Table of contents
Authors create report layouts by incorporating visuals and additional elements such as text boxes, buttons, shapes, and images onto pages. These elements are set up independently of the semantic model fields.
During the design phase, the process of including and adjusting a report visual follows these steps:
Choose a visual type, for example, a bar chart.
Associate semantic model fields, visible in the Fields pane, with the visual's field wells. In a bar chart, these wells comprise Y-axis, X-axis, Legend, Small multiples, and Tooltips.
Customize the associated fields. This includes the ability to rename them or decide whether they should be summarized. If summarization is chosen, you can specify the method of summarization.
Implement various formatting options such as axis properties, data labels, and others.
The following example shows how to configure the analytic query for a report visual.
1.Include a visual representation of a stacked column chart on the report page. Apply a filter to the page using the Fiscal Year attribute from the Date table, specifically choosing FY2020.
2.Organize the visual by incorporating the Month attribute from the Date table into the X-axis well.
3.Provide a summary for the visual by integrating the Sales Amount attribute from the Sales table into the Y-axis well.
Fields represent a comprehensive term denoting model resources utilized for configuring visuals. These fields encompass three distinct model resources:
Columns
Hierarchy levels
Measures
Columns
Utilize columns to filter, group, and summarize values. Numeric columns commonly undergo summarization methods such as sum, count, distinct count, minimum, maximum, average, median, standard deviation, or variance. Text columns can be summarized using first (alphabetic order), last, count, or distinct count, while date columns can be summarized with earliest, latest, count, or distinct count.
During the design phase, the data modeler has the ability to configure the default summarization property for each column. This property can be assigned any supported summarization type or set to "Do not summarize." Opting for the latter implies that the column is primarily intended for grouping, not for numeric summarization. For instance, if a numeric column in the semantic model stores year values, setting its default summarization to "Do not summarize" is appropriate. This decision recognizes that the column is likely used for grouping or filtering, and applying numeric summarization methods, such as an average for years, would lack meaningful significance.
Hierarchy levels
While hierarchy levels are based on columns, they can be used to filter and group but not to summarize. Report authors can summarize the column that the hierarchy level is based on, provided that it's visible in the Fields pane.
Measures
Measures are crafted for summarizing model data and aren't intended for grouping purposes. Nevertheless, there is a unique scenario where measures can serve as filters. Specifically, measures can be utilized to filter a visual, but only when the visual exhibits the measure, and the filter is confined to the visual level, excluding report or page-level filters. In this context, the measure filter comes into play after the analytic query has summarized the data. Its role is to eliminate groups where the measure filter condition doesn't hold true. Drawing an analogy for those familiar with SQL syntax, using a measure to filter a visual resembles the HAVING clause in a SELECT statement.
The stacked column chart illustration has been modified to exhibit monthly groups where sales surpass $5 million. To make this adjustment, navigate to the Filters pane and apply a filter to the Sales Amount field. Set the filter to display items when the value exceeds 5,000,000. Don't forget to click on Apply filter, situated in the lower-right corner of the card.
SWETA SARANGI
25.02.2024