Achieving a Self-Service BI using Metadata: The Metadata Democracy
Self-service, data democracy and tools such as SAC or Datasphere have long been an integral part of the business intelligence bubble. The demand for a self-service BI is high – and has been for a long time: As early as 2016, 80% of the companies surveyed in a BARC study stated that they were already using or planning to use self-service BI tools. We have since seen a strong business orientation of data & analytics tools and a change in the role of business users, who are now expected to shoulder much more responsibility for creating reports and dashboards independently – without the support of IT.
This paradigm shift is certainly a positive one – after all, efficient processes and data-driven decisions are crucial. However, especially with regard to SAP BI tools, the theory of self-service BI is still lagging behind practice. Observations often show that business users are still far from empowered and continue to rely on information and assistance from IT.
This article discusses which obstacles are responsible for this, why they arise and how the path to a genuine data democracy can be achieved. Let us tell you this much in advance: Metadata is the key.
Status Quo: SAP & Self-Service BI
In the context of SAP Data & Analytics, the concept of self-service BI can be broken down into two components: Modeling capabilities and (meta)data transparency.
SAP Analytics Cloud (SAC) and Datasphere enable business users to carry out modeling independently – thanks to an intuitive user interface and low-code approaches. IT skills are no longer absolutely necessary to create interactive dashboards, visualize data and perform ad-hoc analyses. At the same time, the Space concept offers the perfect opportunity to create isolated content areas under the sole responsibility of the business departments.
With regard to data transparency, SAP already offers a number of tools in these technologies that support BI self-service – especially in these areas:
- The graphical representation of data models
- The integrated Data Catalog
- The strategic cooperation with Collibra
Of course, it should also be mentioned at this point that it remains to be seen how the announcements regarding the Business Data Cloud will affect these two self-service components in the future.
So much for the theory – in practice, the above-mentioned data transparency reaches its limits when a very common situation arises: In most companies, in addition to modern tools, older SAP solutions are also part of the architecture. Specifically, we are talking about the continued use of one or more SAP BW systems in almost all large companies. In most cases, these have grown over years and decades and are highly complex. Perhaps HANA Calculation Views or CDS Views that are consumed by SAC stories have also been tried out in the past. And then there may be countless AfO reports in the front end that the business department cannot part with. And, not just to talk about SAP: Power BI, of course, is also popular and widely used.
We can strongly assume that this hybrid approach of old and new technologies will continue to be used by most companies for quite some time still – regardless of the newly announced possibility of lifting the BW system into the BDC as a private cloud edition.
Limits of a Self-Service BI
So much for the framework conditions. Now, if a business user is “let loose” on this architecture and is supposed to actually carry out the BI self-service idea, he is faced with lots of layers below the front-end tools SAC, BO and Co. Without in-depth IT knowledge and access to them, the SAP architecture resembles a Black Box for him.
One can already guess that the promising approach of self-service BI, according to which specialist departments are able to search data and create reports without in-depth technical knowledge, is being undermined by the framework conditions of the non-transparent reality.
In discussions with our customers, we were able to work out what information business users are looking for and thus identify the following key questions that continue to be asked of IT, given the conditions:
- Which reports or dashboards are available for my topic? Who is responsible?
- How are these reports structured technically? Which filters and data sources are used?
- How is my key figure calculated? What is the formula behind it?
- What has changed technically in my report since yesterday?
The sobering realization is that the use of SAP tools with self-service BI functionalities alone is not enough to establish actual self-service BI in the company. The tools do not provide business users with enough information on everyday issues and continue to tie them to IT. Alternatively, business users might not ask at all—instead of searching for existing reports, they simply create new ones, contributing to a growing graveyard of unused dashboards, views, and tables.
You can guess what’s coming next since we have already addressed the solution to this problem in the introduction:
Metadata as the Key to a Functioning Self-Service BI
Now, this is where the main character enters the stage – the metadata, of course.
Why are they so important? Because they contain information about the data & data sources, about relationships and provide valuable contextual information. In short, metadata makes data understandable and findable. Without it, the key to using data effectively is missing.
Unfortunately, this metadata is not served to business users (and developers) on a platter – it has to be found with considerable effort, prepared and then made accessible. This is precisely the challenge we at bluetelligence have been tackling for over 15 years of software development: We search the backend tables of SAP applications, read the metadata and prepare it in a way that it is understandable for everyone.
The question that arises now is of course: How can this valuable information be made available to business users? And this is where data catalogs come into play.
A company-wide data catalog serves as a central knowledge base for metadata and data. It enables companies to provide information in a structured way and make it accessible to a wide range of users. This makes it the basis for genuine data democratization and supports sustainable data management.
In the following section, we look at how a data catalog can help with the challenges we have outlined above:
Basic Requirements of a Data Catalog:
- Simple and centralized access for all business users, e.g. via browser
- User-friendly in terms of UX and UI
- Automatic updating of technical metadata – best case, scheduled via jobs
- Support of common BI solutions (SAP, PowerBI, etc.)
- Easy searchability, ideally with filter options
- Can be called from SAP applications (e.g. in SAC Story)
Find suitable reports/dashboards:
- based on technical name
- based on a source object (e.g. BW Query, Analytic Model…), essentially “Where-Used”
Display the context of reports/dashboards:
- Display the technical structure of all reports/dashboards in an understandable way – e.g. check filters at a glance
- Offer the possibility of documentation beyond technical details – e.g. technical additions or responsibilities – this improves the understanding of data and the possible uses for different user groups
Understand Key Figures
- Display formula of calculated key figures
- Graphical view of the relationships (so-called driver tree)
Data Lineage:
- Identify data sources quickly with the help of graphical representations
A tour through the possibilities of our Data Catalog “Enterprise Glossary”
You are currently viewing a placeholder content from YouTube. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More InformationKey Take Aways
A data catalog that contains an overview of the relevant metadata (data sources, where-used, key figure correlations) and presents it in a way that is understandable for business users is the key to an actual self-service BI.
This empowers and relieves the right people:
- Business users who know about context are capable of taking actions and making well-founded decisions – after all, business users will be expected to have precisely this skill set in the future.
- And IT? Ideally, the elimination of numerous tickets will give them more time for strategic development, innovative data solutions and a sustainable BI architecture.
If you would like to try out our Data Catalog, the “Enterprise Glossary” and the insights it contains on metadata from SAP BW, BW/4, S/4, SAC, HANA and Datasphere as well as Microsoft Power BI for yourself, you can do so easily and free of charge by sending a short demo request via our website.