Data Analytics 101 Series — The ‘Share’ & ‘Act’ Phase

Adith Narasimhan Kumar
Analytics Vidhya
Published in
5 min readJan 15, 2024


What happens when the analysis phase is complete? Is the process complete? No. The analysis performed is useless unless it is shared with the stakeholders for them to arrive at a business decision. Few techniques help us with communicating our insights deduced from the analyzed phase. Let’s have a look!

This phase of the data analytics pipeline is pretty straightforward and I will keep this article as crisp and concise as possible. The analysis that you perform is complex and full of numbers. Let’s be honest! Who likes to look at pages and pages of numbers right? It is for this very reason we use data visualization techniques to convey our analysis to our peers and other stakeholders.

A picture conveys more than pages of words or numbers.

Also, another importance of this phase is to get other people’s views on your analysis. Right after the “Ask” phase, the aspect of an independent set of eyes vanishes. The following phases — prepare, process, and analyze are done by a single person and the same set of eyes look at everything! This makes you immune to any errors or outliers which might result in incorrect conclusions.

Common visualization tools used are Power BI, Tableau, and QlikView.

Types of Visualization Techniques

Each dataset is different. There is no one-size-fits-all concept when it comes to data! The type of graph/chart entirely depends on the use case and the dataset. However, let us look at a few common graphs/charts and their most common use cases.

1. Pivot Tables

Pivot tables are a great choice to summarize a large dataset. It is very handy because pivot tables can handle change in the underlying data range just by a “refresh”.

Source: Career foundry

2. Boxplots

Box plots are used to visualize the key statistics of a dataset. The metrics displayed in a box plot are minimum and maximum values, the median value, and the lower and upper quartiles. Also, box plots are non-parametric, i.e. they display the data without the assumption of the underlying distribution.

Source: Career foundry

3. Scatter Plots

Scatter plots are used to show the correlation between 2 variables in a dataset. Each line item in the dataset is shown as a single point in the plot. This type of plot is particularly useful to visualize the relationship of variables in a large dataset.

Source: Career foundry

4. Line Graphs

Line graphs are best used in time series analysis. They are very close to scatter plots but are connected. This helps us view the trends such as acceleration, deceleration, and volatility of a feature against time.

Source: Career foundry

5. Area Charts

Area charts are similar to line charts but also visualize the volume of the parameter in analysis. This helps us measure the extent of the impact of a particular feature.

Source: Career foundry

6. Bar Charts

Bar charts are used to plot a categorical variable (Name, Type of fruit) against a numerical variable (Number of people with the same name, Type of fruit). It also helps easily visualize data and is one of the most used and common chart types.

Source: Career foundry

7. Pie Charts

Pie charts and bar charts are very similar. Both map a categorical against a numerical variable. However, a pie chart measures the contribution of a component against the entire “Pie”. This is a very common and easily used chart.

Source: Career foundry


The objective of the act phase of the data analytics pipeline is to recommend actions based on the insights gained from our analysis upon which impactful business decisions are made. The ultimate objective of any analysis is to show results and spark data-driven business decisions and hence this phase is as important as all the previous phases.


To conclude, the share phase of the analytics pipeline is very important as it is essential to get an independent person’s input and to convey the outcome of the analysis in a simple, easy-to-understand, and useful manner to the respective stakeholders using which they can act and solve the underlying issue. After all, that’s why we do the analysis right!

This article marks the completion of the “Data Analytics 101 Series”. A big thanks to all of you for your continued support 🙏and am looking forward to the next series!

Do let me know on any topic you want me to cover! will try my best to write on the same. Let’s all learn together! 😁

Hope you liked my article! Do share your thoughts and inputs and I’ll try to have them answered to my knowledge in future articles!

Happy Learning!

Check out my other articles on Blockchain and Machine Learning/Deep Learning. Let me know about any other topics to cover in the future!

Catch my previous article here 👇