Lire cet article en Français 
Introducing Report Validation and Certification for Power BI
Build Trust in your Power BI Content with Validation and Certification Capabilities
Imagine you’re about to enter a meeting. You take a final look at your Power BI report and think, “Is my data correct? Usually, my margin is 10%.” Or imagine yourself in a meeting where one colleague presents a KPI, while another colleague presents a different, almost contradictory figure, and you wonder: ‘But what is the correct number?’ Such issues have always existed in BI and are still very common, despite the major efforts made by organizations in data governance. Without Analytics Governance, they can result in significant risks and costs to organizations, including a loss of trust and eventually the failure of any data-driven initiatives.
At Wiiisdom, we want to help change the way people consume data analytics. Just like when you look for a new restaurant, you’re looking for the most recent reviews. This should work the same way with a Power BI report, you want to be sure it’s accurate, recent, and ready to use so you know you can trust the decisions made from it. We provide Analytics Governance solutions to make data analytics reliable, including the ability to automatically and dynamically test, validate, and certify your Power BI reports and their underlying semantic models, to reassure the end user that their content can be trusted. This is a very unique capability for which Wiiisdom has applied for a patent.
Let’s see how easy it is to validate and certify your Power BI semantic models and reports using Wiiisdom for Power BI.
Power BI Report Validation & Certification with Wiiisdom
I’m going to take you through 3 scenarios of our Analytics Governance solution for Power BI.
Automated validation of semantic model and proactive notifications
First, you need to ensure the quality of your semantic model. If it isn’t working properly or doesn’t have the correct data, your reports built on top of it will reflect the same issues. The importance here is to check the data accuracy and enforcement of business rules as well as technical checks in the semantic model.
To validate your semantic model with Wiiisdom for Power BI, it will be a 3-step pipeline (start + intermediate steps for validation + end):
Example of a pipeline view for semantic model validation.
First, you’ll want to check your Semantic Model Refresh. Is your semantic model refreshed or is it refreshing correctly? This is to ensure it was refreshed in the last X minutes, hours, or days. This is important to start with because if the semantic model is not refreshed, all the subsequent steps will not make sense.
Ensuring that your semantic model refreshed in a chosen period.
Secondly, Row Count. Do your tables contain all your data or is there too much data in your tables? Or, did your data processing fail or have issues? With how Cloud storage works, too much data can become very costly for organizations.
Executing the Row Count test in Wiiisdom for Power BI.
Thirdly, the business rules and data validation are handled through the Semantic Model Values step. For example, you may have different business rules depending on a country so you want to make sure it doesn’t include certain data. Additionally, do you see the expected values in your semantic model?
Implementing Semantic Model Values test in Wiiisdom for Power BI.
Then, you’ll want to test the Column Statistics. Do you often get end users saying “Why do I have a blank option in my visual or slicer?” Executing this test allows you to check statistics such as empty, minimum or maximum values to ensure proper data quality and performance.
Setting up a test for Column Statistics in Wiiisdom for Power BI.
The final step (End) allows you to become proactive by activating notifications via email or instant messaging apps (e.g. Slack, Teams, Google chat) to notify your BI team if a test has passed or failed. Even better, you can integrate it with your ticketing system to create a support ticket as soon as a pipeline fails. This allows you to troubleshoot the issue even before the end user has noticed it! Equally, you can schedule this pipeline to run when you would like and have access to the history of executed pipelines for audit trails.
You can schedule the Pipeline Execution for your Semantic Model whenever you want.
Automated validation of reports and proactive notifications
Now you have tested and validated your Power BI semantic model, you will want to do the same for your reports.
Have you encountered issues with access rights or improperly granted security access, or perhaps faced capacity overloads that require patience? Have you had a colleague modify a semantic model without warning, resulting in broken reports? Become proactive by regularly monitoring your reports to ensure a smooth user experience. For example, check if the report opens within the expected time or simulate a user journey to see if the correct data appears when a user changes a country filter. With our easy drag-and-drop capabilities, you can set up your test pipeline in just a few clicks.
Setting up tests for your Power BI reports in Wiiisdom for Power BI.
Just as you can set up notifications when testing and validating your semantic model, the same can be done for your Power BI report validation. All the execution pipeline history is easily accessible, and every pipeline can be scheduled as and when you like.
Complete validation with dynamic certification
Finally, change the paradigm for your end users and make your validation efforts visible through automated certification. Automated and dynamic certification acts as a visual label proving that a report has been thoroughly tested and validated. A report can be either certified showing the end user they can use it to make decisions, or be decertified warning the end user of potential risks if they do use it. Move from “Can I trust my data analytics?” to “Is it certified so that I can move forward with my decision-making?”
Example of a certified and decertified Power BI report.
It’s really as simple as that!
Validating and certifying your Power BI semantic models and reports doesn’t have to cause a headache. With our no-code, untechnical Cloud solution, in just a few clicks, you can be sure that your Power BI content is accurate and trusted around the clock. If you’re looking for a solution to have validated and certified Power BI content, get in touch with us today to see the product in action.
