Skip to Content
Technical Articles
Author's profile photo Giri Raaj Ragupathi

SAP Analytics Cloud Performance Analysis Techniques

  • Overview :

There will be requirements to improve the Dashboard Performance in SAP Analytics Cloud Implementation Projects. From the design to the implementation of dashboards, this must be carefully examined. In the real world, even if we have the nicest car, if there is a problem with the engine, it must be addressed; otherwise, we will not get the most out of it. This applies to SAP Analytics Cloud Applications as well.

SAP Analytics Cloud is a Great application for meeting a variety of company demands. We will have performance concerns if the application is not created using the best standards. Customers may not be able to use the Dashboard effectively if it takes too long to respond.

We are in a position to answer a few queries while keeping our customers in mind.

    • What is my current Dashboard response time? Is it in the acceptable range?
    • What is the Benchmark response time for similar Dashboards?
    • How can we identify the Response time & related metrics?
    • Where exactly do we need to Focus on Tuning? (Backend, Front end, System Level)


  • Available Tools & Options:

The Performance Analysis can be performed using various methods, I would like to present here, the most used options. Please make a choice according to the varying Customer Needs.

    • Google Chrome Developer Tool
    • Performance Visualizer tool
    • Viz for Analyzing Back-end Performance
    • Perform Analysis Tool (Built Using Analytic Application)

Though the above-listed first 3 Tools will help us to identify the performance, there are some manual steps involved in identifying the bottlenecks. Hence the very effective and easy-to-use tool is the ‘Performance Analysis Tool’ (Green marked above).  We will explore the details of the tool and its benefits in this Blog.

The Performance Analysis Tool (PAT) comes with inbuilt installation from SAP Analytics Cloud.  This is in the above-shown path. This is developed based on an Analytical application.  (*If you are unable to see these contents in your SAP Analytics Cloud Tenant, please check with the admin team for necessary authorization).


  • Step by Step Approach:


  • Step 1: Click the Performance Analysis Tool


  • Step 2: Run Analytic Application

  • Step 3: Access Landing Page – PAT Tool


  • Step 4: Launch the Dashboard (in another TAB) which you intend to Analyze

For Demonstration purposes I have used, one of the standard contents shown below, which has many widgets and is spread across 5 pages.


Access Page 1 – Overview Page, carefully scroll down till all the widgets are properly displayed.  After loading, please switch to page 2 – The context page in this case and follow the same exercise for all the pages in your dashboard.


If your dashboard contains the maps and tables, please allow some time to capture all the details. Then scroll down. (Ref. Page 3)



Once all the 5 pages are navigated in the same way, please switch back to the PAT tab and click the ‘search’ button. This will result in the list of stories in session. As you can see from the below screen, we have accessed ‘Demo_Test_Performance_Story’.


The output of the entire metrics is populated in the below screen.  As you can see, It consists of Page wise response time. It contains key information like the Top 5 widgets which we can focus on tuning.  What is the model-wise back-end time, and network time?

I think the output answers most of our questions, which we raised at the start of this analysis.

    • What is my current Dashboard response time? Is it in the acceptable range?

Yes. My Current Dashboard time is:

      • Page 1 – 4.24 seconds.
      • Page 2 – 7.13 seconds.
      • Page 3 – 16.66 seconds.
      • Page 4 – 15.95 seconds.
      • Page 5 – 8.87 seconds.

Is it an acceptable range? we will discuss this in detail blog. But from a user experience perspective, this is fine. (But there is always a scope for improvement!).

The report contains a detailed widget-wise analysis shown above. This is key information for us to analyze widgets and their associated models. As you can see from the analysis, the Vendor management chart and Worker type chart is more than 7 seconds to load.

Please also have a look at the below Run time analysis. This contains the Dashboard back end and front end and network time consumption.


  • Conclusion:

This tool will be very useful for the performance assessment of SAP Analytics Cloud Dashboard. For any SAP Analytics Cloud Implementation projects, these are the common issues we may face. Based on the above analysis we can focus on the right areas for tuning.

When we say performance tuning – it’s not always application alone, various factors come into play.

    • No. of widgets on a page. (Recommended 8 widgets max)
    • Try to minimize the no. of widgets on the initial page.
    • Push all the front-end calculations to the back-end model.
    • If possible, re-visit the Design to accommodate the Performance issues.
    • Enable the Performance Optimized Mode.

We will dive deep into the details of the Tuning aspects in the Next Blog. Hope this high-level insight helps us in Project Implementations. We will connect soon…Thank you for your time!

–  Giri Raaj. PMP –

Assigned Tags

      You must be Logged on to comment or reply to a post.
      Author's profile photo Divya Lakshmi
      Divya Lakshmi

      Good article, thank you for sharing!

      Author's profile photo Rajesh Roxx
      Rajesh Roxx

      Great Article...Nice...Insights on techniques. Thank you Giri..

      Author's profile photo Venkata Rama Dharmaraj
      Venkata Rama Dharmaraj

      Nice article Giri. Thank you.

      Author's profile photo Mary Infanta
      Mary Infanta

      Performance analysis was a struggle-some. PAT tool is very convenient. Thanks for the step by step details to understand the tool. 👍

      Author's profile photo Alexander Blasl
      Alexander Blasl

      Giri Raaj Ragupathi

      I have a question about "Push all the front-end calculations to the back-end model."

      Is this really true? I added 20 Calculated Measures (also Ex. Aggregations) in SAC Frontend and also in BW Query in Backend (BW on Hana with Ex. Aggregation in SAP Hana ☑). It seems that when I have them in SAC Frontend the Story loads a bit faster. Maybe because my notebook is a powerful workstation? Can you maybe explain your statement a bit more in detail?