Data is no more important than the knowledge one can extract from it. For data to be useful, it has to be organized: first it needs to be contextualized (by determining what information to look for), and then it must be curated (by taking care of the nitty-gritty details about managing a data repository). The efficient execution of these two tasks create value by optimizing the decision making process, allowing those in charge to make sharp analysis on business trends and future strategies. So, data can only be fully capitalized when those who understand the business dynamics are served with data in which they can trust. This allows for a quick decision-making process and the ability to react in real-time to the contingencies of the fast-paced modern world.
Efficient and articulated analysis depends on clean, consistent and trustworthy data. The more confidence there is on the quality of the data , the less time is spent worrying about data quality and the more time is dedicated to the actual work of analysis. In this sense, Unicage provides a full set of robust, modular tools that allow for the creation of tailored solutions to any data quality pipeline. The Unicage system leaves a small footprint on the system’s available resources no matter how long and complex the data preparation process is.
Figure 1: Unicage data triggers in action.
A typical example where the Unicage system excels is in the inclusion of data triggers at any stage of the data preparation pipeline. A data trigger is any process that reacts to a given data pattern. Let’s suppose there is a set of data points that should be received from the database with a specific format; any data point that fails to satisfy such format requirement should be dismissed. In such case, during the data quality pipeline, the Unicage system can not only remove the invalid data points, but also implement, in an efficint way, a data trigger that reports the existence and the frequency of those invalid data points. This allows for a continuous improvement of the data quality pipeline and for the analysts to build confidence on the data repository and on the data itself.
The modular approach of the Unicage system allows the development of data quality pipelines based on the combination of simple, but very efficient, operations. Modularity also means that any pipeline problem can be addressed quickly and accurately. By using such an approach, it is possible to increase the trust the analysts put on data. Trust is key to any relationship, do you trust your data?