|Over half of companies (56 percent) don't seem to find compelling business cases for big data with some 53 percent of them having inadequate analytical knowledge / Photo by: Fabrizio Pivari via Wikimedia Commons|
Over half of companies (56 percent) don't seem to find compelling business cases for big data, according to a survey by German-based Business Application Research Center, with some 53 percent of them having inadequate analytical knowledge.
TechRepublic says one of the reasons why companies turn away from big data is due to its complexity. The maintenance of such technology is grueling—sorting it out, cleaning, categorizing, and aggregating with other kinds of data—to provide sufficient information so that corporate decision-makers are able to make informed resolutions.
The tech news site adds this data preparation and integration, along with iterative testing of queries and algorithms, take longer to develop compared to the traditional third and fourth generation reports off fixed, transactional data that enterprises are used to and have been generating for years.
Moreover, decision makers are not inclined to understand big data analytic reports. "This is a fundamental disconnect because users, especially executive users, want reporting that delivers actionable business output," TechRepublic explains.
Reports on big data analytics are also difficult to read, especially if illustrated in tabular formats (as most are). Business executives would rather read business-incisive reports that can be easily digested in one reading.
These are the most prevalent problems in companies when it comes to big data integration. To solve these issues, TechRepublic provided four ways to "break down these common" barriers:
1. It's important to find data integration tools that can automate a significant amount of data ingestion and integration processes. These tools can reduce IT's need to make custom codes for APIs and interfaces.
2. Identifying and developing a precise business case can be done by directly attacking a business pain point (ie. remedying falling revenues). This should be done at the top level of any big data analytics, with the goal of addressing the pain point with algorithms and queries.
3. Make a visual presentation of data results to make it easier to understand. Greater data visualizations can help in building user confidence in the company's big data analytics.
4. While uptime and meantime to response are valuable IT metrics, it's also important to include big data and analytics metrics should determine the following problems:
a. The number of users who continue to use a report six months following the initial issuance.
b. The number of requests received to further improve a reporting product.
c. The number of reports that result in measurable and actionable steps that management is acting on to enhance business performance.