Third in a three-part blog series on exploratory analytics
As business conditions change and requirements evolve, so must the tools that we use. A farmer tilling a small garden with good soil could finish the job quickly with a pitchfork. In contrast, preparing large agricultural fields or rocky terrain requires powerful industrial cultivators to make sure the land is ready for the growing season.
So it is with data analysis. Enterprises that want to gain insight from their burgeoning data assets are struggling to realize value from traditional analytics tools. Companies are increasingly mining longer datasets with more records and wider datasets with more columns and attributes. As these data assets expand, it becomes more difficult for humans to grasp the meaning of the data residing within.
What’s more, longer datasets require more time for analysis, increasing the time needed to generate query results that can lead to insight. Wider datasets are also difficult to interpret. Business users can’t know in advance which parts of the dataset are important or relevant to addressing a given problem. In fact, analysts may not be able to tell whether the available datasets contain any information that would be useful for their problem.
In my previous blog, we discussed how classic and advanced analytics have been historically used to generate new business insight. But the classic approach is often not up to the task of analyzing longer and wider datasets.
Exploratory analytics as a foundation for insight
That’s why many organizations are investigating exploratory analytics. This technology uses advanced algorithms to automatically highlight the most important findings in the dataset search and suggests the best way to visualize the information. In this way, exploratory analytics extends the methods and insights used in advanced analytics tools to help users focus on critical data resources. Exploratory analytics also overcomes the human limitations imposed by classic analytics by helping analysts to easily make sense of long and wide datasets.
SAP supports the use of exploratory analytics through the predictive analytics software. By enabling workflows, users can analyze datasets to find outliers, influencers, patterns in data, correlations between variables, and data quality issues – all within just a few steps. Users can even quickly answer data-centric questions, such as:
- Does this dataset contain useful information for my business problem?
- Which parts of this dataset are influencing the answer I seek?
And because the discovery of critical information is automated, users often find more than just the answer to their original question. With this strength of exploratory analytics features, users gain unexpected insight.
Today’s growing data assets offer fertile soil for enterprises seeking to understand previously hidden patterns and trends, and exploratory analytics is a critical tool for developing new business insight.
Other blogs in this series: