What is Data Analysis?
Data analysis, also known as analysis of data or data analytics, is a process of inspecting, cleansing, transforming, and modeling data with the goal of discovering useful information, suggesting conclusions, and supporting decision-making. ~Wikipedia
Data analysis is “all the ways you can break down the data, assess trends over time, and compare one sector or measurement to another. It can also include the various ways the data is visualized to make the trends and relationships intuitive at a glance.”
Why Data Analysis is required?
Data analysis helps to make sense of our data otherwise they will remain a pile of unwieldy information; perhaps a pile of figures. This is essential because analytics assist humans in making decisions. Therefore, conducting the analysis to produce the best results for the decisions to be made is an important part of the process, as is appropriately presenting the results.
Its an internal organisational function performed by Data Analysts that is more than merely presenting numbers and figures to management. It requires a much more in-depth approach to recording, analysing and dissecting data, and presenting the findings in an easily-digestible format.
With a data analysis we are able to provide a company with decision-making insight into the following key areas:
- Predict customer trends and behaviours
- Analyse, interpret and deliver data in meaningful ways
- Increase business productivity
- Drive effective decision-making
In research, data analysis provides a basis on which to make sense of the data and to interpret it in ways that are relevant to the research purpose and research questions. It provides the building blocks from which the researcher can construct and substantiate an argument about the findings.
How to do Data Analysis?
While each company analyse data specific to its own requirements and goals, some of steps in the value chain are consistent across organizations:
- Define the objective: start with a clearly defined problem or opportunity
- Collect relevant data: identify & collect relevant data
- Clean & transform data: look for incompleteness, redundancy & errors
- Exploratory data analysis: perform EDA with the help of descriptive statistics & data visualization
- Modeling and algorithms: differentiate correlation & causation, apply inferential statistics to measure relationships between variables
- Communicate or present results: based on the domain & users, communicate or present the results including data visualization
Case Study: Rathi Pizza Inc
Lets get back to our case study to relate how we can perform data analysis.
First, lets define our objective, which can be increasing sales or optimizing operations. Once objective is defined we need to see what are the key matrices to measure the objective? On what data these matrices are dependent upon? In which data stores that data resides & how to collect that data?
During the analysis, we can find certain quality issues with data like data is not complete, data is not consistent, other errors with data. We need to decide how to overcome these quality issues? After cleaning the data, we perform exploratory data analysis (EDA) to see how our features (data attributes) are related, we try to identify the relationships with the help of descriptive statistics & visualization of patterns & relations.
Once above steps are completed, we differentiate between correlation & causation, apply models on data for predictions or insights generation. Please note that data cleaning/transformation, EDA & modeling are performed in iterative manner as these activities are interrelated. Once we have generated some recommendations or insights, we look for means to present or communicate the results or findings to the customer or business.
So after data analysis, we will get actionable insights or recommendation on what steps to take to increase sales or optimize operations.