The last two months we – Norbert Maijoor and I – have been exploring the possibilities of SAP Analytics Cloud (SAC from here further). In this piece we will share with you five practical tips that we find will save you a lot of time and frustration when working with SAC.
Please note: These tips are based on a live data connection with HANA Cloud Platform and with SAC version 1.0.56. Further, we assume that you have followed the OpenSAP course ‘Implementation of SAP Analytics Cloud’. Therefore, we will refrain from a step-by-step explanation. Do you want additional information or help? Just leave a comment and we will try to help you.
1. Be saving
Your best friend in SAC is the save button. Save regularly! At the moment SAC does not have an auto save and undo feature. Meaning that if something goes wrong you will lose every progress you made beyond your last save. Thus, have you been working on a visualization for more than 10 minutes? Or do you want to experiment with your data and visualization? Save your progress.
2. Be blocking
Or more specific use text blocks to your advantage. We like to use them in two ways.
2.1 Aligning chart titles more flexibly
Aligning titles within visualizations is limited. A chart title is shown in the left top corner. See Figure 1.
Figure 1. Standard Chart Title
You can use a text block to position the header more flexibly. For example, above or below the chart as in Figure 2.
Figure 2 Using Text Blocks
Please note: when you use dynamic filters in your visualization you’ll need to show the chart details. These are also shown in the top left corner. Therefore, please make sure you do not put the text block over these details when you need them.
2.2 Hiding parts of visualizations
Some parts of visualizations cannot be hidden using the settings of SAC. For example, certain value labels or parts of a table. You can use text blocks (with or without text) to ‘hide’ these parts.
You’ll need to insert a text block and go to the formatting tab and give the background color of text block the same color as the background of the chart. Then you arrange the text block in front of the chart. This way it will overlap the part of the chart you want it to. Below we provide two examples how we applied this trick.
The bullet chart will automatically show a data label containing both the measures and the dimension values. This is fine since these describe what you’re seeing in the chart. However, we wanted to show a bullet chart based on a measure without dimensions. At the moment this is not possible. So we used a ‘dummy dimension’. The result can be seen in the top of Figure 3.
We had two issues with this result. First, the dimension label (‘1’ in our example) being shown. We did not want this 1 to be displayed. However, at the moment you cannot remove this dimension label. Second, we wanted to use a different label name than the description of the measure for this specific chart but we wanted to keep the description for other charts. Unfortunately, to change the description in the model would mean that it would be changed for all charts.
Figure 3 Example 1
We used two text blocks with background color to easily solve the two issues. The result is shown in the bottom part of Figure 3.
Sometimes you might want so display detailed information in a table. However, at the moment, this is visualized as a spreadsheet. Unfortunately, you cannot format this spreadsheet. See Figure 4.
Figure 4 Table Without Text Blocks
In order to make it more a visualization than a spreadsheet we added text blocks to hide certain parts and to better position the column headers. See Figure 5.
Figure 5 Table With Text Blocks
3. Be calculating
Calculated measures offer you a lot of possibilities. A simple calculated measure that just has the calculation = [‘name of measure’] makes everything better. Below you’ll find the uses for a calculated measure.
3.1 Renaming the measure
Using a connected model, the name of a measure is displayed as it is labeled in the source. This is fine of course but sometimes you might want to shorten or change the measure label that is displayed. At the moment this is not possible with the measures coming from the connected model. This is where calculated measures can help you out. Just enter the name you want under ‘Name’ in the Calculation Editor (Figure 6).
Figure 6 Calculation Editor
In Figure 6 the standard label is ‘Wet- en Regelgeving Percentage’. Which we want to use in some visualizations but in others it is too long. So we abbreviated the label/name to W&R % using a calculated measure which only contains the measure from the model.
3.2 Format the measure
Using a connected model within SAC will result in measures that cannot really be formatted (you’ll have to this in the data source). In the modeler you can tell SAC how a measure should be displayed, but at the moment this isn’t working properly. This will be fixed in a future release.
For now, this entails that measures from your model which should be shown as a percentage are displayed as in Figure 7. Note that the only way to determine that this is a percentage is in the subtitle in the left top corner. The ‘c’ denotes that the value is a percentage. Unfortunately, you cannot do much about this.
Figure 7 Measure as percentage
However, using a calculated measure you can now format your measure properly. See Figure 8.
Figure 8 Formatting measure
We set the format to percentage with one decimal space and the result is far more satisfying:
Figure 9 Formatted measure
3.3 Using more than 1 format.
The format you used for a measure will be applied to all visualizations in the story. This means that if you have formatted a measure to display one decimal, it will do so in all visualizations. Do you want to use the same measure and display one decimal in visualization A and five decimals in visualization B? You cannot do so.
Luckily, you can add another calculated measure. You can use the exact same calculation. Now you can use this new calculation to format the measure with five decimals. Mind you, you’ll have to give it a different name. Continuing the example of Figure 7, I used two spaces in the name: ‘W&R %’. Unfortunately, SAC does not accept spaces before or after the name. The result can be found in Figure 10. The bar chart uses the measure with one decimal and the KPI chart uses the ‘same’ measure with five decimals.
Figure 10 Multiple Calculated Measures
3.4 Using more than 1 threshold.
The same rules that apply to formatting a measure also apply to conditionally formatting a measure. You can define a threshold for a measure to determine what colors should be shown when. This threshold is fixed to the measure. Thus, you cannot have multiple conditional formats for one measure.
Figure 11 Defining a Threshold
Yet sometimes you might want to apply a different conditional format to a measure within the same story. For example, when working with different filters on the same measure. Luckily you can do so with the same method as 3.3. Just add another calculated measure and you can use that to add another threshold.
4. Be coloring
As discussed in tip 2 you cannot hide all text or values of a visualization. Yet sometimes you might want to hide a data label because the meaning of the value is already apparent because of other elements on your page. As an alternative to the text block you can also hide the text by giving that specific test the same color as the background. Important to note is that you’ll have to make sure you’re giving the correct text the color of the background. Fortunately, this is easy to accomplish. First select the visualization and then click on the text you want to hide (in Figure 12 this would be text within the blue box).
Figure 12 Selecting the correct text
In the ‘Designer’ mode under ‘Formatting’ you’ll notice that ‘Chart Properties’ will say ‘Custom’. Now you can edit the color of the selected text and hide it. In addition, we recommend resizing that text as small as possible (size 10 in this version). Remember that the text is still there. So by resizing it you make sure that its impact on the size of the visualizations is limited. See Figure 13 for the options and result.
Figure 13 Formatting text
So when do you use Text blocks and when do you use coloring? It depends on the circumstances and your preferences. We advise the following general rule of thumb. Do you just want to hide the text? Color it. Do you want to use the space of the text (as with the example of the bullet charts in tip 2)? Use text blocks.
5. Be exploring
Currently you might find that SAC cannot do everything what you want it to do. However, SAC offers you a range of other possibilities you might not have thought of before. So we recommend to explore and experiment with all features of SAC. This could give you new perspectives on data and visualizations. And it is fun!
This concludes our 5 practical tips for beginners. What tips and tricks would you share with beginners? And what are your thoughts on our tips? We hope to hear from you!
Clemens Krom & Norbert Maijoor.