Statista’s data suggests big data market volume is set to top $100 billion over the next five years. Companies are willing to invest their resources to get powerful knowledge to know themselves and their customers, improve their processes such as supply chain, forecast sales, predict market behavior to create better customer experiences, services, and/or products, and get advantages over their competitors. This is not easy to achieve and it requires understanding the foundations, so let’s review the 4 levels of Analytics.
Descriptive: This is the baseline where a company should start. Usually, it shows the historical data of a company and answers the question “what happened?” in the form of reports, pie charts, bar charts, tables, and other visualizations.
For example, the sales amount by month over the last 5 years or the top 10 of sold products.
Diagnostic: in the second level, the data is analyzed to answer the question “why did it happen?” and it uses techniques such as drill-down, data discovery, data mining, and correlations.
For example, if the general sales of a company are decreasing, the analysts can go down one level to review the sales by segment determine which ones are affected, and go as deep as needed until they identify a root cause.
Predictive: The third level of analytics is advanced and complex and it answers the question “what is likely to happen?”.
It involves techniques such as regression analysis, forecasting, multivariate statistics, pattern matching, predictive modeling, and forecasting.
For a company to achieve this level is a big step, since it requires a large volume of high-quality data, deep knowledge about statistics, and programming languages such as R and Python.
Probably the largest sector to use predictive analytics in retail, as it is always looking to improve its sales position and forge better relations with customers. For example, forecast sales trends at various times of the year and plan marketing campaigns accordingly.
Prescriptive: the last and most advanced level of analytics. It tries to answer the question “what should be done?”.
This type of analytics is characterized by techniques such as graph analysis, simulation, complex event processing, neural networks, recommendation engines, heuristics, and Machine Learning.
It’s not easy to achieve and its accuracy depends on the maturity of the levels above.
Something to consider is that prescriptive analytics is a valuable tool for data-driven decision -making but those are only suggestions that do not replace human discernment.
In conclusion, data is one of the most valuable assets for a company as it represents all its past, present and even can help to determine its future. But data itself is not helpful if it is not taken, represented, and analyzed properly to transform tons of rows and columns from a database to insights that truly lead to understanding a company’s situation and drive to better decision making.