Está en la página 1de 40

Making Data Beautiful for

Business Users
Tips for Creating Rich, Visual, and Actionable
User Experiences
David Stodder
Director of Research for Business Intelligence
TDWI
September 23, 2014

Sponsor

Speakers

David Stodder
Research Director,
Business Intelligence,
TDWI

Allen Bonde
VP, Product Marketing
and Innovation,
Actuate

Agenda
Visualization: The age of beautiful data
Meeting the goal of improving decisions
Visualization and business intelligence
Meeting diverse user requirements

Best practices for visual data interaction and discovery


Where dashboards are headed amid technology changes

Visualization options: many choices


Reducing the noise and increasing understanding

Recommendations

Visualization: Seeing What Hides in Data


Graphics reveal data.
Indeed graphics can be
more precise and revealing
than conventional statistical
computations. Edward
Tufte

Geolocation of cell phone reports of dengue fever outbreak in


Lahore, Pakistan. Credit: MIT Technology Review

The 20 billion or so neurons


of the brain devoted to
analyzing visual information
provide a pattern-finding
mechanism that is a
fundamental component in
much of our cognitive
activity. Colin Ware

Credit: Stephen von Worley, www.datapointed.net

Confluence of Science and Practice


Human powers: We
perceive meaningful
patterns, structures, and
outliers in what we see
Scientific focus: How we
respond to graphical stimuli;
how we use memory to
process information
Common discourse:
Media employs visual data
representation to explain
current events

Credit: www.adinstruments.com

Cognitive Perspective on Visualization


Humans & Data:
Visualization is the interface
between the human visual
system that finds patterns and Collaboration: People
makes decisions and the
communicating with
powers of data computation
computer visualizations
are much more cognitively
Making sense of the data
powerful; thinking occurs
tsunami: One of the greatest
through interaction
benefits of data visualization
between individuals, using
is the sheer quantity of
cognitive tools, and
information that can be
operating within social
rapidly interpreted if it is
networks. (Ware)
presented well. (C. Ware)
7

Enabling New Perspectives and Analysis


See what is unforeseen:
Visualization allows
perception of emergent
properties that were not
anticipated. The perception
of a pattern can be basis of
new insight. (Ware)

Exposing quality issues:


Visualization often
enables problems with the
data to become
immediately apparent; it
reveals things not only
about the data itself, but
about the way it is
collected. (Ware)
Data relationships: How
are data objects linked or
related? Visualization can
make these more apparent

Visualizations: Some Easy, Some Hard


Visualization as a sensory
language
Understanding pictures
without learning
Symbols that are
hardwired into the brain;
understood across cultures

Visualization as a learned
language; learning curve
Diagrams: symbols based
on social interaction
Semiotics : about how
symbols convey meaning
What is meaningful to one
is nonsense to another
Can be hard to learn, easy
to forget

Visualization: Goal is Decision Making


J.J. Gibsons Affordance
Theory: That we perceive in
order to act, to operate on the
environment; goal-directed
Top-down view of
visualization: We do not
perceive points of light; we
perceive possibilities for action

Pattern perception:
dividing regions into
simple patterns (e.g.,
same colors);
detecting motion

Performing operations: math


ops, merging, inverting,
transforming, splitting into
components, etc.
Forming new kinds of data

10

Multiple Objectives Coming Together


Calibrating to
the users
experience &
knowledge

Presenting
available data
accurately

To gain and keep


the users attention

Context:
Addressing
the users
role

Inspiring
good
decisions
and actions

11

Visualization and Business Intelligence


Key to self-service trend:
Visualization is essential to
making BI easier for
nontechnical users to consume
and interact with data
Shortening path to insight:
What BI has always been
about; visualization accelerates
users progress
Business agility: Viz key to
enabling data-driven decisions
But: Success depends on data;
bad data = bad visualizations

12

Key ROI Focus: Better Operational Efficiency


Organizations want to use visualization to reduce time to
insight for all types of users in many different scenarios

From Data Visualization and Discovery for Better Business Decisions, TDWI Best Practices Report, Third Quarter 2013

13

Most Important Visualization Objectives

14

Visualization: Diverse User Requirements


View of visualization usage patterns in three key areas

Source: Data Visualization and Discovery for Better Business Decisions, TDWI Best Practices Report, Third Quarter 2013

15

Visualization & Display/Snapshot Reporting


Snapshots: Scheduled
rather than requested ad hoc;
users want to personalize
based on roles
Visualizations must be accurate
and consistent

KPIs and scorecards:


Orienting users toward goals
and objectives
Can users or developers make
the look more exciting using
fun visuals?

Drill-down flexibility: Critical

Source: Claimcare.net

16

Operational Alerting: Avoiding Fatigue


Situations that demand
immediate attention: watch
out for alert fatigue
Using color, size, animation,
etc., flexible visualization can
help users prioritize and
recognize sources
Spotting trends and anomalies
in event data streams

Time is of the essence:


Real (or near real) time vital
Mobile devices: form factor
a visualization concern

Credit: www.catchpoint.com

17

Visual Data Discovery and Analysis


Fusion: Analytics, test-andlearn data exploration, and
advanced computation
matched with visualization
A visual path: data
interaction through filtering,
comparing, and correlating
visual data relationships
Business data laboratory:
Enabling exploration of who,
what, when, why behind
events and transactions
18

Visual Discovery Best Practices


Guidance is necessary:
Self-service and freedom
are important, but most
users need guidance
A blank slate with too
many visual options can be
intimidating

Metadata matters:
Common models,
hierarchies, dependency
mapping, etc., enable
users relate different data
sources and metrics

Big data access is often


important: Viz helps users
cope with the data tsunami
Yet, limits to how much data
can really be shown mean
that aggregation, chart
selection, and well-designed
dashboards are critical

Scale and performance:


Ensure data infrastructure
support
In-memory and cloud are
potential options

19

Good Visual Data Interaction: Imperative


Too much is too much:
Drowning users with data even
if presented well is useless if
users cant do what they need to
Goal: Get users beyond putting
pretty pictures on numbers and
toward more immersive data
experiences
Critical: Drill down, slice and
dice, filtering, sorting and
modifying data, customizing
report items

Sanctum, Rogue Pictures, 2011

20

Encouraging Storytelling, New Forms


of Collaboration
Using visualization to narrate the
data story being told
Most visualization stories begin with
some kind of question that orients
the viewer to the topic and context
within which the data is most
meaningful. Steele and Iliinsky
What data are we looking at?
In what time frame does the data
exist?
What notable events or variables
influenced the data?

21

Dashboards: Bringing It All Together


Visual, role-based view of
actionable information
Nexus of self-service BI and
analytics
Performance mgmt: visibility
via access to data-driven,
outcomes-oriented metrics
Integration at the glass:
internal and external data,
metrics, content
Many types of dashboards
and often many dashboards

Source: www.nrel.gov

Credit: www.triadtechpartners.com

22

Old and New Visions of Dashboards


First-Gen Dashboards
Tabular reports with few
and only simple charts
Limited number, variety of
data sources
Limited methods of
finding, interacting w/data
Dependent on IT
developers to create and
modify
Tied to single tool or
application

Where They Are Going


Libraries of chart types;
drag-and-drop selection
Role-based, single view of
data from multiple sources
Integration of search,
advanced analytics tools
Self-service creation,
preferably managed or
guided by IT expertise
Integrated view; seamless
experience on mobile
23

Visualization Options: Many Choices

3D visualization example
Scatterplot example
Heat map example

Bullet graph example. Credit Stephen


Few, www.perceptualedge.com

Gauges and dials example. Credit:


www.mockuptiger.com

Sparklines example. Credit: www.edwardtufte.com

24

Choosing Visualizations: Best Practices


Avoid clutter; no eye
candy
Consider the audience:
executive? A team?
Pay attention to
context; emphasize
what matters
Aim for relevance; dont
mislead or confuse
Step beyond
convention but do so
with purpose

Source: Data Visualization and Discovery for Better Business Decisions, TDWI Best Practices
Report, Third Quarter 2013

25

Recommendations
Gauge where users are right now with data interaction
Are they using spreadsheets? Enterprise BI reporting? What role
does data play currently in their decision processes?
Assess the learning curve: are visualizations more sensory or
learned?

Make visualization part of effort to achieve BI/analytics


democratization
Visual can help nontechnical users who struggle to interact
effectively with data especially big data
Tools alone dont make it easy; remember context, role, purpose

Match users requirements to match the right capabilities


Display, snapshot reporting, or scorecards?
Operational alerting?
Visual data discovery and analysis?

26

Closing Recommendations
Increase data interactivity with broader visualization
options and functionality
But rather than give users a blank slate, ensure that they have
guidance, either through software or from IT developers

Data provisioning: Ensure you have a scalability and data


performance strategy for visual discovery and analytics
Consider expansion to the standard BI/DW architecture, such as
in-memory computing and cloud options

Dashboard clarity and relevance: Employ data


visualization to reduce (not increase) confusion and
clutter and increase speed to insight
Match functionality with users decision processes; allow for
personalization and customization; aim for single view
rather than dashboards for each app or (mobile) platform
27

Thank You!

David Stodder
Director of Research for BI
dstodder@tdwi.org
(415) 859-9933

For a visual to truly be beautiful, it


must go beyond merely being a
conduit for information and offer
some novelty.
When done beautifully, successful
visualizations are deceptive in their
simplicity, offering the viewer
insight and new understanding at a
glance. J. Steele and N. Iliinsky
Beautiful Visualization: Looking
at Data through the Eyes of
Experts, J. Steele and N.
Iliinsky, OReilly Media, 2010

28

Making Data Beautifuland Actionable


Allen Bonde

VP Product Marketing & Innovation, Actuate

29
29
Actuate
Actuate Corporation
Corporation
2014
2014

Computers like
data

people like
answers!
30
Actuate Corporation 2014

30

What do you want to visualize?

31
Actuate Corporation 2014

31

Insights for everyone, everywhere


32
Actuate Corporation 2014

32

Supporting all channels requires SCALE

CLOUD ARCHITECTURE
Interactive
Viewer

Metrics
Manager

Dashboard

BIRT
Studio

BIRT iHub

RDBMS, NoSQL/NewSQL, Hadoop, Cloud, Social Media, Enterprise


Applications, Document Archives, Print Streams, Data Warehouses
33
Actuate Corporation 2014

33

Follow the Crowd


(open is better)

34
Actuate Corporation 2014

34

WHITE LABEL is key to embedding


One of the main
reasons why we chose
Actuatewas the API
structure and the
architecture. Being able
to customize the
workflows, being able to
customize the UI, and
how that was exposed
to our customers was
very, very important.

BIRT Embedded in OEM


and SaaS offerings

and other partners


35
Actuate Corporation 2014

35

Are you being?

helpful
relevant
engaging

Your userswill decide!


36
Actuate
Corporation
2014
Actuate
Corporation
2014

36

Learn more: www.actuate.com


Follow me: Twitter.com/abonde

37
37
Actuate
Actuate Corporation
Corporation
2014
2014

Questions?

38

Contact Information
If you have further questions or comments:
David Stodder, TDWI
dstodder@tdwi.org

Allen Bonde, Actuate


abonde@actuate.com

39

Learn More!
TDWI Big Data Analytics Solution Summit
Big Data Analytics for Customer Insight & Engagement
Scottsdale, AZ | November 2-4, 2014
*
TDWI World Conference
Emerging Technologies 2015
Orlando, FL | December 7-12, 2014
See: http://www.tdwi.org/events

40