WO1992021089A1

WO1992021089A1 - Retail account management system

Info

Publication number: WO1992021089A1
Application number: PCT/US1992/004049
Authority: WO
Inventors: Joseph Anstey
Original assignee: Joseph Anstey
Priority date: 1991-05-15
Filing date: 1992-05-13
Publication date: 1992-11-26
Also published as: AU2007792A

Abstract

A retail account management system (10) comprises structure for collecting data (12, 14, 16, 18), such as shipments, merchandising, execution, cost and financial data, from at least two sources; structure (20), operatively connected to the collecting structure, for selecting the desired collected data and for converting the desired collecting data into a language compatible with the retail management system; and structure (22) operatively connected to the selecting structure, for receiving the desired collected data and for integrating, merging, updating and storing the desired collected data. The system further comprises structure (24), operatively connected to the receiving structure, for updating and modeling the response and forecasting parameters; and structure (26), operatively connected to the updating and modeling structure, for displaying a model in order to evaluate, monitor and plan merchandising programs for retail accounts.

Description

RETAIL ACCOUNT MANAGEMENT SYSTEM BACKGROUND OF THE INVENTION

1. Field of the Invention

While the present invention is subject to a wide range of applications, it is particularly suited as a system for managing retail accounts in the packaged consumer goods categories. More specifically, the system is provided for understanding, evaluating and planning for merchandising expenditures with respect to retail trade.

It is estimated that packaged goods companies spend between forty and fifty billion dollars in the retail trade to promote their goods or products. To date, no method has been available to advise the manufacturers or retailers whether these expenditures are profitable.

Consequently, there are enormous inefficiencies in the distribution systems for the goods. These inefficiencies include: a) sales to consumers at lower prices than would be profitable in view of the increase in associated volume; b) sales from manufacturers to retailers at significant discounts when a large part of the volume shipped on discount is actually sold at full retail price due to forward purchasing and diversion between geographies; c) production inefficiencies and bottlenecks; and d) payment of unnecessary fees by manufacturers. 2. Description of the Prior Art

In the past, retail account management systems, such as the SPAR computer based program, developed and distributed by the SPAR Corporation of Tarrytown, New York, and the Glendinning TP3 computer based program developed and distributed by Glendinning Associates of Westport, Connecticut were used to analyze retail promotional programs. Both of these programs are limited in that their models for estimating incremental sales are based solely on shipments of product from a manufacturer or distributor to a retailer's warehouse. Also, these programs were developed prior to the availability of point of sales scanners and, therefore, fail to utilize this new technology. These systems do not, therefore, reflect the present retail sales situation. Further, the systems are deficient in that they were developed for headquarter's use, are not applicable on personal computers (PC's) , are not user friendly, do not employ optimization techniques, and are not attractive to the field sales force.

SUMMARY OF THE INVENTION

Against the foregoing background, it is an object of the present invention to provide a retail account management system which is capable of understanding, evaluating and planning for merchandising expenditures with respect to the retail trade.

It is another object of the present invention to provide such a retail account management system which incorporates a portable PC based solution, as well as traditional desktop PC and mainframe solutions.

It is a further object of the present invention to provide a retail account management system which can communicate with large-scale proprietary and corporate databases for updating purposes.

It is still a further object of the present invention to provide such a retail account management system which uses as the key measure of incremental sales information that which is collected at a retail outlet.

It is still another object of the present invention to provide such a retail account management system which integrates and incorporates retail sales at the account level with factory shipments and costs, to obtain marginal utilities of different promotional events and/or strategies for each product so that optimization of marketing strategies can be accomplished locally, regionally and nationally.

It is yet another object of the present invention to provide a retail account management system which incorporates tracking, evaluation, forecasting, simulation, mathematical modeling, reporting, graphics and optimization in a PC computer system.

To the accomplishments of the foregoing objects and advantages, the present invention, in brief summary, includes a retail account management system which comprises means for collecting data, such as shipments, merchandising, execution, cost and financial data, from at least two sources; means, operatively connected to the collecting means, for selecting the desired collected data and for converting the desired collected data into a language compatible with the retail management system; means, operatively connected to the selecting means, for receiving the desired collected data and for integrating, merging, updating and storing the desired collected data; means, operatively connected to the receiving means, for updating and modeling the response and forecasting parameters; and means, operatively connected to the updating and modeling means, for displaying a model in order to evaluate, monitor and plan merchandising programs for retail accounts.

DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of the overall retail acc management system;

FIG. 2 is a block diagram of a system for collectin retail sales that which may be used in the management system of Fig. 1;

FIG. 3 is a flow diagram of the software system of t management system of Fig. 1.

FIG. 4 is a flow diagram of the allocator section of the management system of Fig. 1;

FIG. 5 is a flow diagram of the planner section of t management system of Fig. 1;

FIG. 6 is a flow diagram of the monitor section of t management system of Fig. 1; and

FIGS. 7 through 12 are Charts I through VI, respectively, of information used in the management syst of Fig. 1; -1-

BRIEF DESCRIPTION OF THE PREFERRED EMBODIMENTS

Referring to Figure 1, the retail account management system of the present invention is generally represented by reference numeral 10. The system 10 includes means 12 for collecting retail sales and merchandising performance data for one or more products from one or more points of sale or from scanner sales vendors or retailers. The system 10 also includes means 14 for collecting manufacturers' (suppliers) shipment data relating to one or more products incorporated in the retail sales data, means 16 for collecting trade merchandising cost and execution data from the supplier of the one or more products, and means 18 for collecting product financial data from the supplier of the one or more products. It is possible, although not preferably, to combine means 12, 14 and 16 into a single means.

The system 10 also includes means 20, which is directly connected to each of the above means, for transferring and converting the collected data from the language of the above collection means to the language of the present retail account management system 10. Means 20 provides for the back and forth communication between itself and each of the four collection means to convert the collected data to the language of the present system 10. The system further includes means 22, connected to converting means 20, for receiving selected, converted, collected data. The receiving means 22 includes a procedure for integrating, merging, updating and storing the converted, collected data. System 10 also provides means 24 that updates response and forecasting parameters whereby the optimal model for managing retail accounts can be generated and maintained and means 26, operatively connected to means 24, for evaluating, planning and optimizing merchandising programs for and among retail accounts and for displaying the models.

Information or data relating to retail sales is found in means 12 for collecting retail sales data. This data is collected by the retailer and typically purchased by a scan data company on a tape. For example, this data can be sales in approximately thirty grocery stores which provide a representative sample of approximately 230 grocery stores. The scan data company will manipulate and project these data to a selected, specified geography and in turn sell access to its database to manufacturers.

The retail sales to the consumer are, preferably, collected by the universal product code (UPC) search system located at each cash register at the point of sale. The data obtained from the UPC code includes identification of the specific goods and the price. As shown in Fig. 2, the data from each sale and, preferably, from a number of different locations within a single store, as well as in different stores within selected account, market, regional and national locations, is transferred from the scanners 28 at each cash register to a single location 30. The single location 30 can include a computer, preferably a mainframe, which is connected to the scanners 28 through direct means, such as transmission cables 34, or by any other means such as a conventional modem. In addition to the scanned data, other data, such as the quantity of goods purchased and the date and location of each sale, can also be collected by collecting means 12, through scanners 28, or by a conventional program associated with the computer 30. It is also within the terms of the present system 10 to input data from redeemed coupons by conventional means, such as scanning the coupon encoded with a UPC code or manually inputting the information directly into the computer 30. All of the collected data can be stored in the computer 30.

It is also within the terms of the present system 10 to transfer the retail sales data into any transfer means, such as, for example, a storage tape, a data storage diskette, or an electronic transfer modem, which is suitable for transferring the information to another location. The shipment data collected in means 14 is typically from the manufacturer of the packaged and promoted product and may include quantity shipments from the factory and the date of the shipments.

The means 16 for collecting trade merchandising, cost and execution data may comprise data or information such as, e.g. the physical condition of the stores, advertisements in newspapers and price discounts of the products being promoted. The information is not limited to these examples and any desired information which effects the market relating to the product can be collected. The collecting means 16 can be a computer having data storage into which the collected market data can be stored. It is, however, also within the terms of the present system 10 to collect the merchandising performance data by other conventional means, such as, for example, manual collection and then direct transfer of the information into the transferring means 20. The means 14 for collecting manufactures' shipment data and means 16 for collecting trade merchandising, cost and execution data, and perhaps, means 18 for collecting product financial data are typically a computer maintained by the manufacturer, into which such data has been entered via accounting system for financial reporting or by field personnel. The trade merchandising cost and execution counts data found in means 16 is trade administration data such as financial information relating to product, e.g. discounts and execution information relating to promotions and cost discounts. While the collected financial data found in means 18 comprises data generated by the supplier relating to the cost of producing and the profit from the sale of the specific goods. All of the collected information or data can be accessed by the transferring means 20 and converted by that means into system acceptable language and then is inputted into receiving means 22.

Thus, transferring means 20 can convert and transfer the sales data from a scan data company by selectively stripping out, pursuant to instructions, data relating to the specific goods or product line being analyzed. The remaining data will not be transferred. In addition, the scan data company may collect and maintain merchandising performance data and this data will be accessed through means 20.

The transferring means 20 includes a computer program that extracts the information from the various collecting means. The retail sales data and the merchandising performance data in means 12 are typically available from collections of information supplied by scan data companies, such as A. C. Nielsen of Northbrook, Illinois and Information Resources Inc. of Chicago, Illinois. These companies typically purchase the sales information collected on the mainframe computers of the retailers, transfer the information via tape to their own computers and then re-sell the information to their customers, usually packaged goods manufacturers.

The receiving means 22, as illustrated in FIG. 1, is an essential aspect of the system 10. It provides for receiving the collected data from means 12, 14, 16 and 18. The receiving means 22 includes algorithms for integrating, merging, updating and storing the selected data from means 12, 14, 16 and 18.

There are three basic, novel steps or procedures used in receiving means 22. The first step integrates into a single data base, data based on specific time periods.

Specifically, the first step projects the scanned retail sales data to represent the total retail sales for each product in each account in order to provide a set of artificial intelligence that determines the true values for retail sales in all stores for a selected account. In particular, data from means 14 is used to expand the data from means 12 to reflect the total retail sales account on an item by item basis. The first step projects the scanned retail sales data to represent total retail volume by (1) calculating the ratio of total stores for a retail account to stores within the geography defined by the scanner data vendor; (2) if shipments are not available for at least two (2) years, then the ratio is used to project the total retail situation; (3) if shipments are available, the ratio of shipments to the scanner volume is calculated for each item, adjusting appropriately for the time lag between shipments and retail sales; and (4) the reasonability of the adjustment factors is checked by comparing them to the store ratio in (1) above, then use the shipment to scanner volume ratios for reasonable ratios, and store ratio for others.

The second step follows sequentially after the first step and includes editing and correcting the retail sales price data. The second step basically estimates true retail shelf prices from integrated scanner and trade execution data since it determines what the actual retail sales price should be in each period of time, such as a week, based on the periods immediately prior to and after the period in question. This second step also factors in the merchandising performance data from means 12 by comparing the data from means 12 (which has been expanded to include the data from means 14) to the data from means 16 to provide corrected retail sales price data. The second step may include sub steps, such as:

1. Round to "9" the last digits of all prices greater than 40 cents in all weeks. If the last digit is greater than 5, round "up" to next "9" and if the last digit is equal to or less than 5, round "down" to the previous "9".

2. Compare the promoted weeks in the scanner data to the actual events in the execution data to obtain the true price in the promoted weeks.

3. Set the initial week's shelf price based on one of the following listed in the order of priority: (a) the non-promoted scanner price, if merchandising in less than or equal to 40% of account; (b) the moving average scanner price; (c) the previous week's shelf price; (d) the shelf price of the first week where merchandising occurs in less than 40% of the retailer stores if it is essentially the same value as the current week; and (e) future week's shelf price if the the shelf price for three of the following four weeks after the future week is the same as that future week.

4. For each week, accurately establish the actual shelf price level and, importantly, the weeks in which price increases and decreases occur, by the following steps: (a) the plus and minus range percentage (that varies by item) of the actual shelf price is established for each item so that small statistical differences are not read as true differences in price; (b) if merchandising occurred in less than 40% of the retailer stores and the current week shelf price is essentially the same as the previous week's, maintain the previous week's shelf price; (c) if the current week's shelf price is significantly lower than the previous week's, check the following four weeks' shelf prices and if three out of four are essentially the same as the current week's lower shelf price, then use the current week's shelf price for the next four weeks. If not, then use the previous week's price. Conversely, if the current week's shelf price is significantly higher than the previous week's, then check the next four weeks to establish whether the shelf price should be changed upward in the current week. Also, if merchandising occurred in more than 40% of the retail stores, omit the current week and use the shelf price of the first subsequent week in which merchandising occurs in less than 40% of the retailer's stores.

5. If a price change occurs or it is the middle of a calendar quarter, the weekly shelf price is compared to the actual manufacturer suggested retail price for that account. If it is within two (2) times the above range established for price changes, the manufacturer's suggested retail price is used or else the current week's edited scanner price is used.

The third step in receiving means 22 includes editing and correcting the retail sales price data by factoring in feature ad types and prices (merchandising performance data) . The third step provides a set of artificial intelligence that combines the actual merchandising data received from means 16 with the retail sales price data derived from the second step in order to combine and correct the data of the second step.

The third step includes: (1) identifying the merchandise event in scanner data and (2) checking the manufacturer's actual event file for current or previous week. If there is no advertising or event (ad/event) for the current week, check previous week and if a match is found, use the manufacturer type and price. If no match is found, use the scanner type and price. The third step also includes determining that: (3) a coupon promoted price roughly equals the shelf price and there is no manufacturer event, the promoted price equals 1/2 of shelf price; (4) promoted price is always less than or equal to the shelf price; (5) if feature type is buy-one, get-one free and promoted price greater than or equal to 70%, their promoted price equals 1/2 the shelf price; (6) identify events weeks unmatched in scanner for manufacturer's events by (a) taking the manufacturer event date; (b) reviewing scanner week up to 1 week before and 4 week after Manufacturer Event Date; (c) eliminating any week with matched feature; (d) eliminating any week following a matched feature; (e) finding the best performing week based on "% increase in volume over base"; and (f) testing for expected % increase in sales from response matrix calculation for this item, this date, feature type, price and account or market; (g) if the actual response is greater than 50% of the expected response, then insert manufacture event for this week to make it a feature week; and (h) if the actual response is less than 50% of the expected response ignore the manufacture's event.

To account for unresponsive ads, model an expected response for each merchandising event week. If the actual response is greater than 50% of expected, keep the event. If the actual response is less than or equal to 50% of expected response, look at previous week or next week. If the response in the previous and/or the following week is greater than that in the event week and above the expected 50% response, move the feature to the highest week. If the other two weeks are same but greater than 50%, move to the previous week. If all three weeks are the same and over 50% of the expected responses, use the current week. If there is less than a 50% response in all three weeks, change the feature to temporary price reduction.

Referring to means 24, means 24 may be a part of another means (not shown) or, as shown in Fig. 1, a separate means operatively connected to receiving means 22. Means 24 includes basically two novel steps for updating model response and forecasting parameters so that an optimal model for managing retail accounts can be generated and maintained.

Specifically, means 24 provides a forecasting step for forecasting the base volume trend line. This step fits the optimal trend forecast to historical trend data. It is a regression analysis. The second part of this forecasting step models the trend line. The second step of means 24 is used to understand the response of sales. This second step is a modeling step that fits data curves with the correct mathematical formula. This is also a regression analysis of sales response data curves to merchandising and promotional price discounts. This step also contains artificial intelligence that predicts what a response would be in cases where no sample data exists or is found.

The forecasting step has the following substeps: (1) create a fifty-two week moving average total volumes for x-52 weeks, with x being the total number of weeks; (2) fit three types of curves, namely y = A + BX, with y being the moving average trend volume and X being time in weeks. Log y = A + BX and Exp (y) = A + BX,

with B = n x (x*y) - x * v x² - ( x)²

where n = the number of observations

n n and A = XI i=l i=l n

Then, test for the best fit by calculating

R² = (yci - y)²

(Yi - Y)² i

with yi = actual value, yci= predicted value and y = average of actual y's. Thereafter, choose the model with largest R². The next substep is to impose damping on the growth slopes for Log model slope is greater than 0 and exponential model slope is less than 0. Thereafter, calculate Moving Average Trend (MAT) Value using the formula: MAT Value M_+n) = MATVALUE i_+n_! + B, and lastly calculate the weekly value for forecast week i using the formula:i = MAT Value ^ - (MAT Value ^__₁ - Weekly Value i_52) for forecast period.

The modeling step includes modeling and calibrating volume response to various merchandising tactics at different prices. Basically, using the responses under various merchandising conditions defined in the scanner database and corrected for type and price in means 22, the following steps are used to establish price elasticities under various merchandising conditions. Calculate the weighted (usually volume based) average of the increase in sales for all observations where the price discount is zero. Then calculate the average price elasticity for all observations by calculating the weighted average of the volume increase across all observations for a particular merchandising condition and divide the results by the weighted average of the price discount for the same observations. Thereafter, make sure the elasticities for feature advertising and the volume response at zero discount are greater than or equal to that for a temporary price reduction. Then, calculate the elasticity for various types of feature ads accompanied by displays, using the above steps and make sure the response at zero discount and the calculated elasticity are greater than or equal to the values for the feature ads without displays.

A number of alternative mathematical formulations are used, the best "fit" procedure described below is used to select the formula to use.

Using the above responses, calculate the expected total and incremental volumes for each of the past 104 weeks and classify the 104 weeks into various modeled groups, such as TPR, display, feature advertising by type, and feature advertising with display. For each group, calculate the ratio of expected to actual responses and choose the factors derived from the closest model. Where retail sales baseline information is not available from specific retails accounts, means 24 will also create and model retail sales and baseline data for missing accounts.

Referring to FIG. 3, there is provided an illustration of the flow path between the primary components of the entire system known by the trademark PROMOMAX which is owned by Applied Information for Marketing, Inc. The system uses the data of means 22 remodeled or updated in eans 24, the flow amongst the Monitor 40, the account Planner 42, the Allocator 44 and the updated data base 46 as follows. The update means 22 and 24 provides input directly into the Monitor 40 and Planner 42. The Monitor 40 allows the user to evaluate past merchandising events in terms of shipments, retail sales, incremental profits and costs. Once types of events and associated costs are specified in Planner 42, data from the Planner is inputted into the Allocator 44 which then allocates available funds optimally across specific products, types of merchandising events, time of year, retail accounts, and even markets, desired.

Referring to FIG. 4, there is illustrated the Monitor 40 that is programmed for tracking information and evaluating the information against projections. This enables the monitoring and reporting of information, including the historical data base and reported information. The information generated can be reported by any conventional means such as, for example, with graphics on a video monitor. In addition, variance analysis can be carried out with the information collected and reported by the Monitor 40.

Referring to FIG. 5, there is illustrated a flow diagram of the Planner 42 for planning events and programs at market and account levels. The Planner 42 is suitable for planning events, such as promotions within items or lines of products to be included in the planned promotion. This aspect of the system 10 is particularly related to retail accounts, markets and districts. For example, the Planner 42 can merge data from the following blocks: the two year historical trend block 64 including shipments, total retail sales, baseline retail sales volume (without merchandising) shelf price, promoted prices, retail promotion performance, merchandising execution, the merchandising response block 66 including sales response to various types of merchandising and price discounts, the event/tactic timing and cost block 68 including data about when events are planned, their characteristics and how much they will cost, the item strategy block 72 relating to suggested retail pricing, product distribution, etc.

Each of the blocks 64 through 72 are then combined in block 74 through a simulation model that forecasts the following: a) block 76 relating to the sales to the manufacturer; b) block 78 relating to the profits to the manufacturer; c) block 80 relating to the sales to the retailer; and d) block 82 relating to the profits to the retailer. After account planning is completed, the information from the Planner 42 is fed back to the Monitor 40 in order to track actual results or plan and identify variances. As any of the above inputs in block 74 change, the above results will change.

Referring to FIG. 6, there is illustrated a flow diagram of the Allocator 44 to plan the allocation of the trade funds between markets, accounts, products and merchandising strategies. The Allocator or allocation system 44 interconnects information between the following blocks: the planner marginal utilities 46 including information relating to sales and profits for types of merchandising; the sales objectives 48 relating to volume goals for specific markets, accounts and product items within them; the promotion objectives 50 relating to types and numbers of events expected to be for each product; the profit objectives 52 relating to expected regional and market area contributions to profits; the cost budgets 54 relating to market areas, region or nationally, and the marketing objectives 56 relating to global volume and marketing objectives for brands and the line of products marketed.

The data from blocks 46 through 56 are inputted into the optimizer block 58 which allocates funds to each product by account that will provide the highest volume or profits whichever is chosen as primary. The optimizer 58 optimizes the allocations as to markets, accounts, brand/items and strategies. The optimization function is a mathematical linear on non-linear programming algorithm. The output from the optimizer 58 is transferred into the allocations block 60. The information from the Allocator 44 is fed back to the Planner 42 in order to refine and present the merchandisin program for each retail account.

Referring to Fig. 1, evaluation, planning and optimization means 26 is operatively connected to the updating means 24. The evaluation means 26 may include a video monitor or any conventional apparatus such as a printer or plotter for displaying and representing results.

In operation, a manufacturer of prepackaged goods, desires to determine the effectiveness of sales promotions in order to understand, evaluate and plan for merchandisin expenditures with respect to the retail trade. To accomplish this analysis, they can purchase a software system which merges a great deal of data relating to retai sales and merchandising, shipment data, merchandising cost and execution data, and financial data of specific goods o product lines as described herein.

Using the information in the receiving means 22, chart such as the following can be produced for analyzing the effect of various alternative merchandising programs on sales and profits. For example. Chart I, shown in Figure 7, illustrates the historic and planned profit and loss for a specific product based on various marketing variables such as pricing and retail promotions, such as feature advertising and product displays.

Chart II, shown in Fig. 8, shows the weekly trend and merchandising activity and provides historical sales results on a weekly basis. Essentially the chart illustrates the shipments and retail sales and compares them with the base line expected sales. This enables the determination of the increase in sales due to retail promotion. Moreover, this chart compares the shipment information with respect to sales with retail sales. Thus, the degree to which retailers purchase more product than is necessary to support a retail promotion can be assessed. The chart shows the consumer response from promotions and the level of performance from the retailer. This enables the manufacturer to assess the degree of retail performance for specific historical events and the associated consumer sales response relative to the amounts of product purchases in anticipation of the event by the retailer account. Chart III, shown in Fig. 9, shows the percent increase of sales at different discount levels in conjunction with different types of retail promotion and advertising schemes. (This shows the merchandising response factors used in means 66 of planner 42, and updated in means 24).

Chart IV, shown in Fig. 10, shows the effect of a certain type of event, i.e. the sale of goods with a specific redemption allowance and type of advertisement and the sales and profit implications for both the manufacturer and the retailer (This shows the results from the simulation model of Fig. 5) .

The results for all events for an item or all events for a retail customer account can be viewed on a screen shown in Chart V of Fig. 11. The result achieved by the operation on the information with the program described herein is to provide planned weekly data that can then be compared with the historical information. Typically, a graphic display on a video monitor can be used to overlay a trace of the historical and planned weekly information. An example of this type of analysis is provided in Chart VI shown in Fig. 12. It is of course within the terms of the present system to reproduce the graphic display in any conventional means such as printing directly onto paper. The results are also useful in that it can provide a model for monitoring and planning.

Monitoring includes tracking information and evaluating the information against projections. This function includes the use of an historical data base. The information generated can be reported using graphics. If desired, the system can be setup to automatically forecast updated information. In addition, variance analysis can be carried out with the information collected and generated with the system.

The planning function is suitable for planning events, such as promotions, as well as planning programs. This aspect of the system is particularly related to the market or account. For example, the Planner 42 can use the two year historical trend, the two year merchandizing response, the event/tactic cost, the pricing, the line of products and the relationship between the distribution, events and tactics to forecast the following: a) the sales to the manufacturer; b) the profits to the manufacturer; c) the sales to the retailer; and, d) the profits to the retailer.

For planning, the volume of the promoted goods is separated from the volume of the base goods. The base goods are defined as the volume of goods sold at retail if no promotions occurred. The system is able to forecast th weekly trend in the base demand for the "planning horizon". The "planning horizon" is defined as any period from one quarter to one year. The forecast is accomplishe using statistical models which assess the trend in the historical data base incorporating factors that may offset that trend in the future such as various information which affects consumer demand, for example, pricing, advertisement, etc. Typically, the volume growth can be forecasted for up to seventy-eight weeks. Once this projection is accomplished, an estimate can be made as to how much each promotion will yield, e.g. the incremental volume by applying the historical merchandising response for particular types of promotion and price discounts.

Furthermore, since the system forecasts expected retail volume, it helps the manufacturer determine now which product the retailer should buy to fulfill demand for an event thus removing the possibility for the retailer to purchase product at a discount that will be e sold at full price later ("Forward Buy") .

Next, the screens provided in the planner can be changed by altering the merchandising variables so as to evaluate alternative strategies. In effect, the system acts as a financial simulation model. It will yield the incremental sales resulting from the promotion, and also the net profit for the manufacturer and the retailer.

The system is then able to prepare a summary of the various events and rank them in different ways, such as by highest to lowest volume or by profit. This enables the user to optimize a plan by choosing events with highest payout, or removing non-profitable events. The user can also plot a plan versus the historical data. The information can then be presented to the retailer to show why the plan is favorable. This provides a negotiating tool for use with the retailer, especially in laptop format.

The significant aspect of this invention is the ability to combine consumer or retail sales data obtained from the retailer, with financial and shipping data of the manufacturer in-pc-based mathematical models to provide a profitability analysis by event, by item, by account or across all accounts within a district, region or nationally.

A further significant aspect of this invention is its portability by a fixed sales organization since it is designed to run on "laptop" personal computers and can thus facilitate profitable selling. A still further significant aspect of this invention is the ability to relate actual or expected retail sales data to actual or expected manufacturer shipments so as to avoid the economic inefficiency of "Forward Buy" and diversion.

It is apparent that there has been provided in accordance with this invention a retail account management system which fully satisfies the objects, means and advantages set forth hereinbefore. While the invention has been described in combination with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art in light of the foregoing description. Accordingly, it is intended to embrace all such alternatives, modifications and variations as fall within the the spirit and broad scope of the appended claims.

Claims

WHEREFORE . I CLAIM:

1. A retail account management system, comprising:

means for collecting data from at least two sources;

means, operatively connected to the collecting means, for selecting the desired collected data and for converting the desired collected data into a language compatible with the retail management system;

means, operatively connected to the selecting means, for receiving the desired collected data and for integrating, merging, updating and storing the desired collected data;

means, operatively connected to the receiving means, for updating and modeling a baseline volume, forecasting and response parameters; and

means, operatively connected to the updating and modeling means, for displaying a model in order to monitor, plan and evaluate alternative merchandising events and programs for retail accounts.

2. The system of claim 1, wherein the collecting means is a scanner.

3. The system of claim 1, wherein the receiving means includes a plurality of steps.

4. The system of claim 3, wherein the plurality of steps is at least three steps.

5. The system of claim 4, wherein one of the three steps integrates into a single data base, data based on specific time periods by projecting the scanned retail sales data to represent total account sales for each retail product by providing a set of artificial intelligence that determines true values for retail sales in all stores within a selected retail account.

6. The system of claim 4, wherein the second of the three steps includes editing and correcting the retail price data by providing a set of artificial intelligence that compares the edited and corrected data to other data from the collecting means to provide corrected retail price data.

7. The system of claim 4, wherein the third of the three steps includes editing and correcting the retail merchandising data by providing a set of artificial intelligence that combines the actual merchandising data received from the collecting means with the retail performance data derived from the second one of the plurality of collecting means in order to combine and correct the data from the second step.

8. The system of claim 1, wherein the updating and modeling means includes a plurality of steps.

9. The system of claim 8, wherein the plurality of steps is at least two steps.

10. The system of claim 9, wherein one of the two steps is a step for forecasting the base volume or trend line and for modeling the trend line.

11. The system of claim 9, wherein the second of the two steps is a modeling step that updates merchandising response factors for various retail merchandising conditions and price discounts.

12. A retail account management system, comprising:

means for collecting retail sales and merchandising performance data covering one or more products from one or more points of sale;

means for collecting manufacturer's shipment data relating to the products incorporated in the retail sales data at the one or more points of sale;

means for collecting trade merchandising cost and execution data for the products;

means for collecting product financial data for the products;

means, operatively connected to all four collecting means, for selecting and converting the desired collected data;

means, operatively connected to the selecting means, for receiving the desired collected data and for integrating, merging, updating and storing the desired collected data; means, operatively connected to the receiving means, for modeling and updating the response and forecasting parameters; and

means, operatively connected to the modeling means, for displaying the model in order to evaluate, monitor and plan merchandising programs for retail accounts and markets.

13. The system of claim 12, wherein the receiving means includes at least three steps.

14. The system of claim 13, wherein one of the three steps integrates into a single data base, data based on specific time periods by projecting the scanned retail sales data to represent total account sales for each retail product to provide a set of artificial intelligence that determines true values for retail sales in all stores within a selected retail account.

15. The system of claim 14, wherein the second of the three steps includes editing and correcting the retail sales price data by providing a set of artificial intelligence that compares the data from the means for collecting retail sales, sales price and merchandising performance data to the data from the means for collecting trade merchandising cost and execution data to provide corrected retail sales price data.

16. The system of claim 15, wherein the third of the three steps includes editing and correcting the retail merchandising performance data by providing a set of artificial intelligence that selectively combines the actual merchandising performance data with the data received from the means for collecting merchandising cost and execution data to provide corrected promoted price and merchandising data.

17. The system of claim 12, wherein the updating and modeling means includes at least two steps.

18. The system of claim 9, wherein one of the two steps is a step for forecasting the base volume or trend line and for modeling the trend line based on fitting procedures of historical trends.

19. The system of claim 18, wherein the second of the two steps is a response modeling step that fits data curves with the correct mathematical formula.