10 Pitfalls to Avoid in Big Data Analysis
Big companies and startups see the long-term goals in Big Data. But are they asking the right questions? Are they getting the right messages from Big Data? Here is a list of 10 things to avoid.
1.Do not make decisions from Big Data
Many companies do not realize how much information they can extract from Big Data. There are many tools and solutions that can help make decisions from Big Data.
2.Do not engage Data Scientists
Data collection is easier than ever and tools to work on Big Data have also become much more accessible. The problem is often that companies do not have a qualified Data Scientist or someone that can interpret these data. Instead, they rely on untrained staff. The misuse or poorly data analysis is very harmful to the business.
3.Answering the question “what”
The biggest mistake that companies make in the use of Big Data is to try to respond to relatively trivial questions such as “what”. Big Data is not the answer to the “what” but the “why”. Big Data aims to join datasets that have never been joined before, and ask the questions that were never asked; for example, the question of why customers and employees do what they do.
4.Focusing on the processing of data at the expense of analysis
The first part of the Big Data challenge is to find the right algorithms and approaches to ingest vast amounts of information. The second is more neglected – the challenge of finding a way to present the findings to make them usable and easy to understand. Too many companies focus on “how do we deal with all this data?” at the expense of “how can we put into action? ”
5.Failure to follow the time and cost
Many companies have not invested enough in time tracking and cost management, particularly in service businesses. Having the proper tools to track time for true cost accounting is very important. The company will be able to express the true margins of each offer. This information can be used in particular to increase efficiency and profitability.
6.To confuse correlation with causation
When companies work with large data, a major error (and common) is to assume that correlation implies causation. While you can use data to understand the correlation, assimilate to the “cause and effect” can lead to erroneous decisions and unsuccessful results. To distinguish between correlation and causation is essential to the use of data for better results.
7.Overloading the process
Too often, companies invest in expensive tools from the start, requiring a team of analysts, but the team in place or capacity to process information. It is easier (and cheaper) to start with basic tools, and grow according to its learning and missing tools.
8.Thinking Too Big
Big Data is hype and rightly, but companies must think small to leverage their data. A Big Data project can be very expensive and create high recurrent costs. Launch projects by solving real problems and develop solutions to respond is a safer way to see a real return on investment.
9.Be led by the Data
Some companies give too much decision power to Big Data. We must use the data as a factor, but not as an absolute decision in the corporate strategy.
10.Put customers in boxes
The tendency to simplify and ignore the change and the complex nature of individual customer needs is often a cause of partially false results. Do not put clients in boxes! Companies need to listen, understand and find customized solutions, based on what the customer wants. The client is the king.
Source: http://www.huffingtonpost.com/young-entrepreneur-council/12-major-mistake-companie_b_6849616.html
