What is Data Analytics? And Why is it Important?

What is Data Analytics?

Data analytics is defined as the process of analysing raw data to identify trends and patterns.  Increasingly, many different companies and organisations are starting to recognise the value of data analytics, and are seeking to apply it to improve their business / organisational outcomes.

Data analytics can be categorised into 4 different types:

1. Descriptive Analytics
2. Diagnostic Analytics
3. Predictive Analytics
4. Prescriptive Analytics

Descriptive Analytics

Descriptive analytics is used to understand “what happened”. Practitioners would use a combination of summary statistics (e.g mean, median) and metrics to form a big picture understanding of the health of business and operations. It takes a combination of good understanding of statistics and domain knowledge of the industry in question to decide what are relevant statistics and metrics that best describe the context.

One key thing to note is that descriptive analytics does not make decisions or projections for you. It is focused on the meaningful and descriptive presentation of data.

Diagnostic Analytics

The next stage is diagnostic analytics which is concerned about “why it happened”. Here, practitioners would use data visualization techniques to help diagnose the reason behind why certain things happen. 

One of the most classic case in point would be when a team from DSTA made use of several data visualisation to determine the cause behind the spate of train breakdowns in the circle line.

Predictive Analytics

Next, we have predictive analytics, which is what most aspiring data scientists are most interested in. Predictive analytics involves employing various techniques to extract and refine data, and thereafter generate predictions that can help with decision making. Specifically, data scientist would employ machine learning models such as logistic regressions to generate the predictions.

One classic application of predictive analytics is: use machine learning model to predict whether bank clients will default on their loans. In sum, predictive analytics can be said to be concerned with “what is likely to happen in the future”. One fun fact is that while machine learning only gained traction in the recent decade, the machine learning models actually existed several decades ago.

However, it only became viable to use these models to parse through data given the proliferation of cheap and scalable computing power.

Prescriptive Analytics

Next, we have predictive analytics, which is what most aspiring data scientists are most interested in. Predictive analytics involves employing various techniques to extract and refine data, and thereafter generate predictions that can help with decision making. Specifically, data scientist would employ machine learning models such as logistic regressions to generate the predictions.

One classic application of predictive analytics is: use machine learning model to predict whether bank clients will default on their loans. In sum, predictive analytics can be said to be concerned with “what is likely to happen in the future”. One fun fact is that while machine learning only gained traction in the recent decade, the machine learning models actually existed several decades ago.

However, it only became viable to use these models to parse through data given the proliferation of cheap and scalable computing power.

Importance of Data Analytics

Next, we have predictive analytics, which is what most aspiring data scientists are most interested in. Predictive analytics involves employing various techniques to extract and refine data, and thereafter generate predictions that can help with decision making. Specifically, data scientist would employ machine learning models such as logistic regressions to generate the predictions.

One classic application of predictive analytics is: use machine learning model to predict whether bank clients will default on their loans. In sum, predictive analytics can be said to be concerned with “what is likely to happen in the future”. One fun fact is that while machine learning only gained traction in the recent decade, the machine learning models actually existed several decades ago.

However, it only became viable to use these models to parse through data given the proliferation of cheap and scalable computing power.

How Can Businesses Benefit From Data Analytics?

In a broader sense, big data can significantly impact a company’s future. You have the priceless asset of industry knowledge when you have a data strategy and the findings. Keep that knowledge in mind as you monitor economic changes and look for ways for your company to expand – and expand some more.