YOU ARE DOWNLOADING DOCUMENT

Please tick the box to continue:

Transcript
Page 1: Data Visualization Tips and Concepts

COPYRIGHT 2010 RESOURCE INTERACTIVE. PROPRIETARY AND CONFIDENTIAL.

Data Visualization and Dashboard Design

Columbus Web Analytics WednesdaySeptember 22, 2010

Presented by Tim [email protected]

Page 2: Data Visualization Tips and Concepts

2

Beautiful!

2,345

5,262

2,152

3,442

4,424

5,672

5,262

2,152

3,442

4,424

5,672

2,345

3,442

4,424

5,672

2,345

5,262

2,152

Artist Analyst

Page 3: Data Visualization Tips and Concepts

3

Stephen Few Edward Tufte

The Single Most UsefulData Visualization Concept

Data-Pixel Ratio*=

No. of non-white pixels devoted to data

Total non-white pixels

* Term coined by Stephen Few as a derivation of Edward Tufte’s “data-ink ratio”

Page 4: Data Visualization Tips and Concepts

4

Data-Pixel Ratio*=

No. of non-white pixels devoted to data

Total non-white pixels

* Term coined by Stephen Few as a derivation of Edward Tufte’s “data-ink ratio”

Page 5: Data Visualization Tips and Concepts

5

Page 6: Data Visualization Tips and Concepts

6

Adds no value

Tough to read at an angle

Overkill on the size

Redundant info(in this case)

Calibri is a hideous font (IMHO)

Redundant

Heavier than needed

Page 7: Data Visualization Tips and Concepts

7

Page 8: Data Visualization Tips and Concepts

8

X X X XX X X XX X XX XX X X

X X X

X

X X

Page 9: Data Visualization Tips and Concepts

9

Slightly over $150,000

Page 10: Data Visualization Tips and Concepts

10

Slightly over $150,000… but you have to work to figure that out

…and needlessly decreases the data-pixel ratio

Page 11: Data Visualization Tips and Concepts

11

I defy the laws of light!

…and needlessly decreases the data-pixel ratio

Page 12: Data Visualization Tips and Concepts

12

Leave 3D to the Movies

Source: Image captured from recovery.gov (long since removed, thankfully)

Page 13: Data Visualization Tips and Concepts

13

~10%of men in the U.S. have

some form of color blindness

(and a form of red-green color blindness accounts for over half of these)

Page 14: Data Visualization Tips and Concepts

14

On Screen and Color Printout

Black-and-White Printout

Red = Green

Page 15: Data Visualization Tips and Concepts

15

Page 16: Data Visualization Tips and Concepts

16

Page 17: Data Visualization Tips and Concepts

17

Pie Charts = EVIL!

• Rainbows Are Good in Princess Tales — Not in Data Visualization

• Labels, Labels, Labels

• Those Pesky Near-Zero Values

• Seeing Small Differences… and 1D vs. 2D Perception

• Economy (of Space) Is a Virtue

Page 18: Data Visualization Tips and Concepts

18

Source: http://www.math.yorku.ca/SCS/Gallery/

…the sum of the parts must be 100%

If you must use a pie chart…

Page 19: Data Visualization Tips and Concepts

19

Source: Laszlo Thoth, http://tongodeon.livejournal.com/583338.html

Page 20: Data Visualization Tips and Concepts

20

Page 21: Data Visualization Tips and Concepts

21

Mis…LEADDD…ing!!!

Page 22: Data Visualization Tips and Concepts

COPYRIGHT 2010 RESOURCE INTERACTIVE. PROPRIETARY AND CONFIDENTIAL.

Some Thoughts On Dashboards

Page 23: Data Visualization Tips and Concepts

23

Iconic Memory (The “visual sensory register”)

Preattentive processing of information

Short-term Memory(a.k.a. working memory)

Limited to 3 to 9 chunks of visual information

Long-term Memory

Source: Information Dashboard Design by Stephen Few and Brain Rules by John J. Medina

Page 24: Data Visualization Tips and Concepts

24

Dashboards: Less is More

• Aim for a single page/screen:

• The human struggles to process/compare data that is not in its field of vision (short-term memory limitations)

• Should provide “at a glance” information – it’s got to be efficient to consume if you expect people to use it

• Measures included must be tied to objectives and should be organized as such

• All web analytics vendors have pretty crappy built-in dashboards

Page 25: Data Visualization Tips and Concepts

25

Source: http://www.microstrategy.com/DashboardGallery/Dashboards/InteractiveSalesOverview.asp

Page 26: Data Visualization Tips and Concepts

26

Sparklines

• “Historical precision” often not necessary

• Still effectively enables spotting trends, spikes, and dips

• Economical use of space…which means more metrics can be digested/compared at once

Engagement - Increase the number of actively & positively engaged consumers

4-Month Month Prior MthPast 6 Months Trend Total Change

179%

Unsubscribes (posts hidden)

-14%

277%

Likes (Content)

Wall Posts

Comments

243

598

2,499 44%

1,355

Page 27: Data Visualization Tips and Concepts

27

Impressions are down 85% from a

year ago

Email opens and clickthroughs are

droppingDatabase growth

is declining

Page 28: Data Visualization Tips and Concepts

28

Page 29: Data Visualization Tips and Concepts

29

Awareness - Increase brand aw areness Advocacy - Increase the number of actively & positively engaged consumers

4-Week Week Prior WkPast 6 Weeks Trend Total Change

Acquisition - Grow the base of marketable consumers

Total Fans:

Net Increase Over Prior Week: -840

4-Week Week Prior Wk Trial - Increase the number of consumers w ho have been activated by the brand

Past 6 Weeks Trend Total Change

Actual Target Increasing > 15% Trending between -15.0% & 15% Decreasing below -15.0%

<brand> Facebook Page DashboardReport Period Ending August 14, 2010

8%

Shares

66%

-47%

Likes (Overall)

Wall Posts

Comments

173,107

13

245

1,556 114%

316

Facebook Page Traffic

Referrals from facebook.com

-89%

56 -2%

-10%

2

114901

Unique Page Views of Facebook Page 787

New visitor visits from facebook.com

-16%

0%

<brand> .com Referrals

Photo Views

364New Fans 61 -92%

Removed Fans (Unlikes) 5%

Video Plays

125136

157174 174 173

0

30

60

90

120

150

180

210

4-Jul 11-Jul 18-Jul 25-Jul 1-Aug 8-Aug

To

tal

Fan

s (0

00s)

-5

-3

-1

1

3

5

0

500

1,000

1,500

2,000

4-Jul 11-Jul 18-Jul 25-Jul 1-Aug 8-Aug

Total Interactions Sentiment Score

<brand>

Page 30: Data Visualization Tips and Concepts

30

Further Reading

• Information Dashboard Design by Stephen Few

• Brain Rules by John J. Medina

• Blogs:

• Peltier Tech Blog (http://peltiertech.com/WordPress/)

• Presentation Zen by Garr Reynolds (http://www.presentationzen.com/)

• Flowing Data by Nathan Yau (http://flowingdata.com/)

Page 31: Data Visualization Tips and Concepts

Related Documents