5 data analytics myths debunked
Perplexed by Data Analytics? Stuck on statistics? Then fear not, Philip O’Donnell, Forensic Data Analytics Consultant is here to guide you through the fascinating world of analytics, explaining complex concepts, tackling technical terms and showing the power of data in a series of business scenarios. In his first blog, Philip will debunk some of the most prevalent myths surrounding data analytics. Over to you, Philip!
1) Once you have an analytics tool, anyone can be a data analyst
Father of Data Analytics, John Tukey, summed up the aim of analytics in typically succinct manner by stating,
“The greatest value of a picture is when it forces us to notice what we never expected to see.”
Put broadly, data analytics is a process to uncover hidden patterns, unknown correlations, market trends, customer preferences using mathematical and statistical techniques.
However, many people think data analytics is just a tool that turns data into graphs and that once you have this tool, anyone can analyse data. This is a little like saying that by owning a saw, you are a master carpenter! To get the most out of data analytics, it is imperative that the right techniques as well the steps in the process must be understood and used in the right context to be truly effective in any investigation and if performed incorrectly can have misleading discoveries.
2) Data analytics is just for auditors
Where there are people, there is data and this data can be analysed and used to improve the way we operate. Music industry moguls use data analytics to measure listener responses to new music. This then helps them work out which genres, and new artists, are likely to bring them a hit. Analytics is used by all spheres of society, from medical research and environmental studies to more obvious financial applications. Even Hollywood screenwriters have discovered that analytics can produce great success stories. In the Oscar-winning film Moneyball, a poorly performing baseball team hired a statistics expert to help them change their drafting procedures. By using statistics to help select players rather than traditional scouting methods, the team went onto have the longest winning streak in baseball history.
Analytics helps people in all industries make better, more informed decisions and deliver new innovative ways of thinking and doing business.
3) Data context doesn’t matter
The key component to performing any analytics is to understand the environment in which the client operates. Interpreting and advising on findings is a key aspect of the analytics process, so to really add value for clients, sector knowledge is vastly important. The most experienced data analysts need to understand the context of the data, especially in high profile legal investigations, banking cases, corporate compliance, financial analysis, and government projects. Clients looking to get the most out of their data will need to choose a provider who is able to harness industry knowledge and take a pragmatic approach to data science and analytics methodologies.
4) Analysing data can compromise the security and integrity of data estates
This myth does have some truth in that many inexperienced analysts do not understand the importance of a proper data extraction exercise. Direct extraction of raw data from core system is a key step in the analytics process and in the past, I have seen where incomplete and incorrect data extraction has caused data analytics investigation to be invalidated. However, an experienced data analytics provider is rigorous in ensuring data extraction is performed correctly and is accountable in the chain of custody. Done properly, performing extraction ensures that the complete dataset and minimises the risk of an incomplete investigation. Extraction is performed in such a way that it does not compromise existing security of the data as well preserving the integrity of the system. Extraction can be performed on multiple data sources. These include relational databases, data warehouses as well legacy flat files and dynamic xml formats.
5) Analytics techniques don’t change
Data analytics is an incredibly dynamic discipline and new techniques are being developed all the time. A good analytics provider will always stay abreast of the latest trends and methods. So what is in store for 2016?
According to the International Analytics Institute the number one trend for analytics in 2016 will be that the distinction between cognitive analytics and automated analytics becomes blurred. Automated analytics is the changing of an airplane price or stock price based on the real-time analysis of factors such as customer demand or other market forces. Cognitive analytics is the inspired by how the human brain processes information, draws conclusions, and codifies instincts and experience into learning. Cognitive analytics uses machine learning techniques such as Neural Networks, Logistic Regression and historic data. By understanding the human decision making and learning process, data scientists can incorporate this knowledge into their models and achieve even more accurate and in-depth insights.
Click here to find out more information about our Data Analytics service.