Advanced Analytics– Part 1

Advanced Analytics to create business values

March 18, 2022

Today, data & analytics tools and processes are widely used to collect data and analyze it to benefit the business in a variety of ways. The results are often visualized via reports & dashboards generated using tools such as Power BI, Tableau, etc.

Each organization can have different sets KPIs depending on the goals or missions of the organization at that time, such as sales analysis, cost analysis, analyzing opportunities, etc.,

Although valuable, this form of analytics is reactive and only gives you a perspective on what has happened or, in the case of real-time analytics, what is happening right now. But does not provide insights into what will or what might happen in the future and what you can do to either make something happen or prevent it from happening. This is the field of analytics that is referred to as predictive and prescriptive analytics.

This article is an overview of each type of analytics and provides examples of the results of each.

1. Descriptive Analytics is a fundamental analysis to summarize the results, by collecting historical data through various processes to be able to answer questions and describe what has happened or what is happening through information such as:

  • Sales volume in the past to present years
  • Efficiency and working status of the machine.
  • Cost and inventory management information.
  • Customer profiling
  • Products and market demand.

This is the first step to understanding what is happening in the areas of interest and subsequently leads to informed decision making.

2. Diagnostic Analytics determines the “why”, by analyzing the patterns and trends in data to understand:

  • Why are the sales in April lower than the average?
  • Why are we experiencing downtime in the production line?
  • Why is our inventory not meeting the market demand?
  • Why are our products not meeting the needs of customers?

Diagnostic analysis requires the right volume of data to be able to determine the patterns that affect the data we are interested in that can be caused by one or more factors.

For example, frequent downtime of a production line could be due to old machines that have not been properly maintained or caused by human errors from the staff, etc.

3. Predictive Analytics, can predict what will or might happen in the future by analyzing historical trends and patterns to predict and forecast. The accuracy of this depends on the volume of historical data available and the quality of the data to answer questions, such as:

  • What will the company's revenue be in the next three months?
  • Will our current staffing resources be sufficient to meet demands for the upcoming holiday period?
  • When will our machinery be likely to malfunction?
  • Will our inventory be adequate for meeting demands for the next quarter?
  • Will our new products meet customer's demand?

Predictive analysis can be performed through computational processes and statistical procedures and can often be complex and time-consuming. However, automating through machine learning tools and technology or Artificial Intelligence, accuracy can be improved and reduce operational resources.

For example, forecasting inventory management plans for the next year may require at least 5-years of historical data to analyze every inventory item to get the appropriate results. This might take two people and spend more than a month to create, and if the current data has changed from the forecast, they may have to restart the whole process. But if machine learning technology is applied, it can reduce analysis time and bring new data for recalculation immediately to get accurate forecast results and timely responses.​​

4. Prescriptive Analytics is the process of using data to determine an optimal course of action, by combining steps 1-3 mentioned above and results in recommendations for next steps, such as:

  • If you want to increase your profits in the summer, what products and promotions should you recommend to your customers?
  • When is the right time for maintenance to prevent production line downtime?
  • How much inventory is required to reduce the problem of insufficient products for future demand?
  • Which product distribution channel is most suitable for targeting a particular group of customers?
  • What is the appropriate quantity for purchasing raw materials each month to control costs as much as possible?

Each of the above processes is continuous and involves developing procedures and best practices to bring real value to the business. Many organizations may develop different types of data analysis systems according to the objectives and capabilities of the organization.

No matter what stage you are in, by building a data platform to support advanced analytics you can be sure that you are using information contained in the organization to provide a foundation for furthering analysis, achieving real benefits and improving the efficiency of many processes.

Finally, we hope this article is helpful in planning and providing a guideline for making an advanced data analytics roadmap for your more business values.

If any organization is interested in advanced data analytics or you have questions about developing either the readiness, process, or technology/tools, please feel free to contact us for a free consultation by filling out the information below or directly at +66(0)2 117 4344.

We are happy to serve you.