What analytics your company needs

What analytics your company needs

A Boeing 787 aircraft generates half a terabyte of data in mid-range flight. The amount of such data after several flights becomes colossal. A recent study by Newvantage Partners shows that 97% of CEOs invest in building and launching Big Data and AI initiatives. Each of these projects is unique and includes different tools to achieve specific goals. Business Intelligence is a powerful tool when used correctly.

There are four different types of analytics, and each is built on top of the other. It is a pyramid where each level supports the next.
Advanced analytics is standalone or semi-autonomous validation of data or content using sophisticated techniques and tools that typically go beyond traditional business intelligence (BI) to gain deeper insights, forecasting, or recommendations. Advanced analytical techniques include data / text analysis, machine learning, pattern matching, forecasting, visualization, semantic analysis, sentiment analysis, network and cluster analysis, multivariate statistics, graph analysis, modeling, complex event processing, neural networks.
Descriptive, diagnostic, predictive, and prescriptive
Descriptive analytics are at the heart of the pyramid. She answers the question "What happened?" , allows you to analyze past performance data to identify strengths and weaknesses. A narrative report for United Airlines can show how many tickets the airline sold in the last month in different markets.
Descriptive analytics is a preliminary stage of data processing that creates a summary of historical data in order to obtain useful information and possibly prepare the data for further analysis. Data aggregation and data mining techniques help identify patterns and relationships that would otherwise be invisible. Descriptive analytics is sometimes said to provide information about what happened. For example, you may see an increase in Twitter followers after a certain tweet.

Diagnostic analytics is built on a narrative basis. It is a form of advanced analytics that analyzes data or content to answer the question "Why did this happen?" And characterized by techniques such as granularity, data discovery, data analysis, and correlation.

It identifies what factors influence positive and negative indicators. If United's narrative report shows that ticket sales are declining, then the diagnostic report may show that this is due to a reduction in marketing costs. Using descriptive and diagnostic analytics, you can move to predictive analytics.

Predictive analytics is a branch of advanced analytics that is used to predict unknown future events (the answer to the question: "what can happen" based on the accumulated information). Predictive analytics uses a variety of techniques, from data analysis, statistics, modeling, machine learning and artificial intelligence to analyze current data to make predictions for the future. It uses a range of data mining, predictive modeling, and analytical techniques to combine business processes of management, information technology and modeling to predict the future.

If United Airlines wants to increase revenue, its forecast report could show how best to allocate marketing spending.

At the very top of the pyramid is the prescriptive analytics. She answers the question: "what should happen." This is the area of ??business intelligence dedicated to finding the best course of action for a given situation. This is where artificial intelligence and machine learning mine past data to find future solutions. Prescriptive data can tell marketers exactly what prices should be quoted to sell tickets with the highest revenue.

Analytics from the bottom up
Companies often prioritize one level of the analytical pyramid without first examining the major levels.

The process begins with the stabilization of the company's operating data. Only after the operational data is properly structured and organized can you think about more advanced analytics. Companies that fail to build an analytic framework often face huge technical costs, business intelligence software purchases that fail to reach its full potential, and weak overall ideas. Once they have mastered descriptive analytics, companies can figure out if they need higher levels.

Published by Oksana Kvitka


No comments yet.

Leave a comment