Getting Started with Tableau: Interface, Data Connections and Your First Views

What is Tableau?

Tableau is a visual analytics platform: you drag fields onto shelves and Tableau writes the query, aggregates the data and draws the chart. It is owned by Salesforce and, with Power BI, is one of the two BI tools most often named in Indian analyst job posts. Where Power BI leans on DAX formulas and the Microsoft stack, Tableau is known for fast visual exploration and highly customisable charts.

Product	What it is	Cost (2026)
Tableau Public (desktop app + website)	Free version of Desktop; workbooks are saved publicly on public.tableau.com	Free
Tableau Desktop	Full authoring tool; saves locally, connects to databases	Paid (Creator licence)
Tableau Cloud	Tableau's hosted service for sharing, refresh schedules and permissions	Paid, per user
Tableau Server	The same as Cloud, but installed on your company's own servers	Paid
Tableau Prep Builder	Visual tool to clean, join and reshape data before analysis	Included with Creator
Tableau Pulse / Tableau Agent	AI features: metric digests in plain language and an assistant that helps write calculations and build views	Cloud add-ons

TipWhich one should you install?

Learners should install Tableau Public (free). It has almost every authoring feature of Desktop. The two limits: workbooks are saved to your public profile (so never use confidential data) and it connects mainly to files (Excel, CSV, text, JSON, PDF, spatial) and Google Sheets rather than company databases. Your public profile also works as a portfolio for interviews.

Install and connect

Step by stepInstall Tableau Public and connect to Superstore

1. Go to public.tableau.com → Create → download Tableau Public for Windows or Mac and install it; sign up for a free profile.
2. Open Tableau Public. On the Connect pane choose Text file and pick superstore_sales.csv (for an Excel file choose Microsoft Excel).
3. The Data Source page opens: the table appears in the canvas and a preview grid below it.
4. Check each column's data type icon (Abc = string, # = number, calendar = date, globe = geographic). Click an icon to change it.
5. Click Sheet 1 at the bottom to start building.

The data you connect to is the same Superstore India file used in the Excel and Power BI lessons. Check its size before you build anything:

PythonCopy

import pandas as pd
df = pd.read_csv("superstore_sales.csv", parse_dates=["order_date", "ship_date"])
print(df.shape)
print(df.dtypes.astype(str).value_counts().to_dict())
print("Orders from", df.order_date.min().date(), "to", df.order_date.max().date())
df.head(3)

Output

(6323, 19)
{'object': 12, 'float64': 3, 'int64': 2, 'datetime64[ns]': 2}
Orders from 2023-01-01 to 2025-12-31

	row_id	order_id	order_date	ship_date	ship_mode	customer_id	customer_name	segment	city	state	region	product_id	category	sub_category	product_name	sales	quantity	discount	profit
0	1	IN-2023-00001	2023-01-01	2023-01-05	Second Class	C0765	Ishaan Gupta	Consumer	Bengaluru	Karnataka	South	P0059	Technology	Phones	Nova Phone 223	36,000	2	0	3,472.55
1	2	IN-2023-00001	2023-01-01	2023-01-05	Second Class	C0765	Ishaan Gupta	Consumer	Bengaluru	Karnataka	South	P0050	Office Supplies	Appliances	Orbit Appliance 798	3,850	1	0	1,169.54
2	3	IN-2023-00001	2023-01-01	2023-01-05	Second Class	C0765	Ishaan Gupta	Consumer	Bengaluru	Karnataka	South	P0011	Furniture	Tables	Orbit Table 248	62,080	4	0	6,044.47

3 rows × 19 columns

6,323 rows (order lines) and 19 columns: Tableau will show these as fields in the Data pane.

The Data Source page

The workspace

Dimensions vs measures, discrete vs continuous

Tableau sorts every field in two ways, and the colour of a pill tells you which:

A measure dropped into the view is aggregated (default SUM). Right-click the pill → Measure to switch to Average, Median, Count, Count (Distinct), Minimum, Maximum, Standard deviation and so on.

Your first views

Sales by region (bar chart)

Drag Region to Rows and Sales to Columns. Tableau writes the equivalent of SELECT region, SUM(sales) FROM superstore GROUP BY region and draws one bar per region:

by_region = (df.groupby("region")[["sales", "profit"]].sum() / 1e7).round(2).sort_values("sales", ascending=False)
by_region.columns = ["Sales (₹ crore)", "Profit (₹ crore)"]
by_region