Application Developer

Tell the Story

Use Oracle Data Visualization Desktop to present information and insights.

By Mark Rittman Oracle ACE Director

January/February 2017

If, like me, you still remember some of the stories you read when you were a child, it’s probably the twists and turns, the cliff-hangers, the unanswered questions, and the final resolution that make those stories so memorable. As adults, we still enjoy stories, and good ones resonate both intellectually and emotionally with an audience. Stories aren’t limited to fiction, of course, and using Oracle Data Visualization Desktop and some data of interest, you can visualize that data, uncover insights, and then use the tool’s storytelling features to make that data more meaningful when you present it to others.

In this article, I’ll use Oracle Data Visualization Desktop and its visualization and storytelling features to explore, understand, and then tell the story about my attempt to lose some weight by exercising more. I recorded my workouts and weight each day, using wearable devices and a smartphone app. You can download and unzip the sample datafiles used in this article and download Oracle Data Visualization Desktop if you want to try the article example yourself. Or you can use your own fitness tracker or workout app data to build your own story by exporting the data to Microsoft Excel XLSX files and using Oracle Data Visualization Desktop.

Uploading, Analyzing, and Discovering

Let’s start by taking a spreadsheet datafile exported from a cycling workout smartphone app. It contains a record of every cycling workout I recorded over a nine-month period starting in January 2016. I’ll upload the file into Oracle Data Visualization Desktop; review it; and, if necessary, change how columns within it are designated as either measures or attributes and configure the outside temperature measure to aggregate by averaging its values (rather than adding them all together).

  1. Start by double-clicking the Oracle Data Visualization Desktop 12c icon on the Windows desktop to open the application, and then click the Navigator menu button at the top left of the application window to display a menu of options. Click the Data Sources menu item to open the Data Sources page, and then click Create New Data Source.
  2. Because the data I am going to upload is contained in a spreadsheet file, click the From a File option in the Create New Data Source dialog box and then use the Windows Explorer file picker to locate and select the activities.xlsx file you downloaded and unzipped for this article, clicking Open to move on to the next step.
  3. In the Upload a File dialog box, use the scroll bar to review the columns of data retrieved from the spreadsheet file. Oracle Data Visualization Desktop automatically categorizes any column containing alphanumeric data as containing attributes and columns with only numeric values as measures. These defaults are correct for this data set except for two columns. First, the Activity ID column contains numeric values but is, in fact, an attribute column used to join activities to other data the smartphone app records. To correct this first column setting, click the Measure setting below the Activity ID column label and change that setting to Attribute, as shown in Figure 1. To correct the second column, change the aggregation type for the Temp C column from Sum to Avg.


    Figure 1: Changing the column type of an imported data set column
  4. Click Add Data Source to add this file and its column settings to Oracle Data Visualization Desktop’s set of datasources available for projects.
Analyzing the Data, Recording the Insights

Now I can create my first data visualization that will show how the distance logged for the workouts I recorded with the smartphone app varied over the nine months of data. Because the workout app also recorded the outside temperature when the workout took place, I’ll then overlay that information onto the visualization to see whether how hot or cold it was outside affected the distance logged for a workout.

  1. Click the Data Elements menu icon on the left-hand side of the page, and then click the activities datasource to display the list of columns that can be added to the canvas.
  2. To display a line chart of the distance cycled over time, first double-click the Dist km column to add it to the canvas on the right and then double-click the When column. Oracle Data Visualization Desktop will automatically recognize the latter column as time-related and place it on the x-axis with “Month” axis markers while placing the other column on the y-axis.
  3. Next, double-click the Temp C column on the left-hand side of the page to also add it to the data visualization. In this case, the column is intelligently overlaid as colored dots over the main chart data so that each month’s distance total now has an indication of how hot or cold that month was on average. To change the default color to shades of red, better indicating the heat the app recorded for that month, select Color -> Series (Temp C) and then click one of the red colors so that your final data visualization looks similar to Figure 2.


    Figure 2: Data visualization featuring distance, frequency, and temperature
  4. Looking at the data visualization I’ve created, it does seem that more workout activity happened when the weather was warmer, which was, in fact, the case. Let’s save that as an “insight,” a snapshot of a particular data visualization I can combine with other insights at the end to tell the story behind the data I’m analyzing. To do that, click the insights (lightbulb) icon in the left-hand menu panel, click Add Insight, and name the insight Workout frequency increases as weather improves.
  5. Although the frequency of the workouts seems to have increased, they may well have been for shorter distances, perhaps because it was too hot to cycle or I just exercised less as the year went on. To see this overall trend overlaid on the chart as a trend line calculated with linear regression, click the menu button at the top right side of the chart and select Add Trend Line. As I suspected and as shown in Figure 3, although some workouts were for very long distances, the overall trend in individual workout distances logged was, in fact, down in the course of the year.


    Figure 3: Visualization featuring distance, frequency, and temperature plus distance trend
  6. Click the insights icon in the left-hand menu panel, click Add Insight to add this new insight to your project, and name it However, the overall trend was down.

In reality, even though the trend in the workout distance may have fallen, the overall distance logged per month actually rose, due to these additional but shorter workouts. To show this, I first need to create a new calculated column for the month each workout was logged and then use that new column to show the distance traveled per month.

  1. Navigate to the bottom left of the page, and click Add Calculation.
  2. In the New Calculation dialog box, name the calculation Workout Month, select Calendar/Date -> Month from the list of functions to the right, and double-click it to add that function template to the left-hand side of the dialog box. Click the dimension expression, select Activities -> When from the list of columns, click Validate, and click Save to create the new calculation.
  3. Now drag and drop that new Workout Month calculation from the My Calculations folder to replace the When column within the Category (X-Axis) visualization setting and change the visualization type to Bar so that it looks like Figure 4. As you can now see, looking at the distance logged overall each month, the trend is clearly up, with the warmest months recording the greatest overall workout distances. Click the insights icon, click Add Insight to save the current visualization as another insight for the project, and name the insight Overall though, Total workout distance rose.


    Figure 4: Bar visualization showing the monthly distance trend
Adding a Second Data Set to Complete the Picture

So it looks as if my overall workout distance has increased over time, but has this extra exercise resulted in weight loss being recorded in my other smartphone health app, or was all of that working out a wasted effort on my part? To find out, I’ll now add the second datasource containing weight readings over the same period and see whether this increased exercise activity has helped me shed those extra holiday pounds.

  1. To add this new datasource, click the Data Sources icon in the left-hand menu panel, right-click anywhere in the empty space under the existing activities datasource, and select Add Data Source from the menu.
  2. In the Add Data Source dialog box, click Create New Data Source and then click From a File to display a file picker so you can select and upload the second datafile, Weight Readings.xlsx, that you downloaded and unzipped earlier.
  3. When the Upload a File dialog box appears, change the Aggregation setting for the Weight (kg) measure from Sum to Average under that column’s settings, much as how you changed some of the column types after importing the first file (activities.xlsx) earlier in this example.
  4. Now, to join this new datasource to the existing activities datasource, right-click anywhere underneath the two datasources and select Source Diagram. Then in the Connect Sources dialog box, click Add Another Match, select Date from the Weight Readings datasource and When from the activities datasource, click OK to save the connection, and click the close (x) button to dismiss the source diagram from the main application window.
  5. Click the data elements (book) icon in the left-hand menu panel to show this second datasource listed under the one I’ve been using so far. Double-click it to show the three measures that are now available to add to the data visualization, as shown in Figure 5.

    Figure 5: Three Weight Readings measures available for data visualization
  6. Now remove the existing trend line from the visualization by clicking the menu icon at the top right of the visualization canvas, selecting Properties -> Analytics, and clicking the X icon to delete it from this visualization.
  7. Now click Add Calculation on the left-hand side of the page and, in the New Calculation dialog box, name the calculation Weight Month. Select Calendar/Date -> Month from the list of functions to the right, double-click it to add that function template to the left-hand side of the dialog box, and then click the dimension expression to select Weight Readings -> Date. Click Validate and then Save.
  8. Click and drag the Weight (kg) column so that it is placed directly under the Dist km column in the Values (Y-Axis) visualization setting, and, from the visualization menu, select Properties -> Axis -> Secondary Axis Values, using 84 and 88 as the Start and End settings. To enable the secondary y-axis for the weight series, click the Values tab in that same dialog box and then select Weight kg -> Y2 Axis -> On. As you can see in Figure 6, the weight at the end of the period is lower than at the start and the additional workout distance logged over the summer seems to have led to a corresponding fall in weight shortly thereafter.


    Figure 6: Data visualization showing workout distance and weight

Save this as another insight by clicking the insights icon on the left-hand menu panel, clicking Add Insight, and naming this insight And Exercise Seems to Have Worked!

Making the Data Meaningful by Presenting It as a Story

This example created and explored visualizations using the data from two datasources. I saved data of interest as insights so I could come back to them later. Insights are snapshots of data visualizations at a point in time and can be refreshed to show updated data if necessary. Insights can also be used to tell a story about the data that can be more effective than just a set of saved reports in a catalog.

To view and optionally edit the sequence of insights that make up the story for this data set, click the Story Navigator button in the application toolbar to show the insights in the order you created them. Use the back and forward buttons to view the story timeline, and edit or remove insights from the story timeline to create the final story presentation, as shown in Figure 7.


Figure 7: Data visualization with story navigator

Finally, if you want to present these insights and the story they tell about the data to friends or colleagues, you can switch the application to presentation mode by clicking the Presentation Mode button in the same application toolbar to remove all the menus and buttons and focus just on the story and the data.


Data, and the meaning behind it, resonates better with audiences and is remembered longer when you use storytelling techniques and effective visualizations to present your insights to others. Oracle Data Visualization Desktop puts these capabilities into the hands of anyone with a recent desktop PC and some data of interest. Try this storytelling approach out yourself, and see what stories you can tell with your data.

Next Steps

DOWNLOAD Oracle Data Visualization Desktop.

DOWNLOAD the sample data for this article.

LEARN more about Oracle Data Visualization.

TRY Oracle Data Visualization Cloud Service.

READ more by Mark Rittman.


Photography by Ricardo Gomez, Unsplash