Ad Hoc Analysis Release Highlights: New Chart Views & Statistics
The recent 10/17 Adobe Analytics 1.4 release brings enhancements to the chart options and data set in Ad Hoc Analysis (formerly Discover). The areas of chart improvement include display and usability improvements. The new Statistics functions cover several built-in options that overlay against charts and the data set totals. You can now discern data trends and anomalies through efficient visualizations and better identify specific data points.
1. New Chart Views & Usability Improvements
The most visible default change to the chart views is the improved readability and scaling ability of the chart. The chart auto-scales if the left or lower panes are resized in any way. A wholly new feature is also the ability to drag and move the visual area with your mouse. You can zoom in and out by holding down the control button and using your mouse scroll wheel. Charts now interactively highlight a single figure in the data set below when hovering over points within a graph. The chart enhancements fit in perfect with existing menu customization and workflow options in Ad Hoc Analysis. It is important to note chart views can be distorted or obscured if the scale is dramatically altered in the window. You need to be aware of this potential hazard when altering chart views via mouse action.
Ad Hoc Analysis 1.1: Improved Chart View Display
http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2013/10/AdHocAnalysisRelease_1.1.png
If you ever want to reset back to the default chart you can select Chart Options>Reset View display window button right next the ‘Graph Type’ drop down selector and it will bring the chart back exactly to the original full frame view.
Ad Hoc Analysis 1.2: Chart Option Menu to Adjust Graph Type, Reset View, and Toggle between Value & Percent
http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2013/10/AdHocAnalysisRelease_1.2.png
2. Statistics Options
The Statistics option in Ad Hoc Analysis can be enabled in both the chart view area and the raw data set display windows. As illustrated in the 2.1 diagram below toggling the Statistics menu screen shows the options to apply statistical measures against a specific column. Within the chart, the six available statistical functions are shown as bounded ranges and visualization guides. Important to note there is a user checkbox option to include or exclude zero values within the statistical calculations.
Min- The lowest observed value within the given reporting time frame.
Max- The highest observed value within the given reporting time frame.
Median- Middle value separating the greater and lesser halves of a data set.
Mean- Sum of values of a data set divided by number of values.
Interquartile Range- The data spread distance between the 1st quartile (25th percentile) and 3rd quartile (75th percentile). Q3 – Q1 = IQR
Standard Deviation (1, 2, or 3)– The data spread calculated to indicate the extent of deviation for a group as a whole within 3 quartile bands: 1σ (~68% of the observations fall within 1 standard deviation of the mean), 2σ (~95% of the observations fall within 2 standard deviations of the mean), 3σ (~99.7% of the observations fall within 3 standard deviations of the mean).
Ad Hoc Analysis 2.1: Chart Statistics Guides Applied: Min, Max, Median, Mean
http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2013/10/AdHocAnalysisRelease_2.1.png
Ad Hoc Analysis 2.2: Chart Statistics Guides Applied: Standard Deviation (2 Quartiles)
http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2013/10/AdHocAnalysisRelease_2.2.png
The applied Statistics for the data set in total are also provided in summary at the top of the raw data set as show in diagram 3.1 below. If you right click within the data set screen there is a direct Statistics options window that opens with the same six Statistics calculation options that can be selected within the chart view.
Ad Hoc Analysis 3.1: Statistics Total View in Data Set
http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2013/10/AdHocAnalysisRelease_3.1.png
Conclusion
As a data analyst I value how the latest Ad Hoc Analysis release allows for even faster exploration and direct application of visualized statistical checks. The latest release empowers data reviews to process trends and issues. The statistical visualization makes decisions against the data easier for you to note and move on. All in it has been a banner year for Ad Hoc Analysis with the introduction of sequential segmentation, interface refreshes, and the latest visual enhancements. The chart views and statistic options provide immediate benefits to you in the main reports within Ad Hoc Analysis.
Brian Au is a consultant in Adobe Consulting, focused on digital strategy, analytics & optimization for retail & travel clients. He tweets at _@_BrianAu.