Programming
How are decisions made in Data Analytics?
Oct 27, 2025
4-5 min Reading time
Data analytics, one of the most important fields of the current era, is one of the main factors in achieving success in business and various fields. With the rapid development of technology, the volume of information is increasing, and data analytics plays an important role in analyzing this data correctly and drawing conclusions.
For example, a B2B SaaS company can use data analytics to learn which features corporate customers use most in their software, allowing them to prioritize product development and increase customer satisfaction.
Data analytics professionals collect raw data, organize it, and then analyze it. Once the analysis is complete, they transform this vast amount of data into clear and simple insights that enable stakeholders to take concrete actions.
With the advancement of technology, data analytics has become more easily applicable today. This enables data-driven decisions to be made at all levels within an organization.
Types of data analytics
Data analytics is divided into 4 main categories: Descriptive, Predictive, Prescriptive, and Diagnostic analysis.
Descriptive analytics
This type of data analytics looks back to the past and describes what happened. It explains what happened by examining patterns and trends in a data set. This is where most companies start their analysis.
Example: A retail company analyzes sales data from the past year to identify seasonal trends. For example, while demand for winter clothing increases in November and December, interest in beachwear peaks in June and July. Thus, data analytics helps shape inventory planning and marketing strategies.
Predictive analytics
Another type of analysis used in data analytics is predictive analysis. It uses past data to make predictions about the future. This approach allows you to anticipate future risks and opportunities.
Example: A subscription-based streaming platform analyzes user viewing habits, subscription duration, and customer reviews to predict customer churn. This analysis helps to understand when subscriptions are canceled.
Prescriptive analytics
This type of data analytics, unlike predictive analytics, not only shows what will happen, but also which decisions will be most effective. To this end, technologies such as statistical models and machine learning are used to determine the most appropriate course of action for companies.
Example: A logistics company analyzes past traffic data, weather conditions, and delivery times to optimize delivery routes. Predictive analysis predicts delays, while prescriptive analysis suggests the most efficient routes.
Diagnostic analytics
Diagnostic analysis seeks to understand the causes of past events. This type of analysis, unlike descriptive analysis, helps to uncover the reasons behind events.
Example: An e-commerce platform experiences a drop in sales. Data analysis shows that the drop occurred after a website update. The update resulted in slower page loading times. Customer behavior analysis reveals that cart abandonment rates increased during that period.
Diagnostic analysis in data analytics allows businesses to identify root causes of events in order to replicate past successes and avoid negative outcomes.
Key steps of the data analytics process
1. Data collection
First, check the data you need for your analysis. This could be customer data, user experience (UX) interviews, or website impressions. The data can be found in your CRM system, past marketing campaigns, or analytics tools. The process of collecting data for data analytics is a dynamic activity that is constantly updated and occurs in real time. Database Administrators are usually responsible for collecting data.
2. Data retention
The collected data must be stored securely. This often requires converting it into a format that can be used by ETL (extract, transform, load) or ELT (extract, transform, load) processes. The raw data is cleaned, formatted, and organized in storage systems such as data lakes. The storage phase is critical to the reliability of data analytics.
3. Data cleansing
Check for inconsistencies, duplicate data, or missing data. Properly cleaned data ensures that data analytics results are accurate and reliable.
4. Analysis
Use software tools to process and interpret data, thus uncovering key insights that will guide decision-making. This stage is the most strategic part of the data analytics process, as this is where real decisions are made.
