How to use tableau?

Let's pretend the boss gave you a spreadsheet containing lots of

sales data. She asks you to investigate how to improve sales and profits.
She’s puzzled that sales are increasing but profits seem to be stagnant. What do we
do?
Let’s open Tableau, and start exploring data. We will use visualization as our primary
tool for
data exploration. We will visualize the company’s sales through the years.
This opening screen you see is called the Start Page.
Here on the left is the Connect pane, which provides various ways to connect to your
data.
We’ll be using the sample dataset that comes installed with Tableau. In the Connect
pane, find
Microsoft Excel, click that and then navigate to the path I’m showing you right now
/Documents/My
Tableau Repository/Datasources/ etc. etc. on a MAC. In Windows, I’m showing you the
path
right now, the rest of it is the same replace but don’t forget to change the slashes to
backslashes.
Now click on “Sample – Superstore.xls”. This dataset is an Excel file containing 4 years
of
sales data from a fictitious company called Superstore.
Okay, if you’re having trouble finding the repository,
just use this URL. You can download the file from here, it’s being flashed on the screen
right now.
This link is also available in the References file by the way so you can click on that link.
Bear in mind that the Superstore dataset is constantly being updated
by Tableau. For this course, we’ll need to use the standard dataset in the course
download folder.
If you use the different file, you might get or show different results in your visualization.
After choosing the Excel file that we need,
the Data Source screen opens, showing us the contents of the file, in a familiar table
format.
The left panel here, shows the Sheets in the Excel files and represents them as Tables
there are
3 of them here, and here, and here, and the table that contains our Sales data is
Orders.
Drag Orders from the Sheets section to the right side where it says “Drag a table” over
here.
This will tell Tableau we want to use the Orders sheet as a data source.
If you make a mistake, click the Undo arrow here or press Cmd-Z in a Mac
for Windows that’s going to be Ctrl-Z.
This other arrow over here is the Redo button, which is Shift-Cmd-Z in Mac and Ctrl-Y
in Windows.
The bottom half of the Data Source window now displays what’s inside Orders,
showing individual order records. If you’re familiar with databases,
you can drag another table here and use SQL commands to join them. But that’s
beyond
the scope of this course, so I’ll leave it up to you to investigate in your own time.
Start of transcript. Skip to the end.
Let’s click the Sheet 1 tab, right beside Data Source over here.
This opens up the Tableau Workspace. Then we’ll do a quick tour of the Tableau
user interface to make you familiar with it. This horizontal bar over here is what we call
the Toolbar. It contains many commands in an icon format.
Now this upper section of the Workspace contains two shelves called Columns and
Rows over here.
There’s also the Pages, Filters, and Marks shelves beside them.
Think of shelves like real-world shelves because you can place items in and out of
them.
The data from our Orders table shows up at this left panel here, and we call that the
Data pane.
You can drag, let’s say Category, into the Columns shelf like this one here.
Let’s just undo this now so we could create our first view.
Our first view will show Superstore sales plotted through 4 years of operations.
We’ll start by grabbing our column headers from the Data Pane.
Find Order Date right here and drag it now to the Columns shelf. Just follow me. You
will see that
Tableau automatically puts column headers using the Order Date aggregated per year.
Over here, the “Abc” is a placeholder signaling that is waiting for actual data to be
plugged
in. Let’s fill it with Sales figures. The Sales data is here at the bottom of
the Data pane. It’s preceded by a green pound sign, known today as a hashtag of
course.
Drag Sales to the Rows shelf and now we have our first line graph!
It's that easy to do it here.
This line graph seems to show that sales are increasing after a minor plateau in 2017.
So, I wonder why the boss said profits aren’t doing so well. Let’s investigate this later.
Tableau refers to visualization as a view, a visual, or a viz - "V, I, Z" for short. The term
“view”
implies that though we may filter and slice and dice the data, the original dataset stays
intact.
Save your Tableau workbook by clicking File > Save or the Save icon.
Choose the default folder, which is in Documents > My Tableau Repository >
Workbooks.
I’m showing the path right now. For Windows, you will have your own
repository almost similar to this one. Let’s name the file “Superstore Viz”.
Okay, now tableau saves this with the extension .twb, short for Tableau Workbook.
The .twb format saves only the Tableau visuals and they are linked to your original data
source.
If you want to share a Tableau file this way, then you must distribute the original data
source
along with the .twb file otherwise they’re not gonna see anything at all, your viewers.
If you want to embed the dataset into the file, save it as .twbx format,
and then you can share the .twbx file without attaching the original spreadsheet.
Close the workbook by clicking File and then Close, and then Quit Tableau by clicking
Tableau > Quit Tableau.
See you in the next lesson!
Hello again. Let’s open the workbook we saved. In the Tableau Start Page, your
Workbook should show up here, and you can click that to open it, or we could click File
> Open
and choose your Superstore Viz.twb file I’d like us to spend a few minutes on
understanding some basic concepts in Tableau and data analysis in general. This is
important for exploring the
various visualization options later on. It also helps us deepen our understanding of data
and
visualization to become better data detectives. I’m now showing the Data screen.
You can get here by clicking the Data Source tab on the lower-left corner of
Workspace.
A good visualization will help create insights about the data by allowing users to see
the
data from various perspectives – also known as dimensions. Data points usually come
in the form
of Measures done at certain points in time -- sometimes, from a specific location too. In
the
Superstore dataset, these Measurements are found in these columns - Sales, Quantity,
and Profit.
Notice that their value contains numbers or numeric data. Another important thing to
note
is that numeric data can be aggregated into totals, averages, ratios, etc.
Unlike Measures, Dimensions usually contain non-numeric data like Product
Categories,
Customer Types, and Location. We call these nominal data, from the Latin root word
nomen, which means names, to contrast them from numbers.
Word of caution, sometimes numbers can actually be nominal data. For example, ID
numbers like
those in school identification cards – you can total these numbers or take their average,
but it’s not gonna make any sense. Years are also numbers, but they are
actually nominal data. You can get the total of 2019 + 2020 + 2021, but it’s not
meaningful to us.
Our job as visual data analysts is to find trends in the data points.
We can do this by exploring the Measures from various perspectives or Dimensions.
In Superstore, possible Dimensions are Time, represented by Quarters per Year.
Another
Dimension to investigate is the Product Category and Sub-Category, and the
customer’s Location,
which we could get from the address details, which include Country and Region
information.
Notice also that Tableau tries to guess which parts of your file
are Measures and which ones are Dimensions, and it automatically sorts them in the
Data pane.
Let’s click the Sheet 1 tab to return to Workspace.
Look here at the Data pane. These items preceded by blue icons
are the Dimensions, while these with green hashes are Measures.
Under Measures, we see Discount, Profit, Quantity, and Sales.
Under Dimensions, we see Customer, which contains the Customer Name and the
Market Segment;
Location, which we will discuss in the lesson on maps;
and Product, containing Category, Sub-Category etc.
Tableau puts Dimensions at the upper part of the Data pane because they usually
become
column headers and the usual workflow is to pick a dimension and then find
the measures that could give us insight. It’s important to discuss Measures and
Dimensions to master the power of visualization, and we will tackle that in the next
video.
You can read more about Measures and Dimensions in this link:
https://help.tableau.com/current/pro/desktop/en-us/datafields_typesandroles.htm
which is also available in your reference file.
Save your workbook and close it. See you at the next lesson.
Start of transcript. Skip to the end.
Let’s explore the power features of Tableau for creating views.
Open your workbook and go back to Sheet 1. On the upper right corner of the
Workspace, you will find
the "Show Me" panel. Clicking this, Tableau drops down a panel that tells us what
charts we could
create using the data in the current view. At the bottom of the panel, we see the chart
type – horizontal bars for example -- and the minimum requirements to build that visual.
We need at least 0 Measure and 1 Dimension. It’s also hinting that we could add more
Dimensions
and Measures for a richer view of our data. Clicking any of the 2 Maps, Tableau tells us
that
we need at least 1 geo Dimension to construct a map. We will tackle maps later this
week.
Given the view we have right now, these charts are possible according to show me:
text tables,
heat maps, highlight tables, pie charts, and other visualizations.
Click each of them and see how the visualization both changes.
Don’t forget to save your data. See you in the next lesson!
Start of transcript. Skip to the end.
Let us change our line graph to a bar graph
and see more of what Tableau can do. Click Horizontal Bars in Show Me.
This view shows us the amount of Sales (Columns) per year. Our data set actually
contains more detailed day-to-day sales but Tableau automatically aggregated it by
year.
Here, I’m showing you how deeply we could disaggregate the data by clicking this one
Now, let’s sort the Sales figures from highest to lowest, represented by these blue bars.
Click the sort button over here on top of the Columns shelf.
This button will sort the data in descending order
while this button over here on its left will sort it in ascending order. This view will now
present an immediate answer to the question, what are the top 3 years in terms of
sales?
One thing people like about data visualization is that they usually pack a lot of data and
present it in a
meaningful way. The visualization helps reduce clutter in huge data sets. Tableau
makes your
views interactive by default too. That means you can investigate the dataset further by
mousing over
and clicking on the charts themselves. Tableau also automatically makes filters
available.
OK Let’s try some of the automated interactive features that can make
richer visualizations in Tableau. In Dimensions,
look for Segment then drag that into the Rows shelf here. The table now updates
and it will show us a breakdown of the different customers segments right here.
Let’s undo that by clicking the undo button of course, and make a more interesting and
compact bar graph.
Let’s go back and drag Segment again, but this time we drag it directly into the viz
and drop it on top of the horizontal bars. Make sure you drag it directly on top of the
bars
and not here or here. Check if you got a different visualization and undo and try again.
You may pause the video if you need to do a comparison.
Now we’ve got a colorful stacked bar chart conveying more information – it
takes up less space but provides more information for our investigation.
Mouse over each color to see more information contained in each color segment.
Pulling back, we could compare the growth of each segment through the years.
Notice in this section of the workspace, there’s a shelf called Marks with cards called
Color,
Size, etc. Below the Marks shelf, the Segment pill shows up with the icon
corresponding to Color.
This tells us that the color code in the bars are based on Segment. Dragging this
pill out of the Marks card will remove the colors in our graph. At this point,
feel free to pause the video and play around with your visualization. You can always hit
the undo
button or revert to your last saved version. So, another way to make the Segment colors
show in our bar graph is to drag Segment into the Color command inside Marks. There
are many
ways to do things in Tableau, and you could explore these as part of your further study.
Tableau automatically builds some interactivity in its charts.
Let’s click the 2019 bar to see details.
You could see details about Segment, Year, and the Sales value. We could drill down
on the
data behind the bar by clicking the View Data icon over here in the upper right of the
context menu.
The popup screen shows two tabs: Summary and Full Data. Let’s click the Full Data
tab.
This is going to show you the individual records of the Consumer segment for 2019.
We could make the Sales value show up in the chart itself by dragging of course the
sales pill from
the Data pane and dropping it in the Label card here. Now the values show up inside
the bars.
But the blue color for the Consumer segment makes it hard to read the label. Let’s
change that.
Right-click on the Consumer segment of any year -- I’m going to use 2019 – and then
select Format on
the context menu that pops up. Check out the left panel – it’s changed into the Format
pane now.
The item we want to change is the Font color, so make sure you click on the A
inside the square at the top of the Format menu. The menu should indicate here
Format Font.
And look for Pane: and click the dropdown menu here on its right. Change the font
color
by clicking this white square just to the right of the black color square and see
the font color in the labels change to white. Let’s close the Format pane – click the X
icon
on the upper right corner of the Format pane. Save your work and proceed to the next
lesson.