is no bias due to households that love market research, the number of observations gathered is increased compared to prior art methods and the cost-value ratio is better.

Claims

A method of collecting and processing data related to consumer buying behaviour of customers of retail stores, the method comprising:
(a) receiving, at a sample of retail stores (1) during at least one collection session within which data is collected for a predetermined period of time, several data carriers (10) on each of which is stored cash slip information (31, 32) regarding the products (6) and related prices of a customer's individual purchase and each of which is output by a point-of-sale terminal (8) of a store;

(b) reading the data carrier (10) in order to obtain the cash slip information (31, 32) from the data carrier (10);

(c) recording the obtained cash slip information (31,32) in a storage device (21,21'); and

(d) calculating in a computing device estimated product-specific sales numbers (45) based on the recorded data.
The method of claim 1, in which the data carrier (10) is a magnetic storage medium, an optical storage medium, an USB stick, a magnetic card, or a print medium.
The method of claim 1 or 2, in which reading the data carrier (10) further comprises scanning the data carrier with a portable hand scanner (17) in or in front of the stores.
The method of claim 1 or 2, in which reading the data carrier (10) comprises scanning the data carrier (10) with a scanning device located near the point of sale terminal (8) in the stores.
The method of claim 1 or 2, in which reading the data carrier (10) is performed at a remote analysing site (2).
The method of claim 5, in which reading the data carrier (10) comprises scanning the data carrier (10) with a scanning device located at the remote analysing site (2).
The method of claim 5, in which reading the data carrier (10) comprises manually entering the cash slip information (31, 32) into a computing system (19) located at the remote analysing site (2).
The method of anyone of the preceding claims, further comprising:
intercepting a number of customers (9) leaving the stores (1); and

addressing each of these customers (9) in order to receive the data carrier (10).
The method of claim 8, further comprising:
observing the number of customers (9) which leave the store (1) but are not intercepted and/or the number of customers (9) which are addressed but do not cooperate.
The method of claim 8 or 9, further comprising
reading, from the data carrier (10), product-specific price information (33) and product-identification information (34), if the customer (9) agrees;
reading, from the data carrier (10), the total price information (35) of a customer's individual purchase, if a customer (9) refuses to communicate the complete cash slip information (31) on the data carrier but is willing to communicate the total price (35) of his individual purchase.
The method of claim 10, further comprising:
determining, from the total price information (35') and on the basis of the product-specific price information (33') and the product-identification information (34'), estimated product-identification information (37) of all cooperating customers (28, 29) within a collection session.
The method of claim 11, further comprising:
determining, from the total number of customers (N) which do not cooperate and/or are not intercepted and on the basis of the estimated product-identification information (37) of all cooperating customers (28, 29), estimated product-identification information (39) of all customers (28, 29, 30) within one collection session.
The method of claim 12, further comprising:
performing (a), (b) and (c) during a number of collection sessions;

recording the time periods (42) of the collection sessions during which (a), (b) and (c) are performed; and

determining, from the recorded time periods (42) and on the basis of the estimated product-identification information (39) of all customers (28, 29, 30) of one collection session, estimated product-identification information (39) of a complete reporting period.
The method of anyone of claims 11 to 13, in which determining estimated product-identification information comprises using a ratio-estimation (36) or a numerical estimation (38).
The method of anyone of the preceding claims, in which calculating estimated product-specific sales numbers is performed by a computing device (19).
A method of collecting and processing data related to the consumer buying behaviour of customers of retail stores, comprising:
(a) intercepting, at a sample of retail stores (1) during at least one collection session within which data is collected for a predetermined period of time, a number of customers (9) leaving the stores (1);

(b) addressing each of these customers (9) in order to obtain information (31) contained on the cash slip (10) regarding the products and related prices of their purchase in the store;

(c) recording the cash slip information (31) of these customers; and

(d) calculating in a computing device estimated product-specific sales numbers (45) based on the recorded data.
The method of claim 16, wherein addressing each of the customers comprises:
gathering from the customer (9) the cash slip (10), if a customer accepts to hand out his cash slip;

asking the customer (9) for the total price (35) of his cash slip (10), if he refuses to hand out his cash slip.
The method of claim 16 or 17, wherein recording the cash slip information (31) further comprises:
recording the number of customers which leave the store (1) but are not intercepted and/or the number of customers which are addressed but do not cooperate.
The method of claim 17, further comprising:
identifying, from the cash slip information (31') of the gathered cash slips, overall product-specific price information (33') and product-identification information (34') of a collection session;

identifying, from the recorded set of total cash slip prices (32), overall total price information (35') of a collection session;

identifying the total number of customers (N) which do not cooperate and/or are not intercepted within a collection session.
The method of claim 19, further comprising:
determining, from the overall total price information (35') and on the basis of the overall product-specific price information (33') and product-identification information (34'), estimated product-identification information (37) of all cooperating customers (28, 29) of a collection session;

determining, from the total number (N) of customers which do not cooperate and/or are not intercepted and on the basis of the estimated product-identification information (37) of all cooperating customers (28, 29), estimated product-identification information (39) of all customers of one collection session.
The method of claim 20, in which determining the estimated product-identification information (37) of all cooperating customers (28, 29) of a collection session is performed by a ratio-estimation (36).
The method of claim 20, in which determining the estimated product-identification information (39) of all customers (28, 29, 30) of a collection session is performed by a numerical estimation (38).
The method of any one of claims 16-22, further comprising:
performing (a), (b) and (c) during a number of collection sessions;

recording the time periods (42) of the collection sessions in which (a), (b) and (c) are performed; and

determining, from the recorded time periods (42) and on the basis of the estimated product-identification information (39) of all customers (28, 29, 30) of a collection session, estimated product-identification information (43) of a complete reporting period.
The method of any one of claims 19 to 23, in which the collection sessions are properly balanced to significantly cover the days of a week, the hours of the day, and the weeks within the complete reporting period.
The method of any one of claims 16-24, further comprising:
entering (25) into a computer system (19) and recording in a storage medium (21) of the computer system, the obtained cash slip information (31) comprising the actual beginning and ending of the collection session, and for each cash slip (10), the product-specific price information (33), the total price information (35), and/or the product-identification information (34).
The method of claim 25, in which an identification code is assigned to each cash slip (10) and recorded together with the corresponding cash slip information (31, 32).
The method of any one of claims 16-26, further comprising
evaluating time stamps included in the cash slip information (31, 32).