International Symposium on Software Composition

Pivot table

A pivot table is a table of statistics that summarizes the data of a more extensive table (such as from a database, spreadsheet, or business intelligence program). This summary might include sums, averages, or other statistics, which the pivot table groups together in a meaningful way.

Pivot tables are a technique in data processing. They enable a person to arrange and rearrange (or "pivot") statistics in order to draw attention to useful information.

In their book Pivot Table Data Crunching, Bill Jelen and Mike Alexander refer to Pito Salas as the "father of pivot tables". While working on a concept for a new program that would eventually become Lotus Improv, Salas noted that spreadsheets have patterns of data. A tool that could help the user recognize these patterns would help to build advanced data models quickly. With Improv, users could define and store sets of categories, then change views by dragging category names with the mouse. This core functionality would provide the model for pivot tables.

Lotus Development released Improv in 1991 on the NeXT platform. A few months after the release of Improv, Brio Technology published a standalone Macintosh implementation, called DataPivot (with technology eventually patented in 1999). Borland purchased the DataPivot technology in 1992 and implemented it in their own spreadsheet application, Quattro Pro.

Using the example above, software will find all distinct values for Region. In this case, they are: North, South, East, West. Furthermore, it will find all distinct values for Ship Date. Based on the aggregation type, sum, it will summarize the fact, the quantities of Unit, and display them in a multidimensional chart. In the example above, the first datum is 66. This number was obtained by finding all records where both Region was East and Ship Date was 1/31/2005, and adding the Units of that collection of records (i.e., cells E2 to E7) together to get a final result.

Pivot tables are not created automatically. For example, in Microsoft Excel one must first select the entire data in the original table and then go to the Insert tab and select "Pivot Table" (or "Pivot Chart"). The user then has the option of either inserting the pivot table into an existing sheet or creating a new sheet to house the pivot table. A pivot table field list is provided to the user which lists all the column headers present in the data. For instance, if a table represents sales data of a company, it might include Date of sale, Sales person, Item sold, Color of item, Units sold, Per unit price, and Total price. This makes the data more readily accessible.

Report filter is used to apply a filter to an entire table. For example, if the "Color of Item" field is dragged to this area, then the table constructed will have a report filter inserted above the table. This report filter will have drop-down options (Black, Red, and White in the example above). When an option is chosen from this drop-down list ("Black" in this example), then the table that would be visible will contain only the data from those rows that have the "Color of Item= Black".

Excel pivot tables include the feature to directly query an online analytical processing (OLAP) server for retrieving data instead of getting the data from an Excel spreadsheet. On this configuration a pivot table is a simple client of an OLAP server. Excel's PivotTable not only allows for connecting to Microsoft's Analysis Service, but to any XML for Analysis (XMLA) OLAP standard-compliant server.