Understanding Big Data Analytics (Essay Sample)

Understanding Big Data Analytics
Name:
Course:
Tutor:
Date:
Understanding big data analytics
Introduction
Understanding how best and to what extent executives can rely on big data analysis when formulating business strategies is very important (Palem, 2014). There is no doubt that big data sets help analyze terabytes of data within a short period. Most executives find it a worthy investment as compared to traditional methods of data analysis. However, it seems good enough to understand the level to which one can depend on it. It is clear that companies collect data of their activities and keep them in databases. This may include information about how many customers are exchanging ideas concerning their brandonline and even the number of customers an organization has. Generally, big data is defined as large sets of data which common software tools find it difficult to manage and process within the required period of time. Challenges showing up in all aspects of data management such as sharing, capture, curation, transfer and analysis need to be considered when deriving a business strategy. This paper thus provides an insight aimed at making executives understand when and to which levels should they depend on big data analysis for formulating business marketing strategies.
Understanding the concept
It is clear that big data analytics is a field on the rise and enjoys a great deal of diversity. This thus leaves its definition alone not quite helpful enough. Any company wishing to fully utilize it must master the underlying concept correctly (Webster, 2010). There are numeroustechnologies associated with Big Data analytics from which one can draw conclusions on the perfect strategies for a particular company. To slice it down for executives who find it difficult, it is important to look at it from two perspectives; as a storage platform and as a problem solver (Palem, 2014). As a storage platform, it helps stores volumes of data from multiple sources in a reliable manner. For instance, it can be used to keep real-time data from weblogs, sensors and GPS locators for to enable instant access by a large number of users at a given time. From a problem solving perspective, uses a large number of distributed computing machines to lessen the "time- solution" ratio. For example, when exploratory analysis of credit cards data help identify or detect any signs of fraud. Palem opines that this is the main feature that highly differentiates other techniques of data analysis with Big Data. Combining the two, as a storage platform and as a solution enabler, gives what is called Big Data analytics.
With the traditional data warehousing processes being too slow and unable to bring data from structured and unstructured sources together for clear analysis, the Big Data technique seems to be the perfect option.It is only after CEOs delve deep into the field that they can be able to understand when and to which extent Big Data can be useful to the organizations. Researches on how effective data usage can aid in development show that this technique has the capacity to contribute much on organizational development. However, Big Data analysis presents its own unique challenges to the different areas employed. It has become a crucial part of decision making in different areas of development thanks to its cost effective nature. The concept is not that difficult to understand, but the complexity arises when it has to be relied upon in coming up with marketing strategies. An executive can only understand the right stage at which a company requires to embrace Big Data analysis if he/she comes to terms with some of its characteristics.
Characteristics of Big Data
In trying to define Big Data, 2012, Gartner describes it as a high volume of data with high velocity and variety that calls for upgraded forms of processing to provide an insight during the decision making process. Many people can easily relate this to Business intelligence, but there is a clear line between them regarding the manner in which data is used. Big data utilizes inductive form of statistics and frameworks from nonlinear system identification to try and provide relationships in large sets of data aimed at predicting outcomes. To add on the 3Vs (velocity, volume and variety),complexity and variability are the other characteristics of Big Data.
These features are what prompt a CEO to ditch the so termed ineffective traditional tools to try out Big Data analytics. There is no doubt that the effectiveness of the process in managing and acting as a compass in basic decision making has crowned it corporate agenda in most organizations and even governmental sectors (Barton,&Court, 2012). A Harvard Business Review by the two clearly opines that many CEOs admire the manner in which Google, IBM, Amazon and even Hewlett-Packard eclipse their competitors with their strange ability to analyze Big Data for betterment. This trend has generated plenty of hype as every executive wants to pay attention. A study by Erick Brynjofsson and Andrew McAfee of MIT revealed that companies that subscribe to the whole tide of Big Data analytics record more than 6% higher profitability and productivity rates than their competitors in the same field. Though many of them opting for it, there is need to understand the degree to which this can impact on the decision making process of any company. This makes it important to consider its characteristics before defining the level to which marketing or other business strategies are drawn based on predicted outcomes.
General approach to analytics
Many executives are aware of the approaches to analytics, but most of them have to deal with the issue of integration. It may looks simple, but the complexity when drawn against the desired depth of the solution calls for a cautious integration of analytics with the Big Data itself. This has to clearly occur when the real-time data streaming and the persistent historical one are brought together before making decisions (Palem,2010). For instance, real-time data streaming from shopping carts that seeks to help analyze customer behavior is processed against stored historical data. This may be past information on other customers and is always large in volume. Stream processing thus helps identify any patterns in the real-time data that may be similar to that in the pre-stored data. In rare cases, the old data may not be in format ready to be used. A company has to have an analytics processing unit working alongside cloud storage to ensure old data is ready to be used. Results are then converted and computed into statistical scores that can offer insight into the kind of decision to be made. A central command center keeps watch on the whole process while taking care of all alerts and monitoring operational compliance.
Critiques of the Big Data approach
Critiques of the big data on the decision making processes focus mostly on its challenges and the manner in which it is done. This comes as companies continue to invest lots of cash to try and get insight from data streaming in from customers and suppliers. However, Reips and his colleagues are of the view that strong assumptions are made due to its mathematical nature that may not necessarily represent what happens at the lower level of micro-processes. Their critique goes on to spell there is a great need for Big Data to be contextualized according to their political, economic or social environments. Decision making in any company depends on its infrastructure in terms of information flow. In coming up with any IT strategy, it is vital for all the employees in the IT department to have the skills needed in handling Big Data with an aim of providing insights, (Nobel. C,2010). However, companies are investing so much in big data analytics while very little is cultivated into employee skills. According to Harvard Business Review article, the comprehensive analysis of big data has to be complemented by a bigger judgment.
As seen earlier, the main aim of Big data analysis is to provide insight from which marketing decision are to be drawn. However, it seems good to be clear that nearly all of the decisions made are based on historical data thus in the past or at best on what is happening currently. There is no doubt that this large volumes of data can help predict future outcomes. What an Executive should worry about when determining the limits to which big data analysis can be important is if the future will be like the present times. If the organizations is so sure that the future will be similar to the present, then coming up with a strategy on that scope would be beneficial. It has been suggested, as a response to the critique, that combining computer simulations with such as Agent based model with the big data approach for greater efficiency. For instance, records show that agent-based-models are becoming better in predicting outcomes of social complexities through computer simulations from interdependent algorithms. Furthermore, multivariate techniques such as cluster and factor analysis are claimed to be more useful approaches that are miles ahead of the bi-variant ones applied on minor data sets.
Another way in which a CEO wishing to utilize big data analyt...