mod xi 3 analyzing the data business objects xi

Module 3

Business Objects XI Web Intelligence

Data Analysis

2

Analyzing the Data

In Module 2 we viewed, refreshed, and navigated a Web Intelligence document. In module 3 we interact with a report further to discover new insights. You will be exploring data within an existing report to make it more meaningful. Such an analysis can range from viewing information in a different block style such as tabular or chart, to re-sorting, filtering, and drilling down into the details. This report-based analysis is aimed at report consumers, not just power users. Thus, many of the actions are available via right-mouse clicks. Also all the analysis discussed in this module is available via a zero-footprint browser; you do not have to launch the full-blown Java Report Panel

3

Analyzing the Data

However, while drill mode is supported in both the Java Server Pages (JSP) and Active Server Pages (ASP) versions of Web Intelligence, the HTML Interactive Viewer for on-report sorting and filtering is available only in the JSP version, you will have to modify the report to launch the Java Report panel to access some of the features discussed in this module.

4

Analyzing the Data

5

Analyzing the Data

The HTML Interactive Viewer

In order to open a document with the HTML Interactive Viewer, you must set your InfoView Preferences for Web Intelligence as shown in module 1. Note that with the Interactive Viewer, you now have an additional toolbar for formatting the report, an Edit button to modify the report or query, an Insert pull-down menu, and additional options in the Document and View pull-down menus. The left panel in the HTML Interactive viewer is called the Context panel.

6

Analyzing the Data


It has may more capabilities than the standard HTML viewer, including the ability to see additional objects in the query (but perhaps not displayed in the report) and a data summary that displays a description of each object. Report consumers as well as query authors can clearly see the universe object descriptions.

From the Document menu, you now have a Document > Properties option that displays the report author, document description, last modification date, and duration of the query refresh.

7

Analyzing the Data

Report Toolbar Modify a report or query


8

Analyzing the Data


From the View menu, you have additional option to view toolbars (Formatting, report, formula) as well as formatting and query preferences.

9

Analyzing the Data

Actions Available via the Report Toolbar

Button Name FunctionRemove Remove an entire cell or block, or invoke a drop-down

menu to remove a row or column from a tabular block.

Apply Quick Filter Apply a filter to a dimension or measure within the existing block.

Ascending Sort Sort the data values in ascending order. To see the sort priorities, right-click the block and select Sort > Properties from the pop-up menu.

Descending Sort Sort the data values in desceding order.

Insert break Insert a break to a dimension within the existing block.

Calculations Insert a sum, count, average, minimum, maximum, percentage to the table or crosstab.

Insert Row or Column Insert a row or column when yuou wish to add a new variable or formula to the block.

Duplicate Duplicate a block when you want to view gthe same data set in a different style.

Alerters Apply an alerter to the block. The alert must already be defined within the document. The ability to define an alerter exist only in the Java Report Panel.

10

Analyzing the Data

Block Types

In module 2, the different components of a web Intelligence report were explained and the concept of a block introduced. A block is a set of data that contains column headings, row headings, and data values. Business Objects allows several block types: table, crosstab, form, and chart. A master/detail is a type of a report but not specifically a block style. Within a master/detail report you specify a block style to appear in each section. A report can contain multiple block types. For example, in our previous document has both table block and a chart block. Each block can be populated from the same or a different query.

11

Analyzing the DataBlock Types

Use When

Table You want to detail information in a list or spreadsheet style. You may break the list into section in a master/detail report or with breaks. Tables can contain additional columns for many dimension, details, and measures.

Chart You want to discover trends and patterns by exploring summary, not detail. Numbers in a visual format. Charts are ideal for analyzing a limited number of measures by a limited number of dimensions.

Crosstabs Your want to compare measures by different points within a given dimension. A crosstab is a particular kind of table layout that lets you compare actual sales with budget sales, for example, or to compare data from one period to the next.

Form You want to see many details for a particular product, customer, or record. Instead of viewing a very wide report in which a lot of details are displayed on one row, a form lets you view the details vertically.

You can change block styles or duplicate a block and convert a style.

12

Table

When you initially create a new query, Web Intelligence uses the default block type of table. A table is a spreadsheet-style block that lists data in rows and columns. In this example, you will change a chart to a table to further analyze the individual data values. Assume your report displays a horizontal, stacked bar chart. This is a good way to quickly identify that Colorado and Maryland have the high count. However, you cannot readily tell from this chart the specific values. Nor can you easily see the performance by individual month.

Analyzing the Data

13

Analyzing the Data

Chart

14

Analyzing the Data

1. Left click on the chart to select the chart.

2. Right-click to invoke the pop_up menu. Select Turn Chart To.

3. A Turn To dialog appears. Here you can select a different chart style or, in this example, select Tables.

15

Analyzing the Data

16

4. Within each Type, you will see several thumbnails that indicate a subtype. Select Vertical Table and then click OK to close the Turn To Dialog. The horizontal chart now appears as follows, giving a clearer breakdown of the counts per month. Also, you can more easily see that counts are higher for Colorado than for Maryland only in 2005.

Analyzing the Data

17

Analyzing the Data

18

Crosstab

The vertical table in the preceding section displays individual column data values. Inter column relationship comparisons are a challenge. A crosstab is the best block style for quick comparisons and transposes what may originally be rows in a tabular report to column heading. People frequently use crosstabs for the following types of analysis: Incident counts by year, quarter, or month Financial measures by actual and budget Met SLA by region

Analyzing the Data

19

Crosstab

To convert a block to a crosstab:

1. Select the block

2. Right click to invoke the pop-up menu.

3. Select Turn Table To

4. Within the Turn To dialog, select Tables, and then select the Crosstab thumbnail

5. Click OK to close the dialog an see the crosstab.

Analyzing the Data

20

Analyzing the Data

Crosstab

21

Crosstab

With converting a block to a crosstab, you cannot specify which dimensions appear on top and which appear in each row. Web Intelligence takes its best guess, You can easily swap the row and column headings. For example, to have the years appear across the top and the states appear vertically, follow these steps:

1. Select the existing crosstab block that you wish to convert.

2. Right-click to invoke the pop-menu. Select Swap Axis.

3. The crosstab now appears as follows.

Using the Java report panel you can drag and drop column and row headings.

Analyzing the Data

22

Analyzing the Data

23

Chart

Charts are a powerful way to uncover trends and patterns in your data. Chars can transform a dense page of numbers into a visual that quickly highlights opportunities and problems. Web Intelligence provides numerous chart styles.

Often when you want to chart data, you only need a subset of the data that may appear in a table. In such a situation, try using the Duplicate button to first duplicate the table and then convert it to a chart.

Analyzing the Data

24

Chart Procedure

1. Select the tabular block that you wish to graph

2. From the Report toolbar, select Duplicate. If you wish the chart to appear to the right of the table, select To The Right. If you want the chart to appear beneath the table or at the end of the report choose At The End.

3. Modify the duplicated table to remove any extra columns or to further filter the data.

4. Select the duplicated table and right-click to invoke the pop-menu. Select Turn Table To.

5. From the Turn Table To dialog, select the desired chart style, then click OK.

Analyzing the Data

25

Insert a New Report and Block

Using the pop-up menus and Turn To are the easiest, and most intuitive way to view information in different block styles. Alternatively, from within the Interactive Viewer, you can also use the left panel to insert new block or change the style of an existing one. When you insert a new block, you are not creating a new query. Instead, you are displaying data that already exists in the document in a new block style.

Analyzing the Data

26

Master/Detail

A master/detail report is a particular kind of report in which a dimension value is used to separate different blocks. Master/detail reports allow you to analyze data within a particular subset. The earlier report slide is an example of a master/detail report and contains a crosstab and a 3D line chart for each Day.

Analyzing the Data

27

Master/Detail

Analyzing the Data

Click here to set as section

28

Master/Detail

To change an existing column in a report to a section header, click anywhere in the column and then right-click to invoke the pop-up menu. Assuming you have correctly selected a dimension column, the pop-up menu contains an option Set as Section.

In this example, the category existed as a column in the crosstab. However, if it did not exist in the block you can create a master/detail report by dragging an object onto the report page.

Analyzing the Data

29

Master/Detail

1. Display the Left panel by clicking the arrow to show left pane or selecting View > Left Panel.

2. From the Left panel Drop-down, choose Available Objects. This shows you all objects that may be stored as part of the query results but that are not necessarily displayed in the block,

3. Drag the Type object form the Left panel to the report. In order to create a section header, you must drop the object in the upper-left corner, just above the table. Pay careful attention to the shape and position of the mouse pointer.

Analyzing the Data

30

Master/Detail

Analyzing the Data

31

Master/Detail (HTML Report Panel)

Analyzing the Data

Drag and drop

4. When you drop the object into the report, If you do not get the desired results, click undo and try again.

32

Master/Detail (HTML Report Panel)

Note The HTML Interactive Viewer has a multilevel undo, But the Java panel allows you to undo only the last action

Analyzing the Data

33

Sort

Sorting data within a report allows you to rearrange the rows alphabetically, or if your sort is on a measure column, numerically. For example, you can use sort to find locations with the highest counts. When you want to analyze data within a subset by using Breaks, the breaks take priority over the sorts. In the following example, you will sort counts in descending order to quickly identify which locations have the highest counts.

1. Select the column you wish to sort within the table.

2. Web Intelligence highlights the selected column with a gray box.

3. Click Descending Sort from the Report toolbar.

Analyzing the Data

34

Sort Priority

When you want to have multiple sorts within a table, use the pop-up menu to set the sort priority. To access the sort priorities, select the table and then right click. From the pop-up menu, select Sort > Properties. For example, compare the priorities and results shown in the examples below. The data is sorted in descending order of counts and then by state.

Analyzing the Data

35

Sort

Analyzing the Data

36

Sort

Analyzing the Data

37

Sort

Analyzing the Data

38

Breaks

A Break will break the table into multiple pieces so that you can better analyze the data within a group. A master/detail report is somewhat similar, except that a break does not create a separate master cell. Also, a break column must exist in the table, whereas a section header in a master/detail report does not have to exist in the table. Finally, breaks give you additional options for keeping tables together for printing.

Once you create a break, you then use Calculation to generate subtotals. The first break column becomes the primary sort order. Often, this is the first column in the table, but it does not have to be.

Analyzing the Data

39

Breaks

Using the same table from the Sort section, add a break to the state column;

1. Select the State Column

2. Select Insert Break from the Report toolbar. Web Intelligence will insert the break and by default add the state as a break footer.

Caution The break button on the toolbar acts as a toggle to add or remove breaks. If you inadvertently click this when what you really wanted to do was check the break properties, you will lose all your subtotals and other break settings. Use the Undo to restore them and be sure to use the pop-up menu when you want to change Break Properties.

Analyzing the Data

40

Breaks

Analyzing the Data

Remove duplicate

Break Header

Break Footer

Use Breaks to group the data into what appears as mini-tables

41

Breaks

For each state, there is now a separate mini-table. Visually, you can now more clearly see the performance from one year to another.

Note Although breaks make the data appear as if they are in separate mini-tables, in the strict sense, Web Intelligence still treats the table as one block, an important nuance when formatting the table. The Format > Table options will be the same for both the California and Colorado mini tables

Analyzing the Data

42

Break Options

The effects of certain break options, such as break header, break footer, and remove duplicates can be modified by right-click the table and from the pop-up-menu, select Break > Properties.

Analyzing the Data

43

Analyzing the DataBreak Options

44

Break Options Break Header Repeats the column headings within

each mini-table. If your table contains more than one Break Header for each break, certain multiple rows of column headings. To avoid this, set the Break Header only on the first break (or one of the largest grouping for which you want to create a separate mini-table).

Break Footer Inserts a separate row beneath each mini-table. The Calculation subtotals (not yet inserted) will appear in the break footer.

Analyzing the Data

45

Break Options Remove Duplicates If you want the break value

repeated only once in the first row of the mini-table, use this option. As shown, the state name CA appears only in the first row of the table. Without this option set, California would be repeated in each row of the mini-table.

Center Value Across Break Works with the Remove Duplicates option so that the break value is centered within the rows of the table.

(Note This was not working as of XI R2)

Analyzing the Data

46

Break Options

Note Do not confuse the term break “value” with calculation values that may appear in the row of a break footer. A break value is the value from the dimension---in this example California and Colorado etc.. A calculation value would be the subtotalsWe will discuss table formatting in another learning module.

Printing mini-tables, created with breaks, are affected by the following options.

Start a New Page For each break level, print the mini-table on a new page. This print setting is useful when you intended to distribute each break level to a different person—for example, if you have different service managers and have a break on location.

Analyzing the Data

47

Break Options Avoid Page Breaks in Block use this option to force

Web Intelligence to start the break or mini-table on a new page. For breaks that contain many rows of data, particularly long tables it may not be possible to avoid a break in the middle of the mini-table. If the grouping of data is longer than a page and not the first grouping of data in the block, a page break will occur before the break and the mini-table will span multiple pages.

Repeat the Header on Every Page For each page break, repeat the column headings.

Analyzing the Data

48

Break Options Repeat Break value on new Page For mini-tables

that span multiple pages, repeat the break value in the first row of each page. If you choose to repeat the header as well, then the row with the break value is repeated beneath the column headings.

Analyzing the Data

49

Calculations

Calculations allow you to add subtotals to a table that contains breaks. If the table does not contain any breaks, then the calculations are grand totals for the entire block. In most instances, if you ad a calculation to one measure, you will want them on all measures in a report. The subtotals are inserted as formulas in the break footer. The Calculation dialog box is context-sensitive, depending on the object type (character, date, number). For example you may want to count the number of states in which you have stores, but this is not something you would sum or average.

Analyzing the Data

50

Calculations that Allow you to Create Subtotals

Analyzing the Data

Calculation Explanation Formula

SUM Adds the values for a particular measure. SUM is the default calculation when you first apply a calculation to a variable. However, SUM may not make sense for certain measures such as ratios and percentages.

=SUM([Measure Variable])

Count Counts the unique values within a break. Use this with dimension or detail objects

=COUNT([Dimension Variable])

51


Analyzing the Data


Average Calculates the average based on the individual measures displayed in the table. If you hide certain rows of data through a filter, the values from the hidden rows do not , by default affect the subtotal. Average is not a weighted average, so for things like average price, you may want to create a formula to get a weighted average or use extended syntax.

=AVERAGE([Measure Variable])

Minimum Displays the minimum value for a particular measure within the break.

=MIN([Measure Variable])

52


Analyzing the Data


Maximum Displays the maximum value for a particular measure within the break.

=MAX([Measure Variable])

Percentage Calculates the percentage contribution a particular break level makes to the grand total, In addition, it inserts a new column in the table that shows the percentage each row contributes to the individual break level. This calculation uses the Context Operator ForAll.

=Percentage([Measure Variable]) ForAll([Dimension Variable])

53

inserting calculation, Web Intelligence automatically add a grand total to the end of the report. Use the page navigation buttons to scroll to the last page to see the grand totals the following report shows a sum calculation and a percentage calculation in the break footer.

Analyzing the Data

54

Analyzing the Data

55

To add Total SR Count to the table1. Select the SR Count column.

2. Click the Calculation drop-down from the report toolbar.

3. Select Sum

Tip Consider formatting the break footer row so that your subtotals stand out.

Analyzing the Data

56

Note If you do not have your break options set to include a break footer, the calculations are inserted only as grand totals at the end of the report. Also, if you apply a break after you have inserted calculations, the calculations are not repeated in the break footer and remains as grand totals. To have them appear as subtotals in the break footer, first set the break , then insert the calculations.

Analyzing the Data

57

Filter

As you apply breaks, sorts and calculations, you may find that you want to focus on one or more subsets within the report. Filters allow you to restrict the number of rows displayed in the report. They do not affect the query results; they affect only the data currently displayed in the block. It’s important to understand that the calculations will also change to reflect only what is displayed. In the previous report, the table displayed several year of data. To focus on the year 2005, Insert a filter on year.

Analyzing the Data

58

Filter1. Position your mouse in the table on the year column

2. Select Apply Filter from the Report toolbar or right-click to invoke the pop-up menu and select Quick Filter.

3. Web Intelligence displays a list of values available in the report. You can either enter the text to filter on or choose the year from the list in example select 2005 and click >> to add it to the box on the right.

Note normally this is only a subset of the list of values in the entire database.

Analyzing the Data

59

Filter

4. By default, Web Intelligence uses the In List operator to allow you to select multiple values. However, you can use the drop-down to choose Between, Not Null, and so. Operators are discussed later.

5. Click OK to see the effect of the filter.

Analyzing the Data

Enter filter Value

Choose filter value

Available operators

Filter value

60

Removing the Filter

To remove the filter, you must use the left panel, Document Structure And Filters. You cannot use the quick Filter button. For example, assume you want to remove the filter on Year 2005. If you invoke the Quick Filter dialog and simply remove 2005 from the right hand box, Web Intelligence will attempt to filter the years for nulls.

1. Display the left panel by selecting view > Left Panel, or click the arrow icon.

2. From the left Panel drop-down, select Document Structure and Filters.

Analyzing the Data

61

Removing the Filter

Here you can see the underlying component of each report. The top row gives the name of the document. Immediately beneath this are the prompts or query filters. This is different from report filters that affect the display only. Here we are working with report filters. In this example, there is one report tab called “Breaks, Sums and percent” There are two blocks, a table block and a pie Block. Move your mouse pointer to this filter until it is highlighted with a red box.

Note the filter on the Table for Year in List 2005.

Analyzing the Data

62

Removing the Filter

4. Right-click to invoke a pop-up menu and select Remove Filter.

Analyzing the Data

Report tab

Document title

Filter

63

Measure Filters

Filters in Web Intelligence XI Release 2 are simpler in many respects from filters in previous versions of the full client. Measure filters used to require a defining a formula as did nesting filters. Now, you easily filter on measures by entering a value to compare to. For example, to find which states have a number of incidents less than 3,000 per quarter do the following.

1. Select the SR Count column.

2. Click Apply Filter from the Report toolbar or right-click to invoke the pop-up menu and select Quick Filter.

Analyzing the Data

64

Measure FiltersNote the list of values is not available for measures. Enter the value in

the left hand box, in this case 3000, and then click>>.

3. Change the operator form the default of Greater than to Less than or Equal to.

4. Click OK to apply the filter and close the dialog.

Analyzing the Data

Enter Value here

65

Report Filters

Report Filters are global filters that apply in all blocks in a report. In the section “Removing the Filter”, you saw that the report had both a tabular block and a pie block. Each block contains a separate filter on year for 2005. When you want all existing blocks and new blocks to inherit the same filter, apply them to the report via the Document Structure and Filters panel.

Select the component page Body or the report tab (not the individual report block) and right click to select Quick Filter. This dialog now contains all the dimensions and measures for this report, as shown.

Analyzing the Data

66

Report Filters

Under the column Available objects, these objects exist in the query but may not necessarily appear in one of your report blocks.

1. Drag the object for which you want to filter from the left panel, Result Objects, to the right, Report Filters.

2. To choose a filter value from a list, click Values.

3. To add a filter on a measure, enter the numeric value and then click Update Filter.

4. Select OK to close the Report Filter dialog.

Analyzing the Data

67

Nested Filters

Nested filters within a report are new in Web Intelligence. Previously, you could create a nested filter in a query but not within a report. You use nested filters when you need to group sets of conditions. For example, you have seen that locations California and Colorado have the highest incidents. They may be missing incident targets only when incident counts are below 3,000 in a given quarter. For low incident locations, though, the incident targets, you want to filter the report for (California and Colorado with incidents greater than 10,000) or (incidents in other states with revenue less than 3,000).

Analyzing the Data

68

Nested Filters

Analyzing the Data

Double-click to toggle AND / ORAdd Nested filters

Move line up or down

69

Nested Filters

Nesting filters can be rather tricky. First and foremost, you must have a clear understanding of what you want nested and what filters you wanted connected with an AND or OR.

1. To add a nesting or another AND / OR connector, select the block from within the Documents Structure And Filters on the left, and then select Filter to call the Filter Editor. If you have selected just an individual column, your dialog will be different from that shown in the filter editor.

2. Click the Add Nested Filter button on the right of the Filter dialog. This will insert an AND connector leftmost.

3. Double-click the AND to toggle it to OR. Use drag and drop to reposition the connectors and filters.

4. Click Apply.

Analyzing the Data

70

Nested Filters

Analyzing the Data

71

Ranking

Where as filters limit the rows returned in accordance with specific selection criteria. Ranking enables you to limit the rows according to top to bottom values of measures.

Note Ranking is only available in the java Report panel; it is not available in the Interactive viewer. To access the java Report panel, your default InfoView preferences for Web Intelligence must be set to use the Java Report panel for modifying reports (rather than the Query-html or the HTML Report panel … not to be confused with DHMTL Interactive Viewer that you have used so far). Select Web Intelligence Document tab from the Preferences link on the main toolbar, select the java Report panel and apply. Select Edit from the main toolbar to launch the java report panel.

Analyzing the Data

72

Ranking

Analyzing the Data

Select Edit to launch the Java Report panel

73

Ranking

When you apply a ranking, you apply it in terms of the measure you are trying to analyze and whether or not the rank should be within breaks on dimensions. The following table gives some business questions that show which measure you would base the ranking on.

Analyzing the Data

74

Ranking

Analyzing the Data

Business Question Measure to rank

Which Business Units have the Highest expense variance?

Sales Variance

What are the top-selling products? Sales Quantity

What customers have the most Incidents? Incident Count

Which warehouses have the most product on hand? Inventory

Which service providers have the shortest response time? (rank the lowest)

Response Time

75

Ranking

A challenge with the Web Intelligence ranking feature is to understand that there are different ways to rank the data. I would ague that the generally accepted approach is to rank according to a percentage of the total. For example, suppose you want the top 10 percent of incident counts. Earlier versions of Business Objects did not allow this. The ranking calculation mode of Cumulative percentage is new in Web Intelligence XI Release 2. There are four calculation modes for ranking.

Analyzing the Data

76

Ranking Count the number of records to retrieve based on the

measure. Percentage the percentage of records to retrieve

based on the measure. Cumulative Sum the cumulative sum that the

measure must not pass. Cumulative Percentage The cumulative sum of the

measure, represented as a percentage of the total, that the measure must not pass.

Analyzing the Data

77

Ranking

You can add a ranking to a table or chart:1. Select the block you wish to rank

2. Choose Ranking from the Java Report panel Toolbar.

3. Web Intelligence displays the Rank dialog. Choose if you want to filter your data on the top-ranked items or bottom. In this example, select Top and enter a value such 80.

4. In the Based On drop-down, choose the measure on which want the ranking based.

5. Choose the Calculation Mode as described earlier. Select Cumulative Percentage.

6. If you are working with a large data set that has multiple levels, check for Each and indicate the dimension on which to reset ranking.

7. Click OK to apply the ranking.

Analyzing the Data

78

Ranking

Analyzing the Data

79

Ranking

Note This ranking function works only on the rows within the local document, so it is a ranking within a subset of data. If you want a genuine ranking based on all data in the source data, this must be set up at query time, as described next.

Analyzing the Data

80

Server-Based Ranking

Server-based ranking is new in Business Objects XI R2 and requires that your database support the analytic function RANK. When your database supports this, an Add A Database Ranking button appears on the Query toolbar from within the Java Report panel.

Note this button is quite similar to the Add A Subquery button but is different in that it has a star.

Analyzing the Data

81


Analyzing the Data

Add a Database Ranking

Choose Top or Bottom Place Holders

82


In the last section, you used the report-based ranking to rank data within a report. The server-based ranking is much more powerful in that you can rank, data against the full data set. So, you can answer questions such as Who are my top ten vendors What locations have the 12 most incidents What are my bottom 5 cause codes

Analyzing the Data

83

Server-Based Ranking1. To perform a server-based ranking, follow these steps.

2. From the Java Report panel, select Edit Query.

3. Select Add A Database Ranking form the Query toolbar.

4. This will add a ranking row to the Query Filters section. Use the Top/Bottom toggle to select if you want the Top or Bottom ranking

5. By default, the value of Top 10 is selected. You can change the constant of 10 or enter a prompt.

6. Drag the dimension for which you wish to perform the ranking onto the placeholder “Drop a dimension here”, this is category.

7. Drag the measure to base the ranking on to the placeholder “Drop a measure here.” this is SR Count.

8. Select Run Query

Analyzing the Data

84

Drill

So far in this module you added many meaningful breaks, sort and calculations. With multidimensional analysis, you look at the same data but from different viewpoints. The viewpoints may be from different dimensions by drilling across or different levels of detail by drilling down.

A dimension is a kind of object by which you analyze numeric measures. Dimensions often have different levels or groupings associated with them called hierarchies. Multidimensional analysis is the process of analyzing data by different levels within the dimensions. Within Web Intelligence, you can perform multidimensional analysis only with Dimension objects, not with objects the designer has created as detail objects.

Analyzing the Data

85

Drill

Hierarchies allow you to analyze data by different levels of detail. Some hierarchies are very clear-cut, such as Time, Going from Year to Quarter to Month to Week to Day. There is a natural order. Geography hierarchies may also be predetermined, running from Continent to Country to State. When the Geography applies to marketing region. However, each company introduces its own variation. One company may group the Middle Ease and Africa together; another company may include Mexico as part of North America because it is part of NAFTA or will group it under a different management region such as Latin America. Many of these groupings are predetermined as part of your company’s reference data and built in to the ERP or data warehouse.

Analyzing the Data

86

Drill

The universe designer uses the hierarchies to build the default drill path you use for multidimensional analysis.

With Web Intelligence, you can drill down within a hierarchy, for example form year to quarter to month. You also can drill across by analyzing the current year or the past year; you are at the same level of detail, year, but you are changing the selection value that you are analyzing. You can analyze data by one dimension at a time (Time) or by several at once (Time, Geography, Product). You can also drill against multiple report formats, including tables, crosstabs, and charts.

Analyzing the Data

87

Drill

Fixed reports often deal with standard, recurring information needs and give you an overview of what is going on in your business. Multidimensional analysis is more exploratory and answers why, where, and when. For example, you may start with a standard management report that shows product sales for this year and last year. Sales for one product is lower this year than last year. You need to know why. So you begin to explore: where the sales bad for a particular region, sales person , or quarter?

Analyzing the Data

88

Understanding the Data

`You can easily get lost when drilling within a report. Therefore, it’s helpful to first understand how the different universes objects relate to one another and where you are drilling from and to. Here are two sample hierarchies time and location.

Analyzing the Data

89


Note Understand the hierarchical nature of you dimensions before drilling within a report

Analyzing the Data

90


It’s also important to understand that most drilling occurs within a local microcube. You are not querying the database again. For example, using time dimensions that contain Year, Quarter, Month. When you construct a query, you first specify which columns of data you want to retrieve from the database. Initially, the default report displays all columns of data. However, as you format the report, remove column, and create objects, you may hide certain columns. These columns still exist in the query and therefore in the microcube but are not displayed in the report. This allows you to drill from year to Quarter to Month without having to execute a new query.

Analyzing the Data

91

Analyzing the Data

QueryYear, Qtr, Month, Count

RDBMS(Data Mart or ERP)

92


If, however you want to drill down to a level of detail that does not exist in the document, to day for example, then new in release 2 is the ability to drill through. Web Intelligence will issue a new query and seamlessly display the additional level of detail. This drill-through to detail had previously been possible in the full client but not in Web Intelligence.

Analyzing the Data

93

How is the Data Aggregated

Web Intelligence aggregates measures in two ways: SQL, functions and projection aggregates are set by the universe designer. For the most part, you do not need to worry about projection aggregates if your universe has been designed correctly.

Analyzing the Data

94

What is drill?

You use drill to analyze the results displayed in reports. Drilling on reports lets you look deeper into your data to discover the details behind a good or bad summary result displayed in tables, charts, or sections.

Drill Down

With drill-down, you are looking for lower levels of detail within an existing hierarchy. In the following example, you will use Drill to discover which regions and service areas in Portland have the highest incidents.

Note Be sure you have first set the desired drill preferences for infoView.

Analyzing the Data

95

Drill Showing or hiding the Drill toolbar

When you start drill mode, the Drill toolbar automatically appears at the top of the drilled report. The toolbar displays the value(s) on which you drilled. These values filter the results displayed on the drilled report.

For example if you drill on the 2003 date for year, the values for Q1, Q2, Q3, and Q4 are the quarters for year 2003. This means that the quarterly values you drilled to are filtered by 2003.

Analyzing the Data

The Drill toolbar displays the value that filters the drilled results

96

Analyzing the Data

Drill

The Drill toolbar allows you to select other values in order to filter the results differently. For example, if you use the Drill toolbar to select 2002, the Q1, Q2, Q3, and Q4 results displayed on the drilled block would be quarters for year 2002.

Note If you are drilling in query drill mode, the drill toolbar shows only the drilled value in the list of values of the drilled dimension. For example, if you drill on 2002, you see 2002 only in the list of Year values.

97

Analyzing the Data

Drill

If the drilled report includes dimensions from multiple data providers, a ToolTip appears when you rest your cursor on the value displayed on the filter. The ToolTip displays the name of the query and the dimension for the value.

The ToolTip displays the names of the dimension and query for the value displayed on the Drill toolbar

98

Analyzing the Data

Setting your drill options

You set your drill options in InfoView. When you modify your drill options, the modifications are implemented the next time you start Drill mode.To set your drill options:

1. On the InfoView toolbar, click Preferences.

2. Click the Web Intelligence Document Preferences tab.

The Web Intelligence Document Preferences page appears.

3. In the Select a view format section, select the view format that corresponds to the tasks you want to perform on reports:

99

Analyzing the Data

Setting your drill options

If you want to... then select...perform drill analysis only, HTML perform drill analysis and interactive reporting, (Interactive reporting possibilities include sorting and filtering, adding predefined and custom calculations, modifying the selection

Interactive

of data displayed in tables and charts, and modifying formatting).

100

Analyzing the Data

Setting your drill options4. In the For each new drill session section, select the drill option

that corresponds to whether you want to keep a record of the report results or not before you modify those results using drill:

If you want Web Intelligence to... then select...retain a copy of the original report, so that you can compare the drilled results to the results on the original report,

Start drill on a duplicate report

drill on the existing report, so that the original report becomes modified by your drill actions,

Start drill on the existing report

101

Analyzing the Data

Setting your drill options5. In the General drill options section, select the options that you

would like to apply during your drill sessions:

If you want Web Intelligence to... then select...prompt you whenever a drill action requires a new query, so that Web Intelligence can extend the scope of analysis and add more data to reports,

Prompt if drill requires additional data Note Web Intelligence ignores the setting for this option if you are working in query drill mode because extension of the scope of analysis is implicit in query drill

synchronize drilling on all report blocks,

Synchronize drill on report blocks

hide the Drill toolbar when you switch to drill mode

Hide drill toolbar

102

Analyzing the Data

Setting your drill options1. To apply the new drill settings, click OK. The modifications are

implemented the next time you open a Web Intelligence document and start Drill mode.

Making documents drillable

To make a document drillable you need to:

• specify the scope of analysis included behind the displayed results

• understand the drill hierarchies on the universe upon which the document is based

103

Analyzing the Data


Specifying the scope of analysis for a document

The scope of analysis is the ability to include additional data in a document, which is not displayed in the reports. When you start Drill mode, you can drill on any of the displayed results down to the detailed data included in the scope of analysis. You specify the scope of analysis when you define the query for the document. The larger the scope of analysis, the more data is included in the document and the larger the document size.

You can select a predefined level for the scope of analysis or define your own custom level.

104

Analyzing the Data


Specifying the scope of analysis for a document Predefined – the scope of analysis is the same for each

dimension in the report. For example, if you set the scope of analysis at two levels, each dimension in the report will contain additional data two levels down.

Customized – you specify the dimensions in each class to which you want to be able to drill. Setting a custom scope of analysis reduces the length of time it takes Web Intelligence to run the query or refresh the reports. For example, a document might contain up to a thousand rows of data. If you know that you or other users will need to analyze only certain dimensions in the reports (for example, [Customer] or [Geography]), you can specify those dimensions in a custom scope of analysis.

105

Analyzing the Data


Setting the scope of analysis

In a universe, the objects within each class are usually represented in a hierarchy. For example, a class called Time period contains the objects Year, Quarter, Month, and Week arranged in a hierarchy as follows:

Note The hierarchy does not always correspond to the order of dimensions within a class. This is because the hierarchies can be customized at the universe level by your administrator, using Business Objects Designer.

106

Analyzing the Data

Making documents drillableWhat is a scope of analysis?

The scope of analysis for a query is extra data that you can retrieve from the database to give more details on the data returned by each of the objects in a query. This extra data does not appear in the initial result document, but it remains available in the data cube, so you can pull this data in to the report to allow you to access more detail at any time. This process of refining the data to lower levels of detail is called drilling down on an object.

In the universe, the scope of analysis corresponds to the hierarchical levels below the object selected for a query. For example, a scope of analysis of one level down for the object Year, would include the object Quarter, which appears immediately under Year.

107

Analyzing the Data

Making documents drillableWhat is a scope of analysis?

You can set this level when you build a query. It allows objects lower down the hierarchy to be included in the query, without them appearing in the Results Objects pane. The hierarchies in a universe allow you to choose your scope of analysis, and correspondingly the level of drill available.

You can also create a custom scope of analysis by selecting specific dimensions to the Scope of Analysis pane.

Note You cannot define a scope of analysis when working in Query Drill mode. Query Drill defines the scope automatically in response to drill actions.

108

Analyzing the Data

Making documents drillableSetting levels of analysis

When you set the scope of analysis for a query, you include dimensions one or more levels down the hierarchy tree for each dimension that you have added to the Result Objects pane.

When you run the query, the dimensions included in the scope of analysis are returned to the cube for that document, but are not projected onto the reports that the document contains. They can be added to the reports at any time, without having to run the query again. Values for the dimensions in the scope of analysis can also be viewed by switching a report to Drill mode, and then drilling down to them from the values displayed on the report.

109

Analyzing the Data

Making documents drillableHow many scope of analysis levels can you set?

You can set the following levels for scope of analysis:

Level Description

None Only the objects that appear in the Results Objects pane are included in the query.

One level down Two levels down Three levels down

For each object in the Results pane, one, two, or three objects lower down the hierarchy tree are included in the query. The data from these objects is stored in the cube until you add them to the document.

Custom All objects added manually to the Scope of Analysis panel are included in the query.

110

Analyzing the Data

Making documents drillableHow many scope of analysis levels can you set?

Note Including a scope of analysis in a document increases the document size significantly. This is because the data necessary for the scope you specify is saved with the document, even though it is not visible in the reports unless you start drill mode and drill down to the data to display the corresponding values.

In order to minimize the size of documents and optimize performance, it is recommend that you only include a scope of analysis in documents where you are certain that users will need to drill.

111

Analyzing the Data

Making documents drillableSetting the Scope of Analysis

To set the scope of analysis:1. Verify that you are in Query View.

The Edit Query button on the Web Intelligence toolbar is pressed in when you are in Query View.

2. Click the Show/Hide Scope of Analysis Pane button so that it appears pressed in. The Scope of Analysis panel appears at the bottom of the Result Objects pane. The default scope of analysis is None.

Each dimension in the Result Objects pane appears in the Scope of Analysis pane.

3. Click the down arrow in the Scope of Analysis drop-down list box.

4. Select a level for the scope of analysis.

The level appears in the list box and the dimensions that are hierarchically below each dimension in the Result Objects pane appear in the Scope of Analysis pane.

.

112

Analyzing the Data

Scope of Analysis Pane

113

Analyzing the Data

Making documents drillableSetting the Scope of Analysis

5. If you want to add selected dimensions to the scope of analysis or create a custom scope of analysis, select dimensions in the Query Manager and drag them across to the Scope of Analysis panel.

[Actual Start Date Month Number] and [Actual Start Date] appear under [Actual Start Date Year]

114

Analyzing the Data

Making documents drillableAbout drill paths and hierarchies

When you analyze data in drill mode, you move along a drill path. These paths are based on the dimension hierarchies set by the designer of the universe. Universe designers organize objects in classes in a hierarchy with the most summary objects at the top and the most detailed at the bottom. So if you want to make a high-level report, you know that your query should include objects at the top of the list. If you want to see more detailed information, you can then switch to Drill mode and drill down on each dimension value displayed in the reports.

115

Analyzing the Data

Making documents drillableAbout drill paths and hierarchies

For example, if the data from [Quarter] did not sufficiently explain a result, you could drill down to [Month] or [Week], depending on how the universe designer set up the hierarchy. When you drill to a different level, measures, such as a [Revenue] or [Margin], are recalculated accordingly.

Drill paths usually follow the same hierarchy order as the classes on a universe. For example, a class called Time typically includes the [Year] dimension at the top of the class, followed by the [Quarter], [Month], and [Week] dimensions. The hierarchies for drill within the Time hierarchy typically follow the same order, because users want to drill annual results to analyze details for quarter, month, and so on. However, the universe designer can also define custom hierarchies.

116

Analyzing the Data

The Time HierarchiesThe Actual Start Date Class

117

Analyzing the Data


Viewing drill hierarchies

To view the drill hierarchies on the universe on which a document is created:1. Open a document in Edit mode using the Java Report Panel.

2. Verify you are in Query View. The Edit Query button is pressed in when you are working in Query View.

3. Click the Data tab. If the Data tab is not visible, click the Show/Hide Manager button.

4. Click the Hierarchies radio button. The hierarchies appear.

5. To view the dimensions on each hierarchy, click the + sign next to each hierarchy.

118

Analyzing the Data



The hierarchies organize the dimensions with the most summary data at the top and the most detailed data at the bottom.

119

Analyzing the Data



Note A dimension can belong to several hierarchies. When you drill a result on a dimension that belongs to more than one hierarchy, Web Intelligence prompts you to select the drill path.

120

Analyzing the Data


Beginning your drill session

You can analyze the data behind the results displayed on the tables and charts in reports.

Note We assumes that your security profile includes the necessary access rights to perform drill.

121

Analyzing the Data


Starting drill mode

Drill mode lets you look deeper into your data to discover the details behind a good or bad summary result. To start drilling on a report you either: switch the report to Drill mode

or if the report is saved in drill mode, drill directly

Note Before you begin a drill session in Web Intelligence, you need to set your drill options in InfoView. For full information about each option and to find out how to select the appropriate options for your analysis.

122

Analyzing the Data


Switching to drill mode 1. With a document open in the Web Intelligence Java Report

Panel, verify you are in Report View. The Edit Report button is pressed in when you are working in Report View.

2. Select the report you want to drill. The Drill button appears in the Web Intelligence toolbar.

3. Click the Drill button. By default, a drill icon appears on the tab of the drillable report. Depending on the drill options you selected on the Web Intelligence Document Preferences page in InfoView, Web Intelligence either makes the selected report drillable or creates a duplicate of the selected report, in drill mode.

123

Analyzing the Data


Opening a report saved in Drill mode1. With a document open in the Web Intelligence Java Report

Panel, verify you are in Report View. The Edit Report button is pressed in when you are working in Report View.

2. Select the report you want to drill.

• If you have permission to drill, the report displays in drill mode.

124

Analyzing the Data


In Drill mode, the report tab displays the Drill icon

125

Analyzing the Data


If you do not have permission to drill, the report displays in Results View.

In view mode, the report tab displays the Results icon

126

Analyzing the Data


Drilling on dimensions in tables and sections

Dimensions typically represent character-type data, such as customer or business names, and dates. Web Intelligence makes calculations based on the dimensions in a report. For example, if you create a report that calculates a region’s total sales revenue for a given year, Web Intelligence calculates the [Sales revenue] measure based on the two dimensions: [State] and [Year].

127

Analyzing the Data



128

Analyzing the Data



When you drill on a dimension to see more data behind the displayed result, the sales revenue is calculated according to the values to which you drill.

On this report, when you drilldown on year 2003 to view thedetailed data for each quarteryear 2003, the sales revenue,sum, and average measures are calculated for each quarter.

129

Analyzing the Data



Note You cannot drill on detail objects.

You can drill on dimension values in tables, section cells, and charts on a report.

130

Analyzing the Data



Drilling on a table

131

Analyzing the Data



Drilling on a Section

132

Analyzing the Data


This section explains how to:• drill down to detailed information

• drill up to summary information

• drill by different dimensions to view a specific slice of data

• choose a drill path when more than one path is available

• drill to dimensions already displayed on a report block

• retrieve more data to the document to continue drilling

133

Analyzing the DataMaking documents drillable

Drill down to detailed information

You drill down to see the lower-level data that makes up the summary results displayed on reports. This helps explain why high or low results occurred.

Example: Using drill analysis to find out why sales decreased dramatically in 2003

In this example, you receive a report that shows sales revenue results for the accessories line at the eFashion retail store. The above crosstab shows that the Accessories line decreased in 2003.

134



To analyze more precisely when the decrease occurred, you drill down on the cell value 2003, to view the detailed data for each quarter.

135



When you drill down on the cell value 2003, a filter appears in the Drill toolbar to show that the quarterly values you have drilled to are filtered for the year 2003. The drilled chart clearly shows that the problem arose in Q4 of 2003.

The Drill toolbar displays the filter on Year, to show that the quarterly values displayed on the drilled chart are for the year 2003.

136



To find out which of the of the categories within the Accessories line was responsible for the drop in revenue, you drill down again on the cell value Accessories. The drilled crosstab shows which categories were responsible for low revenue in Q4.

137



Note If you try to drill to a dimension that is already displayed in another column or row of the same table, Web Intelligence automatically displays the next available dimension in the drill path.

138


Drilling down on a dimension value in a table or section cell

To drill down on a dimension value in a table or section cell:1. Verify you are in Drill mode.

The Drill button is pressed in when you are working in Drill mode and the drillable values display on the report as hyperlinks.

2. On a table or section cell, place your pointer over the dimension value on which you want to drill.

A ToolTip appears, showing the next dimension in the drill path.

139



3. Click the value.

The drilled table or section displays data one dimension level down.

The Drill toolbar, at the top of the report, displays the values from which you drilled. These values filter the values displayed on the drilled table.

140



Note If the drilled report includes dimensions from multiple data providers, the ToolTip displays the name of the query and the dimension for the value.

141


Drilling up

You drill up on a dimension value to see how the more detailed data aggregates to a higher-level result. For example, you may have drilled down on Year to examine data for each quarter. If you want to see how this data aggregates to yearly results, you can drill up.

When you drill up on a dimension value, you move along the drill path from lower- to higher-level data. For example, you may have drilled down on [Year] to [Quarter]. If you drill up on [Quarter], you return to [Year].

142


Drilling up

You can only drill up on a dimension value if:• you have previously drilled down to that dimension

or

• you have defined the appropriate drill path in the scope of analysis.

143


Drilling up on a dimension value

To drill up on a dimension value:1. Verify you are in Drill mode.


2. On a table or section cell, right-click the dimension value on which you want to drill up, then on the shortcut menu click Drill up.

Or

Click the Drill Up icon next to the dimension value you want to drill up.

144



Note If the table is a crosstab without headers that display the names of the dimensions on the table, then the Drill Up icon appears next to each value from which you can drill up.The report now displays data one dimension level up. The filters that filtered the value you drilled up from, are removed from the Drill toolbar.

The Drill Up icon appears next to the name ofeach dimension from which you can drill up.

145


Drilling by

When you drill down or up, you move through a hierarchy one dimension at a time. However, you can get another view of the data by slicing it in a different way, and then look at the data in other hierarchies. To do this, you drill by the other dimensions that interest you.

Example: Drilling by the Products hierarchy to slice sales revenue results by product

You work as regional manager for California in a retail clothing store, and have been sent the following report that shows quarterly sales revenue by state:

146


Drilling by

147


Drilling by

You are only interested in analyzing the results in the state of California. In addition, you want to analyze the sales revenue broken down by each product line you sell. To drill on California data, you place your pointer on the table cell that says California.

If you drilled down now, however, you would drill to results for each city within California, because [City] is the dimension below [State]. Instead, you select Drill by from the drill menu and then you navigate through the dimensions on the Products hierarchy by selecting the sub-menus until you reach the [Lines] dimension.

148


Drilling by

149

Analyzing the DataDrilling by

The drilled report displays the detailed sales revenue results for each product line sold in California.

150

Analyzing the DataMaking documents drillable Drilling by a dimension value

To drill by a dimension value:1. Verify you are in Drill mode.


2. On a table or section cell, right-click the dimension value you want to drill by.

151


A shortcut menu appears, displaying the available drill paths.

152


3. Place your pointer on Drill by, then on the class to which you want to drill.

4. Click the dimension to which you want to drill.

Note You can only Drill by to a dimension that is included in the scope of analysis of the document.

The Drill toolbar displays the value that filter the results displayed on drilled table. This drilled table displays results for 2001 Q1.

153


Choosing a drill path when more than one is possible

A dimension can belong to several classes. When you drill down on such a dimension value, Web Intelligence does not know which drill path to follow. You must define the drill path.

Choosing a drill path when more than one is available

To choose a drill path when more than one is available:1. Verify you are in Drill mode.


2. Drill on a dimension value in a table or section cell.

154

Analyzing the Data


Choosing a drill path when more than one is possible

The Select Drill Path dialog box appears.

155


Choosing a drill path when more than one is possible3. Select the path you want to drill.

4. Click OK.

Note If the dimension value you choose to drill on is the result of a previous drill, the drill path is already known. Therefore you do not need to select a drill path.

156


Retrieving more levels of data to the report

When you are drilling a Web Intelligence report, you may want to drill up or down to a dimension that lies outside the scope of analysis defined for the document. To return the additional data, Web Intelligence must run a new query that includes the additional dimension(s) you specify. This is called extending the scope of analysis.

When you extend the scope of analysis, you can select filters to limit the data Web Intelligence retrieves when running the new query. This saves time, if you are working with a large document. If you have several reports in one document, and you select filters when you extend the scope of analysis, those filters affect all the reports in the document.

157



When you extend the scope of analysis, you can select filters to limit the data Web Intelligence retrieves when running the new query. This saves time, if you are working with a large document. If you have several reports in one document, and you select filters when you extend the scope of analysis, those filters affect all the reports in the document.

Note You will not be given the option to select filters, if your Drill options are set not to display the Extend the Scope of Analysis prompt message during drill. In this case, Web Intelligence automatically returns data for all the values on the dimensions you are drilling.

158


Retrieving more levels of data to the report Drilling out of the scope of analysis

To drill out of the scope of analysis:1. Verify you are in Drill mode.


2. Drill on a dimension value.

A ToolTip informs you that a new query is necessary to return the additional data to the document.

159


Retrieving more levels of data to the report Drilling out of the scope of analysis

If your Drill options are set for Web Intelligence to prompt you when a drill action requires a new query, the Extend the Scope of Analysis dialog box appears.

160

Analyzing the Data Drilling out of the scope of analysis

Dimensions you can opt to include in the new query

Filters you can opt to apply to the new query

161


The dialog box lists the dimensions in the hierarchy above and below the drilled value. The dimensions already included in the document are checked.

The dialog box also displays the filters that you can select to filter the new query. For example, in the illustration above Web Intelligence suggests that you filter the query by limiting the additional data for Lines to data for the Accessories line. If you don’t select the filter, Web Intelligence will return data for all product lines to the document.

162


3. Select the check box(es) next to the dimension(s) to which you want to drill.

4. Select the check box(es) next to the filters you want to use to filter the query.

5. Click OK.

Web Intelligence runs a new query and returns the additional data to the document.

The results for the dimension you drilled to appear on the table.

Note You need permission from your administrator to drill out of the scope of analysis during a drill session.

163


Drilling on measures in tables and sections

When you drill on a measure value, Web Intelligence performs the drill action one level down for each related dimension in the block and displays the new measure calculation for the displayed dimensions.

Example: Drill on annual sales revenue results to see the breakdown by city and quarter

For example, you drill down on the year 2003 sales revenue value for California, which is displayed on a crosstab that shows sales revenue by year in by state.

164



The Drill ToolTipshows you whichdimension levelsyou will drill to

165



The drilled report displays sales revenue by quarter by city for California – the state on which you drilled.

The Drill toolbar shows you which values for year and state are filtering the results displayed in the drilled crosstab

166


167

Analyzing the Data



168



169

Drill Down

To begin drill mode, select Drill mode from the main toolbar

Analyzing the Data

170

Analyzing the Data

In the General drill options section, select the options that you would

like to apply during your drill sessions:

mod xi 3 analyzing the data business objects xi

Business

data chart

data report toolbar

data block types

masterdetail report

chart block

report consumers

report author

existing report