Making Data-Driven a Reality
Against the backdrop of "data-driven", a debate is unfolding that is rich in buzzwords: from Artificial Intelligence to Natural Language Processing to Robotic Process Automation. All of this quickly seems like hype. However, in this case, the hype actually has substance. Companies can derive enormous benefit from data in a variety of ways.
Competencies and technologies
The fact that there is immense potential, but that it has not yet been fully exploited, suggests in our view that companies need to invest quickly in two dimensions in order to make the most of their data and thus gain a competitive edge: in expertise and technology.
Competence on the one hand is necessary to analyze data in a target-oriented manner. First of all, this requires a deep understanding of mathematics and a broad understanding of business. Employees must therefore be able, for example, to form hypotheses about customer satisfaction and verify or falsify them with the help of the data. Or they must be able to derive conclusive interpretations for the business from the patterns identified in the data.
Technology, on the other hand, is indispensable because data can only be processed with dedicated solutions. This applies not only to the pure analysis of the data, but also to the steps required beforehand - collection, harmonization and processing. Technology thus bridges the gap to competence when, for example, augmented analytics in the SAP Analytics Cloud provides data preparation that enables employees to make fast, data-based decisions through machine learning scenarios without having to process and interpret the data themselves in a tedious and subjective manner.
Data Democratization with the Analytics Cloud
SAP has significantly updated its data-oriented product portfolio in recent years and now offers innovative solutions for all disciplines. Central to this is SAP Hana, for one. The in-memory database, which combines OLTP and OLAP, is now a component of SAP applications.
On the other hand, the SAP Analytics Cloud (SAC) occupies a dominant position, which as software as a service is also based on the Hana in-memory database. With SAC, the data to be analyzed can be accessed in two ways: With Import Data Connection, they are loaded from a source system into the cloud and analyzed there. With Live Data Connection, there is no replication of the data in the cloud. Instead, the SAC works on the source system.
The SAC functionally covers the areas of reporting, analysis, planning and predictive analytics. In addition, there are the possibilities of the Application Design component, with which dashboards can be created. The individual areas can be integrated excellently, which avoids breaks in the workflow that were the rule in the past when using different stand-alone solutions.
Overall, the SAP Analytics Cloud is characterized by a differentiated user concept. This means that analyses can be performed not only by experts from the IT department or Controlling - as has almost always been the case to date. Self-services also allow employees from the specialist departments and management (citizen data scientists) to evaluate and visualize data independently.
Only in the course of this data democratization will data- and insight-based decisions be firmly anchored in everyday business. This is an important step on the way to the Intelligent Enterprise - because it means that data science can be applied out of the box in the business departments, even without in-depth data analysis expertise.
SAP Analytics Cloud
The BI Survey 19 by analyst and market researcher Barc suggests that the SAP Analytics Cloud is very well suited. In it, SAC was compared with a number of other tools in 34 criteria - on the basis of a user survey - and came out on top.
For example, 97 percent of participants would recommend SAC to others. 91 percent of them rate the dashboard creation capabilities as excellent or good. And 88 percent consider SAP's ability to understand the needs of the business to be very good or good.
Based on our experience from customer projects, we can absolutely understand this excellent performance of the SAP Analytics Cloud. However, the customer projects have also shown us that a few points are critical to success during implementation. For this reason, Nagarro ES has developed a procedure based on SAP Activate, which ensures that all relevant aspects are taken into account (see box).
SAP Data Intelligence
SAP positions the Analytics Cloud as a central solution. Nevertheless, it is also worth taking a look beyond the SAP Analytics Cloud in order to meet the requirements that data scientists place on a technology. As a rule, the potential of SAC is not sufficient for their demanding projects. SAP Data Intelligence is better suited for them.
The solution runs on the SAP Cloud Platform (SCP), connects the somewhat more conservative business world with the open source world and is open for corresponding applications: for example, for Jupyter Notebooks from Project Jupyter, for Python and for Python-based machine learning frameworks such as pandas, scikit-learn or TensorFlow.
The data models created in this way can be transferred to the SAP environment with SAP Data Intelligence, where they can be further processed, automatically tested, and finally delivered to a highly available and scalable productive landscape - including subsequent performance monitoring. And, very important in a business context: AI scenarios can be audited so that they meet the requirements of an audit, for example.
Introduction of the SAC in four phases
Discover
Fitting Workshop: Discussion of fundamental questions; recording of the current situation; formulation of the target situation; definition of the SAC architecture Jam (reporting/planning): Exchange on business processes and requirements; alignment of knowledge and experience from successful SAC projects.
Prepare
Develop an initial SAC prototype as a proof of concept (PoC).
Explore
Design of applications; establishment of a live connection to the test system; basic alignment of individual aspects of the SAC - such as clarification of security issues and tailoring of user access.
Realize
Set up SAP Analytics Cloud; implement applications and permissions; validate data, set up live connection to production system.