User Experience Architecture For Discovery Applications

Joe Lamantia
User Experience Lead Enterprise Products
Endeca
User Experience (UX) jlamantia@endeca.com
Best Practices for
Latitude Discovery
Applications Mark Burrell, Ph.D.
VP Worldwide Lead, User Experience
mburrell@endeca.com


• How can you harness the power and flexibility of
Latitude to create useful, usable, and compelling
discovery applications for your business users?
User Experience (UX)
Best Practices for
Latitude Discovery • This session goes beyond the technology to
Applications explore how you can apply analytics best
practices, user interface design patterns, and the
fundamental principles of information design and
visualization to compose discovery applications
that optimize user discovery, success,
engagement, and adoption.


User Experience at Endeca

Who we are
• You may know us from Professional Services
• We’ve moved to Product Development
• We’ve expanded the team
We need you
• User panel for future research, design reviews, and evaluation
To join, fill out a brief survey at tinyurl.com/endecaux
• We’re also conducting interviews at the conference
Talk to Kirsten Robinson, Joe Lamantia, or Mark Burrell


Endeca for Enterprise Discovery
Enabling Business & People via Information Visibility & Discovery

5


? ?
?

5


INFORMATION VISIBILITY

? ?
?

5


DISCOVERY

? ?
?

Analyzing Evaluating

Exploring Monitoring

5


DISCOVERY

?

? ?

5

Discovery Shift: From BI to Discovery

From: To: 
 Search  Exploration, decision support and
discovery

 Content and data  Information assets, facets, &
relationships

 Static “reports”  Interactive-dynamic summaries,
dialogue & exploration

 Discrete features and functions  Integrated solutions and holistic
experiences

 One size fits all presentations  Personalized, adaptive solutions

 Solitary user  Collaboration and community

6

Identify Information Assets: Compose and Configure

?
? ?

7 © 2011 Endeca Technologies Inc. All Rights Reserved. CONFIDENTIAL AND PROPRIETARY - DO NOT REPRODUCE.

Agenda

Understanding User Needs
• Segmentation Models
• Modes of Discovery

Design Patterns for Discovery Applications
• Information Layout, Workflow and Interactivity
• Screen-level patterns, application structure patterns and component group patterns

Design Principles for Information Design & Visualization
• Fundamentals of Visual Perception
• Chart selection


Understanding User • Segmentation Models: End Users
Needs
• Scenarios and Modes of Discovery


Understanding Discovery: Key Questions

What are the business strategies & objectives?

Who are the critical users/people/communities?

What are the strategic user goals & user scenarios?

? What do people need to know and what key
questions do they need to answer?

What decision/discovery support and information
assets will help people achieve their goals?

How do people need to interact with information
assets to achieve their goals?

How do people need to interact and collaborate
with other people to achieve their goals?


Discovery Spectrum: Key Dimensions

User Types Goals & Modes of Discovery
Scenarios Discovery Assets

 Electronics Engineer  “Find part #35456…”  Locating  Products
 Purchasing Agent  “Discover compatible  Analyzing  Rich Media
 Novice Shopper parts …”  Evaluating  Textual Info
 Technical Enthusiast  “Understand part  Comparing  Relationships
 Etc. obsolescence …”  Verifying  Community
 etc.  Etc.  Etc.


User Profile: Strategic Leader Role
Description
• Sets the overall strategy and direction of an Primary Modes of Discovery:
organisation, by developing long-term plans and • Monitoring, Analyzing, Evaluating
capabilities to position the organization for future
Example Scenarios (SVP Sales)
success
• Monitor how well we are tracking to revenue and

Examples margin targets by division

• Director of Brand Image, Conoco-Phillips; SVP • Identify problem/success areas and where to

Private Sector, OTPP; General Mgr, Division, Auto intervene and reward

Co; SVP Sales, Beverage Maker

Key Questions
Goals
• Are we hitting or exceeding target revenue and
• Sets strategic direction & objectives for an
quantity and market share?
organisation to optimize success
• What is our market share profile across divisions?
• Develop long-term plans to achieve strategic
Are we trending in the right direction?
objectives.
• Which brands/categories are underperforming?
• Monitor achievement of strategic objectives,
• Are our strategic categories performing well or not?
progress wrt market trends and the overall
(e.g. healthy lifecycle products increasing or
organisational context and direction.
decreasing? By how much?) Are we hitting targets?
• Identify strategic problems & opportunities.


Enterprise Discovery Workers: End-Users

Strategic Leader:
End Users
Analyst:
• Set strategic direction & objectives for • Monitor, measure, & inform others
an org to optimize success re: org functioning in re: key metrics &
• Develop long-term strategic plans value chain
• Monitor achievement of strategic • Generate insights and
objectives, trends/directional progress, recommendations to help optimize
org context Strategic strategy & management
• Identify strategic problems & Leader • Advise or answer questions for
opportunities Strategic Leaders & Mgrs

Designer:
Managers: Operations Product/ Relationship • Design-create activities or artifacts
• Plan & manage execution of strategic Manager Service Manager that specify execution of processes,
plans -- processes, people, products- Manager support relationships, or define
services , & relationships -- to optimize products, services
outcomes/value
• Monitor detailed execution,
diagnose issues , identify
opportunities; inform Strategic
Leaders
Rep:
• Monitor-manage-support
relationships with specific customers-
Analyst Designer Rep partners to optimize value


Modes are a tool for designing discovery solutions that

• leverage what is common about people’s discovery needs
and activities (human cognition and perception)

• allowing for what varies (information structures, frame of
reference [business domain]


Discovery “Modes”

“a broad, but identiﬁable literary method, mood, or
manner, that is not tied exclusively to a particular
form or genre.
[Some] examples are the satiric mode, the ironic,
the comic, the pastoral, and the didactic. (CB)”

http://en.wikipedia.org/wiki/Mode_(literature)


Rhetorical Modes
Exposition
The purpose of exposition (or expository writing) is to explain and analyze information by presenting an idea, relevant evidence, and
appropriate discussion.

Argumentation
The purpose of argumentation (also called persuasive writing) is to prove the validity of an idea, or point of view, by presenting sound
reasoning, discussion, and argument that thoroughly convince the reader. Persuasive writing is a type of argumentation with the additional
aim to urge the reader to take some form of action.

Description
The purpose of description is to re-create, invent, or visually present a person, place, event, or action so that the reader can picture that
which is being described. Descriptive writing can be found in the other rhetorical modes.

Narration
The purpose of narration is to tell a story or narrate an event or series of events. This writing mode frequently uses the tools of descriptive
writing. Narration is an especially useful tool for sequencing or putting details and information into some kind of logical order, usually
chronological.

http://en.wikipedia.org/wiki/Rhetorical_modes

Identifying Modes

“Understand the quality performance of a part and module set in manufacturing
and the field so that I can determine if I should replace that part.” (Engineering)

“Understand a lead's underlying positions so that I can assess the quality of the
investment opportunity.” (Portfolio Manager)

“Understand a portfolio's exposures to assess portfolio-level investment
mix.” (Portfolio Manager)

“I need to understand the cost drivers for this commodity so I can negotiate better
terms with my suppliers and forecast business risk based on market
indices.” (Procurement)

Mode: ‘Comprehend’ (understand)

Modes of Discovery: Lookup

Locating
To find a specific (possibly known) item
e.g. I need to find a new part with particular technical attributes
and then source it from the most qualified supplier -
Engineering

Verifying
To confirm or substantiate that an item or set of items meets some specific criterion
e.g. How can I determine if I am looking at the latest information for a part or supplier? - Supply Chain
Specialist

Monitoring
To maintain awareness of the status of an item or data set for purposes of
management or control
e.g. I need to monitor at risk/failing customers/dealers so I can prompt my Account Reps to fix the problems -
Sales Manager


Modes of Discovery: Learn

Comparing
To examine two or more items to identify similarities & differences
e.g. I need to compare our module set teardowns with competitive teardown information to see if we’re staying
competitive for cost, quality and functionality - Engineering

Comprehending
To generate insight by understanding the nature or meaning of an item or data set
e.g. I need to analyze and understand consumer-customer-market trends to inform brand strategy &
communications plan – Director, Brand Image

Exploring
To proactively investigate or examine an item or data set for the purpose of
serendipitous knowledge discovery
e.g. I need to understand the cost drivers for this commodity so I can negotiate better terms with my suppliers and
forecast business risk based on market indices - Procurement


Modes of Discovery: Investigate

Analyzing
To critically examine the detail of an item or data set to identify patterns &
relationships
e.g. I need to know the cost drivers for a part such as materials that impact cost. Is the relationship a
correlation or step function for a part cost driver? - Engineering

Evaluating
To use judgement to determine the significance or value of an item or data set with
respect to a specific benchmark or model
e.g. I need to determine my current state in my prints so I can evaluate if I have price variation to negotiate a
better price - Procurement

Synthesizing
To generate or communicate insight by integrating diverse inputs to create a novel
artifact or composite view
e.g. I need to prepare a weekly report for my boss (sales mgr) of how things are going - Account Rep


Comparison–driven Optimization

1. Replace a problematic part
Analyze
(from sourcing, cost or technical
perspective)

2. with an equivalent or better part Compare

3. without compromising quality
Evaluate
and cost.



1. Analyze
Analyze

2. and understand gaps between
current cost of commodity Compare

3. versus best in class Evaluate
manufacturing costs.


Analyze Compare Evaluate

Core Engineer: Identify parts used for same function as candidates for commonization and complexity
reduction
Engineering: Replace a problematic part (from sourcing, cost or technical perspective) with an equivalent
or better part without compromising quality and cost.
Engineering: Compare our module set teardowns with competitive teardown information to see if we’re
staying competitive for cost, quality and functionality.
Portfolio Manager: Compare a lead's performance claims with relevant benchmarks to assess the lead's
claims
Procurement: See the difference between what we are spending and what we should be spending to
maximize savings (between actual PO and should costs).
Cost Estimators: Analyze & understand gaps between current costs of commodity versus best in class
manufacturing costs


Exploration-driven Optimization

Explore Analyze Evaluate

Core Engineer: Identify opportunities to optimize use of tooling capacity for my commodity/parts
District Manager: Identify sales opportunities and targets (increased key customer market share across
categories/brands; upsell-cross sell; promotional targets
Category Manager: Evaluate & optimize our product portfolio: Which products should we de-list and
retire? What new products should we be making/selling?
Brand Manager: Identify the best customer/consumer/region targets for our brand/products
Core Buyer: Determine suppliers to use for parts in my program and execute sourcing agreements
Program Administrator: Identify customers/marketers/dealers failing & at risk of de-branding based on
performance problems


Strategic Insight

Analyze Comprehend Evaluate

Engineering: Track module cost versus functionality over time to determine trends.
Engineering: Understand the quality performance of a part and module set in manufacturing and the field
so that I can determine if I should replace that part.
Portfolio Manager: Understand a lead's underlying positions so that I can assess the quality of the
investment opportunity
Portfolio Manager: Understand a portfolio's exposures to assess portfolio-level investment mix
Procurement: I need to understand the cost drivers for this commodity so I can negotiate better terms
with my suppliers and forecast business risk based on market indices.


Strategic Oversight

Monitor Analyze Evaluate

SVP Sales: Monitor how well we are tracking to revenue and margin targets by division
Supervisor/Inspector: Monitor and grade incoming incidents; close incidents, add incident close codes
Core Engineer: Monitor global commodity use in relation to plan/guidelines to identify gaps that require
corrective action
District Manager: Monitor how well we are tracking to revenue and margin targets by division
Brand Manager: Monitor & evaluate how our brand is performing in re: revenue, margin, and market
share targets
Financial Analyst: Monitor & assess commodity status against strategy/plan/target


Comparison–driven Synthesis (~4%)

Analyze Compare Synthesize

Director, Brand Image: Analyze and understand consumer-customer-market trends to inform brand
strategy & communications plan
Engineering: Find out how many parts I have in my module set of parts and find ways to reduce cost
across them
Core Buyer: Formulate scope & strategy for sourcing and gap closure
Brand Image Analyst: Analyze and understand a market: marketer network, competitive position,
customer sat, & share, etc. to inform brand strategy and communications plan


Mode Networks

Verify

Analyze Comprehend Synthesize

Compare Evaluate

Monitor Analyze

Explore


Modes and User Roles

Monitor Comprehend

Strategic
Leader

Explore Evaluate

Product-
Service
Manager Comprehend
Compare
Operations Relationship
Manager Manager

Designer Analyst Rep


Common Collaborations

Strategic
Leader

Product-
Service
Manager

Manager Manager

Compare
Verify

Explore Analyze Evaluate


Strategic
Leader

Product-
Service
Manager

Manager Manager




Strategic
Leader

Synthesize

Product-
Service
Manager
Synthesize
Manager Manager Synthesize
Synthesize



Design for Effective • Application Structures
Discovery
• Information Layout, Workflow and Interactivity
Applications
• UI Design Pattern Library


Design Principles for Discovery Solutions

Match screen composition to discovery needs
• No one size fits all

Assemble screens using common combinations of components
• Show ‘relatedness’ by grouping components visually

Assemble discovery applications using several types of screens

Support user groups by creating different applications


Page 2 Page 3

Screen Structure: Page 4.A | Page 4.B

Metrics Bar
ʻMulti-purposeʼ Search
View 1 View 2 View 3
Purpose
Breadcrumb
Supports exploration, comparison
and visualization. Results Table

Modes
Locating
Verifying Guided Chart Chart
Navigation
Evaluating
Analyzing
Exploring

Components Cross Tab

Range Filter
Breadcrumb
Guided Navigation
Metrics Bar Record Detail
Results Table


Data
Source
Application Structure: B
Classic

Users Multi-
purpose
Analyst / Designer roles
Manager roles

Multi- Multi-
purpose purpose
Screen types
Multi-purpose

Data Data
Source Source
A C


Data
Source
Application Structure: B
Classic
Analyze
Users Multi-
purpose
Manager roles

Multi- Multi-
purpose purpose
Screen types
Multi-purpose Evaluate Compare
Data Data
Source Source
A C


Discovery Solution Trends

Larger volumes of data
• Including unstructured data
Growing discovery application structures
• Depth, breadth
Varied discovery needs
• Summation vs. detail, time frames, domains
Navigation flow between pages
• E.g. ‘Drilling for oil’, U-shaped navigation
Large communities of discovery workers
Diverse Perspectives


Screen Structure:
Status & Alerts
Dashboard
Page 1 Page 2 Page 3
Purpose
Present global KPIs
Metrics Bar Metrics Bar
All business overview
Summary of one domain / several Metrics Bar Metrics Bar Alerts
domains / processes

Metrics Bar

Metrics Bar Chart Alerts
Modes
Metrics Bar
Monitoring
Exploring

Components Cross Tab
Metrics Bar
Alerts
Chart
Cross Tab


Screen Structure:
Visualization
Dashboard
Page 1 Page 2 Page 3
Purpose
Present global KPIs
Metrics Bar Metrics Bar Metrics Bar
All business overview
Summary of one domain / several Chart Chart Chart
domains / processes

Modes Chart Chart Chart
Monitoring
Exploring

Components
Metrics Bar Cross Tab
Chart
Cross Tab


Application Structure: Data
Hub and Spoke Source
C Metric Metric Metric
Data Data Data
A B C
Users
Analyst / Designer roles Multi-
Manager roles purpose

Screen types
Dashboard
Multi-purpose

Multi- Multi-
purpose purpose

Data Data
Source Source
A B


Application Structure: Data
Hub and Spoke Source
C Metric Metric Metric
Data Data Data
A B C
Users Analyze
Analyst / Designer roles Multi-
Manager roles purpose

Monitor
Screen types
Dashboard
Multi-purpose

Multi- Multi-
purpose purpose

Evaluate Compare
Data Data
Source Source
A B


Design Pattern: Analytics Dashboard

• Identify the primary user segments that will
be served by the dashboard and develop an
appropriate strategy for dealing with
variation among those users.
• Identify the data and metrics that are most
relevant to the priority user segment(s).
Organize the content and dashboard
elements into meaningful groups so that
connections between related elements are
immediately apparent.
• Draw attention to exceptions or important
changes so that the user can gain
immediate visibility into deviations from
expected or desired values.
• Provide clear indicators of the default time
frames applied for varied dashboard
elements and the freshness/recency of the
information (e.g. date/time of updates).

http://patterns.endeca.com/content/library/en/home/patterns/analytics_dashboard.html


Endeca User Interface Design Pattern Library
A growing set of user interface design patterns providing guidance and
best practices for creating great search, faceted exploration, and
Discovery Solutions

http://patterns.endeca.com

44

Screen Structure: Page 1 Page 2 Page 3

Visual Discovery Page 4.A | Page 4.B

Metrics Bar
2 Axes Search
Purpose
Breadcrumb Metrics Bar
Insight into single core process, org
unit, aspect of ecosystem, global kpi.

Modes
Chart Chart
Comparing
Analyzing Guided
Navigation
Verifying

Components
Breadcrumb
Results Table
Guided Navigation
Range Filter
Metrics Bar
Chart
Cross Tab



Visual Discovery Page 4.A | Page 4.B

Metrics Bar
4 Axes Search
Purpose
Breadcrumb
Breadcrumb Metrics Bar
Visual insight into single core
process, org unit, aspect of
ecosystem, global kpi.
Chart Chart

Modes
Analysis Guided
Navigation
Evaluation
Verifying Chart Chart

Exploration

Components
Breadcrumb Range Filter
Cross Tab
Guided Navigation
Metrics Bar
Chart
Cross Tab


Application Structure: Br

Differentiated
Metric Metric Metric
Purpose
Provide differentiated discovery
capability for summarization vs. Visualization
detailed engagement with data. 2 axes

Users
Dashboard
Manager roles

Screen types Visualization Visualization
2 axes 2 axes
Multi-purpose


Application Structure: Br

Differentiated
Metric Metric Metric
Purpose
Provide differentiated discovery Analyze
capability for summarization vs. Visualization
detailed engagement with data. 2 axes

Users Monitor
Dashboard
Manager roles
Strategic Leader roles
Evaluate
Visualization
Compare
Visualization
2 axes 2 axes
Screen types
Dashboard
Visualization
Multi-purpose


Design Pattern: Analytics Applications

• Provide a clearly identifiable “Home” or “launch” page that supports the initiation of
common discovery tasks and acts as a hub for ongoing discovery work. The
optimal design of the “Home” or “launch” page will depend on its purpose and
context, but the following are common types...
• Provide links to past discovery work (e.g., “bookmarked” views, results sets,
saved searches etc.) to enable resumption or recall of past analyses.
• Allow the user to independently control the visibility of selected elements of the
display, i.e. the visualizations, results table, faceted navigation, etc.
http://patterns.endeca.com/content/library/en/home/patterns/analyticsApplications.html


Unstructured Data
Search
Discovery
Purpose Breadcrumb Results List
Understand unstructured data

Modes
Comprehending
Exploring Compare

Synthesis
Guided
Navigation Record Detail Find Similar

Components
Search
Breadcrumb
Record Detail
Guided Navigation
Results List
Compare
Record Detail


Application Structure:
Multi-perspective
Visualization
Users 2 axes

Manager roles
Strategic Leader
Visualization
Dashboard
2 axes

Screen types
Visualization
Dashboard Unstructured
2 axes
Visual 2 axes
Visual 4 axes
Unstructured
Visualization Visualization
4-axes 4-axes


Evaluate
Multi-perspective Analyze
Visualization
Users 2 axes

Manager roles Synthesize
Strategic Leader
Visualization
Dashboard
Monitor
2 axes

Screen types Explore
Visualization
Dashboard Unstructured
2 axes
Visual 2 axes
Visual 4 axes Comprehend
Unstructured
4-axes 4-axes
Compare
Comprehend Verify


Comprehensive
Multi-
Visualization purpose
Users 2 axes
Manager roles
Strategic Leader
Visualization
Dashboard
2 axes

Screen types
Dashboard Visualization
Unstructured
2 axes
Visual 2 axes
Visual 4 axes
Unstructured
4-axes 4-axes


Explore
Comprehensive
Purpose Multi-
Visualization purpose
2 axes

Users Monitor
Analyst / Designer roles Visualization
Dashboard
Manager roles 2 axes

Strategic Leader Compare
Visualization
Unstructured
2 axes

Screen types
Comprehend
Dashboard
Analyze
Visual 2 axes Visualization Visualization
Visual 4 axes 4-axes 4-axes

Unstructured Evaluate


Screen Structure: Application 1 Application 2

Discovery Hub Metrics Bar

Metrics Bar Alerts
Purpose
Metrics Bar
Hub connects linked apps & Syndicated Content

communities
Chart Chart
Modes
Exploring
Monitoring Application 3
Metrics Bar
Components
Metrics Bar
Syndicated Content
Syndicated Content
Chart
Alerts
Bookmarks
Bookmarks


Discovery Environment Visualization
2 axes

Purpose
Integrate multiple discovery
Visualization
Dashboard
2 axes

applications with overlapping but Visualization
2 axes
Unstructured
Visualization
2 axes

distinct communities of users Visualization Visualization
4-axes 4-axes Visualization
Dashboard
2 axes

Visualization
Unstructured
2 axes

Users Visualization
4-axes
Visualization
4-axes

Analyst / Designer ? Rep roles
Manager roles
Strategic Leaders

Visualization
2 axes

Screen Types Visualization
2 axes
Dashboard

Unstructured
2 axes

Unstructured

Dashboard (variants)
4-axes 4-axes

Discovery Hub


Discovery Environment Visualization
2 axes

Purpose
Integrate multiple discvoery
Visualization
Dashboard
2 axes

applications with overlapping but Visualization
2 axes
Unstructured
Visualization
2 axes

distinct communities of users Visualization Visualization
4-axes 4-axes Visualization
Dashboard
2 axes

Visualization
Unstructured
2 axes

Users Visualization
4-axes
Visualization
4-axes

Analyst / Designer ? Rep roles
Manager roles
Strategic Leaders
Synthesize
Visualization
2 axes

Screen Types
Explore
Visualization
Dashboard
2 axes

Unstructured
2 axes

Unstructured

Dashboard (variants)
4-axes 4-axes

Discovery Hub Monitor


Implementation Tips • Suggestions “from the field”


Implementation Tips

Layout/Configuration of components should be consistent within a community.

Filters First. When you find yourself making lots of pages and tabs, first ask
whether you are providing a different visualization or a different slice of data. If the
latter – can navigation be used instead?

Drop-Downs Second. When you find yourself making lots of pages and tabs, look
for commonality across the visualizations. If only a few dimensions/metrics change
– can drop-downs be used instead?

Tabs Group Data and Visualizations. The right way to use tabs is when the
visualization is too dense or big to fit on the screen.


Key Design Guidelines - Consistency

• Use of “listen to data source” should
be consistent

• All numeric fields should have a
formatter
• Traffic lights should be used
consistently
• Metric and dimension naming
conventions should be consistent and
user-friendly
• All dimensions should have a
dimension group
• Dimensions displayed should be
consistent in pages/tabs that share a • Unlike OLAP – all Endeca analytics
datasource are based on dynamic record sets
• High cardinality dimensions should • Showing users the records that are
leverage dimension search and driving analytics helps them
more… link appreciate exploratory capabilities
• Multi-or dimensions are typically
preferred by users
• Is a dimension is expanded – this
should be done consistently


If analytics statements are the same except for 1-3 dimensions/metrics, consider
whether the statement can be parameterized – with varying dimensions/metrics
exposed as options in drop-down box.

If analytics statements are the same (same metrics/dimensions) except for the
WHERE clause – the WHERE clause typically represents a rule-oriented
classification of the underlying records.

Can this WHERE clause be converted into an additional attribute of the record?
(Classify records based on these rules during data ingest.) If so – tabs can often be
eliminated.

Highlight these new dimensions in the application so users are aware of them.


The correct use of tabs is to offer different views of the same data.

When you see scroll-bars in the browser – it is a good signal that tabs may be
needed.

If there is >1 visualization on the page, ask yourself is user need to correlate
information between those visualizations. If so, they should be on the same page.
If not, separate them into different tabs – less is more when using screen real
estate.


Dimension overload. If there are lots of dimension in the GuidedNavigation menu –
consider using precedence rules. Are there groups of those dimensions that make
sense to display after user has drilled into a certain category of data?
Additionally – remember that the “lookahead” dimension search in the Searchbox
component spans all dimensions in the application.
Grouping by month dimension. The default sort order for any textual dimension is
alphabetic. This is typically not desired for Month dimensions. Manually create/
order these dimension values in the DeveloperStudio.
Sub-pages versus outer-tabs. Generally speaking, sub-pages should be avoided.
The user needs to hover over page in order to learn that sub-pages exists. Outer
tabs are always visible. If need to link to page/content – then sub-page may be only
option.


Record Type. Are the users going to understand this concept? If no – exclude from
navigation component.

Search. This is the most differentiated feature of Latitude. Leverage it – there
should be a search box on every page. Record search and compound dimension
search are very powerful features. They are also novel features from user
standpoint – provide them with examples of valuable search strings.

DEFINE statements aren’t free. The MDEX Engine still needs to do work, even if
the results aren’t returned. Make sure that DEFINEs are needed for your analytics
statement.

Toggle edit controls. This should be disabled for end-users. It adds a lot of visual
clutter to the application.


Design Principles for • Fundamentals of Visual Perception
Information Design &
• Chart Selection
Visualization


Visual Perception

Cognitive processing
• Iconic memory (“graphics buffer”)
• Short-term memory (“RAM”)
• Long term memory (“hard disk”)

Stephen Few, “Tapping the Power of Visual Perception”


Visual Perception

Perception is tuned to detect certain visual primitives

Jock McKinlay, “Designing Great Visualizations”

Preattentive Attributes

Stephen Few, “Tapping the Power
of Visual Perception”

Which can be used to encode quantitative information?

Accuracy of Perception

Jock McKinlay, “Designing Great
Visualizations”

What might this mean for design?

The Ubiquitous Pie Chart

Jorge Camoes, “Eye Candy
and the Pie”

Which is biggest – Item 3, 4 or 5?
• Why is this difficult to interpret?
• How might it be improved?


Short term memory

Limited to 3-7 chunks of data at a time

Stephen Few, “Tapping the Power of Visual Perception”


Information Design and Visualization Principles

1. Use 2-D position and the line length to encode quantitative
values
2. Use lines graphs to show patterns formed by a set of values
across a continuous range
3. Don't try to communicate relationships between multiple values
by showing one value at a time.
4. Multiple graphs shown together are often a better solution than a
single graph, especially when several variables are involved.
5. Different views of the same data set give insight into
relationships that can’t be seen from one perspective only.
6. Allow the user to filter out irrelevant data and focus on their key
questions.

Stephen Few, “Information Visualization, Design, and the Arts”


Chart Selection

Andrew Abela, “Choosing a
Good Chart”


Chart Selection

What/Who How much Where When How Why

Dan Roam, “Back of the Napkin”


Chart Selection

What How much Where When How Why

30% 35% 13% 13% 2% 7%

bullet 7 5% Bar 60 23% heat map 20 29% Line 22 29% Flowchart 3 0% Bubble 15 0%

thermometer 7 5% pixel bar 7 3% region map 15 22% Sparkline 19 25%
number 8 6% column 60 23% pushpin map 20 29% Gantt 11 15%
colored shape 57 41% pixel column 6 2% Timeline 13 17%
schematic 7 10%
text 60 43% thermometer 6 2% Swimlanes 9 12%
organization chart 1 1%
brick 9 4% Cycle 0 0%
quadrant 0 0%
number grid 11 4% Progression 1 1%
connections map 4 6%
stacked bar 22 9%
venn diagram 1 1%
clustered bar 22 9%
stacked column 23 9%
clustered column 22 9%
variable width
column 1 0%
bullet 6 2%
pie 1 0%
concentric pie 0 0%


Chart Selection

Grid Map TimeLine Bubble

What How much Where When How Why

30% 35% 13% 13% 2% 7%

Flowchart 3 0% Bubble 15 0%
bullet 7 5% bar 60 23% heat map 20 29% Line 22 29%
thermometer 7 5% pixel bar 7 3% region map 15 22% Sparkline 19 25%
number 8 6% column 60 23% pushpin map 20 29% Gantt 11 15%
colored shape 57 41% pixel column 6 2% schematic 7 10% Timeline 13 17%
text 60 43% thermometer 6 2% organization Swimlanes 9 12%
brick 9 4% chart 1 1% Cycle 0 0%
number grid 11 4% quadrant 0 0% Progression 1 1%
stacked bar 22 9% connections
clustered bar 22 9% map 4 6%
stacked column 23 9% venn diagram 1 1%

clustered column 22 9%
variable width
column 1 0%
bullet 6 2%
pie 1 0%
concentric pie 0 0%

Design Patterns: Spatial Information Visualization

Point Location Map
• Define iconic “pins” that can clearly convey precise
spatial location along with relevant information about
each record http://patterns.endeca.com/content/library/en/home/patterns/
pointLocationMap.html

Region Map
• Define boundaries for the regions at an appropriate
level of granularity to align with the users’ discovery
scenarios, mental models, and expected level of
analysis http://patterns.endeca.com/content/library/en/home/patterns/regionMap.html

Heat Map
• Define a function by which record density values can be
mapped to “intensity” values that align with the users’
discovery scenarios, mental models, and expected level
of analysis
http://patterns.endeca.com/content/library/en/home/patterns/heatMap.html


References

Books
Information Dashboard Design, Stephen Few, O'Reilly Media 2006
The Back of the Napkin, Dan Roam, Penguin 2008
The Visual Display of Quantitative Information, Edward Tufte, Graphics Press 2001
Beautiful Evidence, Edward Tufte, Graphics Press 2006
Readings in Information Visualization: Using Vision to Think, Stuart Card, Jock
Mackinlay & Ben Schneiderman, Morgan Kaufmann 1999

Papers
The sensemaking process and leverage points for analyst technology as
identified through cognitive task analysis, Pirolli, P., & Card, S. (2005) https://
analysis.mitre.org/proceedings/Final_Papers_Files/206_Camera_Ready_Paper.pdf
Exploratory search: from finding to understanding, Gary Marchionini,
Communications of the ACM, Volume 49 Issue 4, April 2006
http://www.ischool.utexas.edu/~i385t-sw/readings/Marchionini-2006-Exploratory_Search.pdf


References

Papers (cont.)
Tapping the Power of Visual Perception, Stephen Few, September 4, 2004
http://www.perceptualedge.com/articles/ie/visual_perception.pdf
Designing Great Visualizations, Jock D. Mackinlay with Kevin Winslow, February 18,
2009
http://mkt.tableausoftware.com/downloads/designing-great-visualizations.pdf
Information Visualization, Design, and the Arts: Collision or Collaboration?
Stephen Few, Visual Business Intelligence Newsletter , January 2010
http://www.perceptualedge.com/articles/visual_business_intelligence/information_visualization_and_art.pdf

Save the Pies for Dessert, Stephen Few, Visual Business Intelligence Newsletter,
August 2007
http://www.perceptualedge.com/articles/08-21-07.pdf


References

Papers (cont.)
Creating Successful Portals With a Design Framework, Joe Lamantia, International
Journal of Web Portals, 2009
http://www.igi-global.com/bookstore/article.aspx?titleid=37471
The Building Blocks of Successful Portals, Joe Lamantia, Intranets Today, 2009
http://www.intranetstoday.com/Articles/Default2.aspx?ArticleID=6087
Enhancing the Portal Experience, Joe Lamantia, International Journal of Web
Portals, 2009
Containers and Connectors as Elements in a Portal Design Framework, Joe
Lamantia, International Journal of Web Portals, 2010
http://http://www.igi-global.com/bookstore/article.aspx?titleid=40319
Evolution of a Portal Suite, Joe Lamantia, International Journal of Web Portals, 2010


Endeca User Interface Design Pattern Library
A growing set of user interface design patterns providing guidance and
best practices for creating great search, faceted exploration, and
Discovery Solutions

http://patterns.endeca.com

81

User Experience at Endeca

Who we are
You may know us from Professional Services
We’ve moved to Product Development
We’ve expanded the team

We need you
User panel for future research, design reviews, and evaluation
To join, fill out a brief survey at tinyurl.com/endecaux
We’re also conducting interviews at the conference
Talk to Kirsten Robinson, Joe Lamantia, or Mark Burrell


Contacts:
Joe Lamantia
User Experience Lead Enterprise Products
Endeca
jlamantia@endeca.com

Q&A Mark Burrell, Ph.D.
VP Worldwide Lead, User Experience
mburrell@endeca.com


User Experience Architecture For Discovery Applications

More Related Content

Viewers also liked

Similar to User Experience Architecture For Discovery Applications

More from Joe Lamantia

Recently uploaded

User Experience Architecture For Discovery Applications

Editor's Notes