Microsoft-Power-BI-Dataflows

In order for an organization to achieve a data-driven culture, employees throughout the organization need access to information that enables them to make data-driven decisions, in any given situation.

With a growing interest for data among business users, so-called self-service BI-technologies have evolved quickly. Tools such as Power BI, Qlik Sense and Tableau allow users without extensive IT-knowledge to collect data directly from various data sources such as their ERP-system, transform it to the desired format and display it in reports with powerful visualizations.

“As more and more business users in an organization start creating reports, all pulling data from the same sources, some issues with the self-service BI-approach will become apparent.

However, as more and more business users in an organization start creating reports, all pulling data from the same sources, some issues with the self-service BI-approach will become apparent. Extracting data can be a too complex task for a business user, and numerous data extractions from e.g. the business ERP system may affect its performance. Not to mention, report creators will waste time doing the same data transformations as their colleagues as they must be performed once for every report. Then, what happens if one employee does a transformation slightly different than their colleague? Reports will show inconsistent numbers.

Complement to data warehouses

In a traditional BI-system, these issues are solved using a data warehouse, which stores data in the correct format and can be used as the organization’s single source of truth. Business users can connect reporting tools such as Power BI to the data warehouse and avoid slowing down the ERP system or having to deal with inconsistent reports. Setting up a data warehouse can however be connected to large perceived and actual costs, especially for smaller businesses. In an attempt to close this gap, Microsoft announced their new self-service data prep tool for Power BI, called Power BI Dataflows, in July 2018.

What is Power BI Dataflows?

Power BI Dataflows is a new feature in the online Power BI service, that can be used to extract and combine data from various data sources, transform it using the no-code Power Query interface and then store it in the cloud. The Dataflows can then be used as input to Power BI reports. As of now, the feature is available in Preview, and only for users with Power BI Pro or Premium licenses.

What are the benefits?

Using Power BI Dataflows, issues such as the ones discussed above can be avoided. A Dataflow can be re-used in multiple reports and work as a single source of truth. Since Dataflows are designed for business users and not IT-professionals, it is also supposed to make ETL-processes and reliable data accessible to more people within an organization. With built-in connections for a wide variety of common data sources, the feature can simplify data extraction.

Are there disadvantages?

In the current version, functionality is limited and only the creator of a Dataflow can edit it, making it person dependent. KPIs and measures can not be stored in a Dataflow, they still have to be recreated in each Power BI report. This means that the risk of having inconsistent reports within the organization is not completely mitigated. The output from Dataflows are stored in an Azure Datalake and can be used by other more advanced Azure tools, but not by any other BI-tools that the organization might already have.

microsoft powerbi data
 

The use cases for Microsoft Power BI Dataflows depend on the organization and its BI-goals

However, it is not a replacement for a complete data warehouse. For larger organizations, Dataflows should be seen as a complement to more advanced and governed BI-solutions that includes data warehouses. A common scenario is that business users have to request access to data stored in data warehouses from IT-professionals. Data preparation is a time-consuming activity and making it more accessible to business users can increase their power over the data and give them faster access to new data sources, which can help bridge the gap between them and BI-professionals in an organization. Another recommended use case for Dataflows is in a separate sandbox environment where data and analytics ideas can be tried out before implemented in the governed BI-solution.

Is Power BI Dataflows suitable for my organization?

Technology is an important part of a BI-strategy, but before making the decision about how to execute the strategy and which BI-tools to use, the why and who questions need to be addressed. What purpose does the BI-solution serve in your organization? Who are its main users? Depending on the answers to those questions, Power BI Dataflows may or may not be the right choice of technology

At Capacent, we support companies on their journeys to become more data-driven by helping our clients answer the above questions about why, who and how a BI-solution should be implemented.

Author: Josefin Sundström is an analyst working for Capacent Sweden. 

If you are wrestling with translating data into insights or simply do not have the resources to do that, do not hesitate to contact Mikko Eriksson, Head of Business Analytics for Capacent Finland, for more information on how we can help you to create sustainable value in your business. He has over ten years of experience in Business Intelligence solutions in various industries and functions, e.g. within IT, Finance and Commercial.