Scribd
Upload
34 views10 pages

Visual Analytics - Process

The document outlines the process for creating effective data visualizations. It discusses analyzing the available data to determine quality, variables of interest, and appropriate chart types. It recommends sketching visualization ideas on paper before building prototypes in tools like Tableau, PowerBI, or Qlik. An iterative process of sketching, getting feedback, and revising is suggested. The document provides tips for ensuring visualizations clearly communicate the intended message before focusing on design aesthetics. It also discusses best practices for complex visualizations with multiple charts or interactivity.

Uploaded by

jmlusu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
34 views10 pages

Visual Analytics - Process

The document outlines the process for creating effective data visualizations. It discusses analyzing the available data to determine quality, variables of interest, and appropriate chart types. It recommends sketching visualization ideas on paper before building prototypes in tools like Tableau, PowerBI, or Qlik. An iterative process of sketching, getting feedback, and revising is suggested. The document provides tips for ensuring visualizations clearly communicate the intended message before focusing on design aesthetics. It also discusses best practices for complex visualizations with multiple charts or interactivity.

Uploaded by

jmlusu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 10

Visual Analytics

THE PROCESS FOR


DATA VISUALIZATIONS
May 1st, 2022
THE
PROCESS
WHAT DATA DO YOU HAVE?
Once you have a completed brief, look at your data, and answer some questions:

• Is the data quality good? You can’t build a good viz with bad data. Garbage in, garbage out.

• What data types do I have? Categorical, discrete, continuous, string, boolean, integer, number, date… the data
types you have will limit the types of charts you can build. For more on this, look at the Chart and Graph Types on
slide 9.

• What are the variables* I want to center my visualization on? Often, the temptation with a large dataset
is to use as many variables in your viz as possible. But this can lead to cluttered, confused viz’s.
Ideally, choose a one/two central variables in your data, and look at it from different angles.
• For example, if viewing sales data by city, the variable “city” may be your central variable, and you can
analyze that variable by “sales”, “profit”, “discount given”, etc.

You should now be ready to start building!

* Think of each column in your data as a variable. e.g. “store name”, “customer name”, “sales”, “profit”, “date ordered”, “city”…
SKETCH FIRST…
Once you have your brief, and you know what data to work with, you can start building.

The best tool for building your first charts? Pen + paper.

A suggested approach:
• Sketch out your first viz ideas, trying to build charts that deliver a clear message.
• After your first sketch, show it to someone. See if your message comes across clearly, or if it needs work.
Do not worry about design – just focus on nailing the message.
• Iterate – draw new sketches, show them to people, get feedback, draw new sketches…

Continue this iteration until your sketches deliver the right message. Then, you’re ready to prototype.

Sketch Show Prototype…


…AND THEN BUILD A PROTOTYPE
Once you have a sketch you’re happy with, you can begin to build the actual viz. There are many tools
available, although Tableau, PowerBI and Qlik are the most popular.

Again, adopt an iterative approach. Build, show, get feedback, build again. This time, focus on two things:
• Does it clearly deliver the right message?
• Is it beautiful? For help with the useful links in slide 10.

The number of iterations you run through will depend on how much time you have. But, by the end, you should have
a polished, ready-to-go, effective visualization!

Prototype Show Finished viz!


TIPS FOR
BUILDING
GOOD VIZ’S
MESSAGE FIRST, DESIGN AFTER
When building your viz, first make sure your data visualisation clearly delivers the
message/understanding you want it to.

Only then should you focus on making it beautiful.

Think of your data viz as a building. The strength of the message is the building’s foundations, and the design is the
building’s decoration. A building with shaky foundations will still fall, no matter how pretty it is.

Message first, Decorate after.


WHEN BUILDING COMPLEX VIZ’S
If you’re building a visualization with that’s complicated (multiple charts, interactivity, etc.), these rules may
come in handy:

• Don’t just add in every chart possible. Resist the temptation to use every possible chart type in a viz. Use
only what clearly delivers your message.

• Center your data visualization around a single focus. The best viz’s take a single variable and analyze it
from through different lenses, giving the viewer a deep understanding of a central theme. A single
complete picture is better than multiple incomplete pictures.

• “Overview first, then zoom and filter, and details on demand”*. With a complex viz, it’s usually best to
start a data viz at the high level, and then drill-down into more specific areas. If building in interactivity, allow users
to click/filter/highlight what they’re interested in, providing more detail when needed.

*
This is a popular mantra data viz enthusiasts (known as Shneiderman’s Mantra).
Useful Links
https://cfpb.github.io/design-system/guidelines/data-visualization-guidelines
https://eazybi.com/blog/data-visualization-and-chart-types
https://www.eea.europa.eu/data-and-maps/daviz/learn-more/chart-dos-and-donts
https://www.tableau.com
https://powerbi.microsoft.com/en-us/
https://docs.google.com/document/d/1pMeLLcOGyXSv6DKqe4UB_4SxfqNAsVw-ccTNWDnNo
bo/edit

You might also like