The aim of data preparation is to provide users and analytical systems with clean and usable data - as a basis for determining relevant customer needs. Without the appropriate data basis, meaningful analyses are difficult to realize. The methods of "data preparation" can support the preparations.
Data Preparation includes all activities that serve to improve data quality in master data management, its usability, accessibility and transferability. These activities include (but are not limited to):
- Data integration (the merging of information from different databases),
- Data profiling (the largely automated process for analyzing the quality of existing databases),
- Data cleansing (process for removing and correcting data errors in databases or other information systems) and
- Data governance (internal company guidelines for handling data).
The aim of data preparation is to provide users and analytical systems with clean and usable data - as a basis for determining relevant customer needs.
Don't waste precious time, because the amount of data is growing every day
According to the "First half 2018 market survey" by Nucleus Research, spending on business intelligence (BI) tools fell by 80 percent in 2018. The problem is not that there is less interest in analyzing information. Instead, 83% of users want analytics tools directly in the applications they already use, rather than spending precious time switching to BI stand-alone solutions.
Furthermore, reports, dashboards and visualizations remain an important part of the analysis process. However, more and more users are no longer only concerned with these "front-end" results of the analysis, but increasingly with the "back-end" activities that are fundamental to the analysis. This is where data preparation comes into play (see info box). In 2017, the Würzburg-based research and consulting institute Barc conducted a global study on the topic of "data preparation in the specialist area" (695 participants in 50 countries). According to the study, seven out of ten companies are already using the method, while one in ten companies is currently in a project or planning phase. The most important drivers for data preparation are
- 47%: higher expectations of tangible business impact and improved competitiveness through analytics;
- 46 %: growing number of data sources with increasing data volumes, diversity of data types and sources and increasing
- The speed at which data can be generated, evaluated and processed (big data).
Data preparation for the qualitative improvement of master data
According to zetVisions, these results do not sound like new hype, but rather tangible business requirements. Yet data pre-processing by IT and specialist departments is not a new invention. What distinguishes data preparation from its technical ancestors is its relevance and usability for a broader spectrum of non-technical decision-makers. This was made clear by the Boston-based Aberdeen Group in two studies (March 2016, March 2017). Traditional data management was the domain of IT specialists and database administrators. Data preparation is not just for data professionals, but gives a wider circle of managers control over the quality and usability of their data. This includes master data on products, suppliers, customers[1], employees and finances. Marketing managers, sales directors or CFOs - they all take an active role in improving their data.
But there are still a few stumbling blocks to overcome. Keyword "data silos": 46% of the companies surveyed by Aberdeen in 2016 say that the difficulty of making data accessible across departments is their biggest challenge. In addition to the resulting data quality problems, the time factor is playing an increasingly important role. As the business world becomes increasingly dynamic and fast-paced, there is less time to make the hopefully "right" decisions. As a result, the pressure to provide the "right" information in the shortest possible time is growing.
The mantra of data professionals
The right information, at the right time, in the right place - that is the mantra of data professionals. After all, with regard to the insights gained from the data and the decisions made on their basis, the following has always applied: "Garbage in - garbage out". The quality and usability of the data used in the analysis has a direct influence on the accuracy of the findings that emerge on the other side, as Aberdeen rightly points out. Additionally, companies today struggle with the length of time it takes to provide data for analysis. Poor data quality, inadequate accessibility to the right data sources and insufficient speed bring the analysis process to a standstill. It is precisely these problems that companies want to eliminate or prevent by using data preparation. According to Aberdeen, excellence in this area is sometimes defined by shortening the time it takes to make data available for analysis.
However, two results of the Barc study are worthy of note. Although the survey participants are aware of the relevance of data governance for data preparation, only 32% are satisfied or very satisfied with the current status in this regard. It is also surprising which tool for data preparation is by far the most popular: Excel. There is no need to discuss the limited functionalities of Excel for everything to do with data management here. The claim of data preparation to improve the quality, usability, accessibility and transferability of data can only be met in the long term by those who introduce data governance and are also prepared to invest in professional software solutions - not least for master data management.
