7

HELP YOUR DATA TELL ITS STORY

Tips for more effective visualizations

In today’s data-driven world, how you show and share your data is critical in determining its overall impact and effectiveness. But just putting your data into charts and graphs isn’t enough. Effective data visualization takes skill and storytelling, an eye for detail, and an understanding of your audience. Ready to learn the 7 secrets that can make the difference between great visualizations and just another set of charts? Let’s dive in.

1

Start with the story you want to tell. Whatever your specific goals may be, when you create a visualization, you’re really trying to capture a snapshot of your business – a quick, visual story that will drive understanding, discussion, and decision-making. To do this, you need to understand the data you’re working with and determine what story you want to tell. Start by asking a few questions.

UNDERSTAND YOUR DATA

• How many data sets are you working with and can they be combined? • Is your data qualitative or quantitative? • Is there external data you want to include? • How accurate or fresh is your data?

DETERMINE YOUR STORY

• What are the key messages you want to convey? • What do you want your audience to think, do, or feel? • What details will lend credence and support to your story?

WHAT STORY DO YOU WANT TO TELL? Good visualizations highlight relationships between data which can tell a powerful story. Outliers illuminate deviations from the norm. Trends indicate changes over time. Patterns show repeated, consistent characteristics. Correlations communicate relationships between two or more variables.

2 Variable width chart

Visualizations Bar chart

Bar chart Circular area chart Line chart According toTableQlik Visualization Advocate Dr. Patrik Lundblad, there are four fundamental Line chart Variable width chart or tables with Bar chart horizontal charts Visualizations types of dataembedded visualizations, each of which can provide specific insights: Bar chart

Table or tables with embedded charts

Bar chart horizontal

COMPARISON VISUALIZATIONS

Table Table or or tables tables with with embedded embedded charts charts

ANSWERS Which products sell best? How big is our market share? QUESTIONS Stacked 100% Stacked 100% bar chart area chart LIKE Line Line chart chart

Stacked 100% bar chart

Bar Bar chart chart horizontal horizontal

Composition How are our sales compared What percentage of the to last year?

Scatter plotbudget is going toward Visualizations bubble size

x, y, and z?

RELATIONSHIP VISUALIZATIONS

How many customers exist Is there a correlation between Waterfall chart across a demographic range? advertising spend and sales? Scatter plot Variable width chart

What would you How many days late are What’s Distribution the variation between Relationship customer payments? like to show?expenses and revenue

Line Stacked chart Source: ©A. Abela, 2010. www.ExtremePresentation.com Stacked bar chart

Stacked Circular area chart area chart

Stacked 100% bar chart area chart

Line chart

Bar chart horizontal

Bar chart

Stacked Stacked 100% 100% area area chart chart

Pie chart Line chart

Distribution

Stacked bar chart

Circular area chart

Composition Comparison

Scatter plot

Stacked 100% bar chart Pie chartBar Line chart Composition Bar histogram histogram

S

Tree map

w/subcomponents

Line chart

Stacked 100% area chart Stacked 100% bar chart Tree map Line histogram

Waterfall chart

Line histogram

Rela

w/subcomponents

Scatter plot bubble size Line chart

Scatter plot Stacked area chart Scatter plot Bar histogram

Pie chart

Stacked 100% bar chart w/subcomponents

Stacked 100% bar chart

Waterfall Waterfall chart chart

Source: ©A. Abela, 2010. www.ExtremePresentation.com

Bar histogram

BarLin cha

Scatter plot

What would you Relationship Visualizations like to show?

ource: ©A. Abela, 2010. www.ExtremePresentation.com

Table or tables with embedded charts

across regions?

Comparison Table or tables with embedded charts

Ba

Waterfall chart

Stacked 100% areaStacked chart Barbar chart chart

EXAMPLES

ons

00% 00% rt rt

COMPOSITION Comparison DISTRIBUTION VISUALIZATIONS VISUALIZATIONS

Line

Bar histogram

Linechart histogram Line Bar histogram Variable width chart Table or tables with Bar chart horizontal What would youIdentifying commonalities Identifying the highest and Seeing theembedded relationship Showing correlations charts Relationship Distribution Visualizations between data like pointsto thatshow? and outliers and clusters Scatter plot lowest values Comparison change over time bubble size Seeing current versus Seeing the general shape of Identifying outliers older values Identifying the relative a range Bar chart Circular area chart Line chart Scatter plot Bar chart Circular area chart Line chart Line histogram difference between parts of Composition What would you Spotting trends a whole Relationship Distribution Scatter like toplotshow? Comparison

Scatter plot bubble size

Variable width chart

Circular area chart

Line chart

FREQUENTLY USED FOR

Scatter plot

ar area chart

Line chart

Visualizations Choose the right visualization.

Scatter plot

hh chart chart

Circular area chart

Line histogram

Tree m

Stack are

3

Use color – carefully. Color can be a powerful aid to your visualizations, giving clear cues about relative value, points of emphasis, and differentiation between data sets. Choose your colors carefully, though – and more important, sparingly – because color used improperly can actually cause more confusion than it clears up.

DO: USE DISTINCT COLORS FOR EACH SEGMENT OF A PIE CHART

DO: USE DIFFERENT SHADES OF THE SAME COLOR TO DESIGNATE NUMEROUS VISUAL ELEMENTS.

DO: USE THE SAME COLOR FOR EACH BAR OF A BAR CHART.

DON’T: COLOR SEGMENTS USING DIFFERENT SHADES OF THE SAME COLOR.

DON’T: USE DIFFERENT COLORS WHEN DEALING WITH NUMEROUS GRAPHICAL ELEMENTS.

DON’T: USE A DIFFERENT COLOR FOR EACH BAR.

4

Keep it simple. With each visualization, ask yourself, “Which elements here are most important to the story I’m trying to tell?” Trying to cram too much into the same visualization only adds confusion and makes it harder to spot the insights. Making simple visualizations doesn’t have to be complex: • Limit the number of visualizations in a dashboard to 9 or less • Add callouts to emphasize the information that’s most critical to your message • Use trend lines to highlight important correlations between variables, and make your graphs more scannable • Use size and color to clearly define unique data variables • Avoid decorative fonts, text treatments like underlines and italics, and visual embellishments like drop shadows

WHOA – SLOW DOWN THERE We get it, it’s easy to get excited about visualizations. But it’s possible to get a little too fancy, with too many bells and whistles. In fact, there’s even a name for the phenomenon: Chartjunk. Here are a few of the most common chartjunk offenders:

5

Make it look good. Design matters. So use a few basic design principles in your visualizations to add clarity and avoid confusion. For example:

DO: ORDER SLICES FROM LARGEST TO SMALLEST

DO: USE TRANSPARENT COLORS

DO: USE 2D LINES

so each element remains visible

to clearly convey priority and ranking

DON’T: USE SOLID COLOR WITH OVERLAPPING DATA

DON’T: USE 3D LINES

for easier comparison

DON’T: RANDOMLY ORDER THE SLICES

6

Show your data at the right scale. Improper scale can make major insights seem mediocre and minor deviations feel massive.

300 300 200 200

300

To correctly represent the story your data is showing: • Make sure the scale of your axes accurately reflects the size of the data you’re visualizing • Choose the axes that will best represent trends in your data. Will absolute numbers or percentages make your data clearer? • Don’t use cropped axes. Always start the Y-axis at 0

DO: START Y-AXIS VALUE AT 0

150 150 100 100

200

150 100

50 50

300 50 0 200

00

150 380 100 260

380 380

50 140 DON’T: TRUNCATE THE SCALE

0

PRO TIP: Vertical type can be difficult to read, so use horizontal labels whenever possible.

260 260

120

100 380

140 140

50 260

140

120 120 JAN

FEB

MAR

FEB

100 100 50 50

120

100 50

7

Create visualizations with your audience in mind. In the end, it all comes to down to knowing what your audience is actually interested in – and giving them what they want. Even the clearest, bestlooking visualizations can still be duds if they’re not helpful or interesting to the people you’re trying to reach. Ensure you’re delivering a persuasive story by: • Highlighting and ordering your information based on your audience’s interests • Visualizing your data in an intuitive, conventional manner to facilitate understanding • Testing your visualizations with a few trusted colleagues to ensure the story they’re seeing is the one you wanted to tell

When done well, a good visualization transforms messy, massive data sets into discussions, understanding, and well-informed decisions.

By creating visualizations that don’t just present data, but tell a clear, compelling story, you can explore, explain, and express critical information and make intelligent decisions that can have a big impact on your business.

See the whole story that lives within your data.

qlik.com © 2016 QlikTech International AB. All rights reserved. Qlik,® Qlik Sense,® QlikView,® QlikTech,® Qlik Cloud,® Qlik DataMarket,® Qlik Analytics Platform,® Qlik NPrinting,™ Qlik Connectors™ and the QlikTech logos are trademarks of QlikTech International AB, which have been registered in multiple countries. Other marks and logos mentioned herein are trademarks or registered trademarks of their respective owners.