Identifying Relevant Information To answer the question, “Which salesperson is making the most sales?”, the data warehousing team decided to design a report that lists total sales by emp
Trang 1Example Data Warehouse 4 Identifying Sources 45
Which Salesperson Is Making the Most Sales?
Identifying Relevant Information
To answer the question, “Which salesperson is making the most sales?”, the data warehousing team decided to design a report that lists total sales by employee The following display shows an example of such a report
Display 5.1 Total Sales by Employee Report
The next step is to identify how such a report could be created
Identifying Sources
The data warehouse team examined existing tables to determine if they could be used to the create the report shown in the previous display They identified several tables that could be used These tables are described in the following sections
Source for Staff Information The STAFF table contains information about employees, such as name, ID, department, supervisor, and salary, as shown in the following display
Display 5.2 The STAFF Table
Trang 2Source for Organization Information The following ORGANIZATION table identifies the organization to which an employee belongs
Display 5.3 The ORGANIZATION Table
Source for Order Information The following ORDERS table contains information about orders placed with salespersons, including date, salesperson ID, type of order, and customer
Display 5.4 The ORDERS Table
Trang 3Example Data Warehouse 4 Identifying Sources 47
Source for Order Item Information
The following ORDER_ITEM table contains information about orders placed with the company, and includes product ID, amount ordered, price of items, and other data
Display 5.5 The ORDER_ITEM Table
Source for Customer Information
The following CUSTOMER table contains information about the customers who are placing orders with the company Information includes name, address, birthdate, and other data
Display 5.6 The CUSTOMER Table
In reviewing the previous tables, the data warehousing team identified the following issues:
3 The salesperson and salesperson ID must be correlated to determine sales
3 The sales totals for each order must be correlated with the correct salesperson
3 The sales for each salesperson must be totaled
3 Some information does not exist in current source tables New columns and tables must be created
Trang 4The next step is to specify the new tables that must be created in order to produce the desired reports
Identifying Targets
To simplify the SAS Data Integration Studio job that will be used to create the desired report, the team decided to combine certain columns from existing tables into a smaller number of new tables:
3 A new table will be created that joins the CUSTOMER, ORDERS, and ORDER_ITEMS tables: the ORDER_FACT table
3 A new table will be created that joins the STAFF and ORGANIZATION tables: the ORGANIZATION_DIM table
3 In order to answer the question of who made the most sales, the two new tables will be combined to create a third new table on which the report will be based The third new table will be called Total_Sales_by_Employee
By combining tables, the warehouse team can easily answer the specified question,
as well as create a diverse range of reports to answer other business questions Details about each new table are provided in the following sections
Target That Combines Order Information The ORDER_FACT table is created by joining the CUSTOMER, ORDERS, and ORDER_ITEMS tables The new table will include all order data, including salesperson
ID, customer, price, and quantity
Target That Combines Organization Information The ORGANIZATION_DIM table is created by joining the STAFF and ORGANIZATION tables The new table will include all employee information including name, ID, salary, department, and managers
Target That Lists Total Sales by Employee The Total_Sales_by_Employee table is created by joining the ORDER_FACT table and ORGANIZATION_DIM table It will be used to produce a report similar to the draft report shown in Display 5.1 on page 45 “Example: Using the Target Table Designer to Register SAS Tables” on page 140 describes how to specify metadata for the Total_Sales_by_Employee table “Example: Creating a Job That Joins Two Tables and Generates a Report” on page 150 describes the job that uses the
Total_Sales_by_Employee table to produce the report
Creating the Report
For an example of the report can be created in SAS Data Integration Studio, see
“Example: Creating a Job That Joins Two Tables and Generates a Report” on page 150
Trang 5Example Data Warehouse 4 Identifying Sources 49
What Are the Time and Place Dependencies of Product Sales?
Identifying Relevant Information
To answer the question, “What are the time and place dependencies of product sales?”, the data warehousing team decided to design a report that reports sales across
a time dimension and a geographic dimension The following display shows an example
of such a report
Display 5.7 Time and Place Dependencies for Sales Report
The next step is to identify how such a report could be created
Identifying Sources
Further questioning of the executive team revealed that it would be helpful to track sales across a customer dimension and an internal organization dimension as well as across the dimensions of time and geography Questions that require multiple
dimensions to be analyzed together can often be answered with online analytical processing (OLAP) Accordingly, the data warehousing team concluded that the question, “What are the time and place dependencies of product sales?”, could be answered most efficiently with OLAP
The data warehouse team examined existing tables to determine whether they could
be used as inputs to an OLAP data store that would produce reports similar to the report shown in Display 5.7 on page 49 They identified several tables that could be used These tables are described in the following sections
Sources Related to Customers The following tables can contribute to the customer dimension of the OLAP data store:
3 CUSTOMER table
3 CUSTOMER_TYPE table