Financial data normalization systems and methods

• US 10,332,135 B2
• Filed: 03/05/2012
• Issued: 06/25/2019
• Est. Priority Date: 04/12/2010
• Status: Active Grant

• Current Assignee: First Data Corporation (Fiserv Incorporated)
• Original Assignee: First Data Corporation (Fiserv Incorporated)
• Inventors: Tavares, Silvio, Fahy, Susan, Carlson, Dennis
• Primary Examiner(s): Delich, Stephanie Z
• Application Number:
  US13/412,451
• Publication Number:
  US 20120191506A1
• Time in Patent Office: 2,668 Days
• Field of Search: None
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

Systems and methods are described for generating indexed sales data by aggregating point of sale (POS) datasets are aggregated from transactions at a plurality of POS terminals. The POS datasets for each transaction include a transaction amount, a merchant classifier, and a transaction time. An industry subset of the aggregated POS datasets is obtained for a given timeframe based on the merchant classifier. This industry subset comprises transactions for a given industry. A sales value is calculated for the industry subset, and a monthly fluctuation factor is applied to the sales value. Also, a normalization factor is applied to the sales value based on a percentage of the sales value relative to an overall market size to obtain the indexed sales value.

Claims

1. A method for normalizing point of sale (POS) sales data, the method comprising:

aggregating, by a computer system, point of sale (POS) datasets from a plurality of POS terminals, each POS terminal being configured to collect transaction data as a function of transactions effectuated via the POS terminal, wherein the POS datasets for each transaction comprise a transaction amount, a merchant classifier, and a transaction time, and wherein the POS datasets comprise a percentage of transactions effectuated within an overall market;

obtaining, by the computer system, an industry subset of the aggregated POS datasets for a given timeframe based on the merchant classifier, wherein the industry subset comprises transactions for a given industry;

calculating a reliable portion of the industry subset, the reliable portion comprising only data having a statistically insignificant variability from a baseline;

calculating, by the computer system, a sales value for the reliable portion of the industry subset;

applying, by the computer system, a time-based fluctuation factor to the sales value to account for sales fluctuations that are related at least in part to seasonality;

applying a normalization factor to the sales value based on a percentage of the sales value in terms of dollars relative to a size of the overall market to obtain an indexed sales value for the given timeframe;

generating an interactive formatted graphical report showing one or more of a trend of the indexed sales value or a projected sales volume based on the indexed sales value, wherein the interactive formatted graphical report is automatically formatted using auto-graphics zones, and wherein the interactive formatted graphical report comprises a plurality of selectable views with each of the plurality of selectable views displaying a different subset of data when selected; and

transmitting the interactive formatted graphical report over a wireless communication channel to a user device;

wherein the interactive formatted graphical report causes the interactive formatted graphical report to display on the user device such that each of the plurality of selectable views is selectable by the user device to show each of the different subsets of data;

wherein the time-based fluctuation factor is calculated by using a time series of historical daily data from the POS datasets, and wherein the time-based fluctuation factor is further calculated by selecting a previous time frame, calculating daily sales for the reliable portion of the industry subset for at least some days in the time frame, and performing a statistical analysis of the daily sales for those days to obtain the time-based fluctuation factor.

2. A method as in claim 1, further comprising comparing the indexed sales value for the given timeframe to indexed sales values of other timeframes.

3. A method as in claim 1, further comprising using the indexed sales value to predict sales during a future timeframe.

4. A method as in claim 1, wherein the indexed sales value is in dollars and is calculated by dividing the sales value by the time-based fluctuation factor.

5. A method as in claim 1, further comprising applying a scaling factor to the indexed sales value.

6. A method as in claim 1, wherein the given timeframe comprises a calendar month, and wherein the fluctuation factor is a monthly fluctuation factor, and wherein the fluctuation factor is balanced against a norm of 1 such that a fluctuation factor less than 1 suggests a month will have less spend than an average month and a fluctuation factor greater than 1 suggests a month will have more spend than the average month.

7. A method as in claim 1, wherein the statistical analysis performed when calculating the time-based fluctuation factor comprises one of:

a least squares analysis, an autoregressive integrated moving average, and a seasonal autoregressive integrated moving average.

8. A system for projecting future sales data, the system comprising:

an aggregation subsystem, communicatively coupled with a point of sale (POS) network comprising a plurality of POS terminals, and configured to aggregate POS datasets from the plurality of POS terminals, each POS terminal being configured to collect transaction data as a function of transactions effectuated via the POS terminal, wherein the POS datasets for each transaction comprise a transaction amount, a merchant classifier, and a transaction time, and wherein the POS datasets comprise a percentage of transactions effectuated within an overall market;

a data storage subsystem, communicatively coupled with the aggregation subsystem, and configured to store the aggregated POS data from the plurality of POS terminals;

a processing subsystem, communicatively coupled with the data storage subsystem, and configured to project sales data by;

obtaining an industry subset of the aggregated POS datasets for a given timeframe based on the merchant classifier, wherein the industry subset comprises transactions for a given industry;

calculating a reliable portion of the industry subset, the reliable portion comprising only data having a statistically insignificant variability from a baseline;

calculating a sales value for the reliable portion of the industry subset;

applying a time-based fluctuation factor to the sales value to account for sales fluctuations that are related at least in part to seasonality, wherein the time-based fluctuation factor is calculated by using a time series of historical daily data from the POS datasets, and wherein the time-based fluctuation factor is further calculated by selecting a previous time frame, calculating daily sales for the reliable portion of the industry subset for at least some days in the time frame, and performing a statistical analysis of the daily sales for those days to obtain the time-based fluctuation factor;

applying a normalization factor to the sales value based on a percentage of the sales value in terms of dollars relative to a size of the overall market to obtain an indexed sales value for the given timeframe;

generating an interactive formatted graphical report showing one or more of a trend of the indexed sales value or a projected sales volume based on the indexed sales value, wherein the interactive formatted graphical report is automatically formatted using auto-graphics zones, and wherein the interactive formatted graphical report comprises a plurality of selectable views with each of the plurality of selectable views displaying a different subset of data when selected; and

transmitting the interactive formatted graphical report over a wireless communication channel to a user device;

wherein the interactive formatted graphical report causes the interactive formatted graphical report to display on the user device such that each of the plurality of selectable views is selectable by the user device to show each of the different subsets of data.

9. A system as in claim 8, wherein the processing subsystem is further configured to compare the indexed sales value for the given timeframe to indexed sales values of other timeframes.

10. A system as in claim 8, wherein the processing subsystem is further configured to use the indexed sales value to predict sales during a future timeframe.

11. A system as in claim 8, wherein the processing subsystem is further configured to calculate the indexed sales value in dollars.

12. A system as in claim 8, wherein the processing subsystem is further configured to apply a scaling factor to the indexed sales value.

13. A system as in claim 8, wherein the given timeframe comprises a calendar month, and wherein the fluctuation factor is a monthly fluctuation factor.

14. A system as in claim 8, wherein the time-based fluctuation factor is calculated by using a time series of historical daily data from the POS datasets.

15. A system for projecting sales data, the system comprising:

a data storage subsystem configured to store aggregated point of sale (POS) data from a plurality of POS terminals that are communicatively coupled with a POS network comprising a plurality of POS terminals, wherein the POS data comprises, for each transaction occurring at one of the POS terminals, a transaction amount, a merchant classifier, and a transaction time, and wherein the POS datasets comprise a percentage of transactions effectuated within an overall market;

a processing subsystem, communicatively coupled with the data storage subsystem, and configured to project sales data by;

obtaining an industry subset of the aggregated POS data for a given timeframe based on the merchant classifier, wherein the industry subset comprises transactions for a given industry;

calculating a reliable portion of the industry subset, the reliable portion comprising only data having a statistically insignificant variability from a baseline;

calculating a sales value for the reliable portion of the industry subset;

applying a time-based fluctuation factor to the sales value to account for fluctuations between the given timeframe and other timeframes, wherein the fluctuation factor is balanced against a norm of 1 such that a fluctuation factor less than 1 suggests a month will have less spend than an average month and a fluctuation factor great than 1 suggests a month will have more spend than the average month;

applying a normalization factor to the sales value based on a percentage of the sales value in terms of dollars relative to a size of the overall market to obtain an indexed sales value for the given timeframe;

using the indexed sales value to predict a projected sales value for a future timeframe;

generating an interactive formatted graphical report showing the projected sales value, wherein the interactive formatted graphical report is automatically formatted using auto-graphics zones, and wherein the interactive formatted graphical report comprises a plurality of selectable views with each of the plurality of selectable views displaying a different subset of data when selected; and

transmitting the interactive formatted graphical report over a wireless communication channel to a user device;

wherein the interactive formatted graphical report causes the interactive formatted graphical report to display on the user device such that each of the plurality of selectable views is selectable by the user device to show each of the different subsets of data.

16. A system as in claim 15, wherein the processing subsystem is further configured to calculate the indexed sales value in dollars.

17. A system as in claim 15, wherein the processing subsystem is further configured to apply a scaling factor to the indexed sales value.

18. A system as in claim 15, wherein the given timeframe comprises a calendar month, and wherein the fluctuation factor is a monthly fluctuation factor.

19. A system as in claim 18, wherein the monthly fluctuation factor is calculated by using a time series of historical daily data from the POS datasets.