Upload
tony-russell-rose
View
109
Download
2
Tags:
Embed Size (px)
DESCRIPTION
Classic IR (information retrieval) is predicated on the notion of users searching for information in order to satisfy a particular “information need”. However, it is now accepted that much of what we recognize as search behaviour is often not informational per se. Broder (2002) has shown that the need underlying a given web search could in fact be navigational (e.g. to find a particular site) or transactional (e.g. through online shopping, social media, etc.). Similarly, Rose & Levinson (2004) have identified the consumption of online resources as a further common category of search behaviour.In this paper, we extend this work to the enterprise context, examining the needs and behaviours of individuals across a range of search and discovery scenarios within various types of enterprise. We present an initial taxonomy of “discovery modes”, and discuss some initial implications for the design of more effective search and discovery platforms and tools.
Citation preview
A Taxonomy of Enterprise Search
Tony Russell-Rose, UXLabsJoe Lamantia & Mark Burrell, Endeca
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 2
Agenda
The landscape of search The information-seeking process
Models and frameworks Search strategies and tactics
Patterns of search behaviour Mode networks
Design implications Conclusions
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 3
The Landscape of SearchUnderstanding the territory
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 4
The Classical View
Web Search• Multiple engines• Single source
Enterprise search• Single engine• Multiple sources
Key differences Rich link structure Redundancy Adversarial search
& spam User goals & tasks:
fact finding to exploratory search
5
James Joyce’s Ulysses?
Irish writer?PR6019.O9 U418 G42
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 6
Fact finding Discovery
Irish writer?James Joyce’s
Ulysses?
Search Tasks
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 7
Redefining ‘search’
Keyword search works when you know what you want, and how to ask for it Fails when you don’t know what you
want, or even what questions to ask
Part of a larger cycleinformation
retrieval
analysissense-making
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 8
The Information-Seeking ProcessModels and frameworks
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 9
The Information-Seeking Process
Standard Model Broder 2002,
Sutcliffe & Ennis 1998
Cognitive Model Norman 1988,
Marchioni 1989
Dynamic Model Bates 1989, O‘ Day
and Jeffries 1993
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 10
Search Strategies and Tactics (Bates, 1979)
•to keep a search on track and efficient
Monitoring
•for navigating through the file or link structure of the individual resources
Information structure
•to support the overall query or search specification)
Query reformulati
on
•to aid the selection of words or phrases in the current query
Term
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 11
Search Modes and Analysis Techniques (O’Day & Jeffries, 1983)
Search Modes• Monitoring a known topic or set of variables over time• Following a specific plan for information gathering• Exploring a topic in an undirected fashion
Analysis Techniques• Looking for trends or correlations• Making comparisons• Experimenting with different aggregations/scaling• Identifying critical subsets• Making assessments• Interpreting data to find meaning
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 12
A Taxonomy of Web Search (Broder 2002)
•The immediate intent is to reach a particular site
Navigational
•The intent is to acquire some information assumed to be present on one or more web pages
Informational
•The intent is to perform some web-mediated activity
Transactional
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 13
Search Activities (Marchionini 2005)
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 14
Diverse Perspectives
Marchionini, 2005• Lookup• Learn• Investigate
Morville, 2010• Quit• Narrow• Expand• Pearl-growing• Pogo-sticking• Thrashing
Spencer, 2006• Known-item• Exploratory• Don’t know
what you need to know
• Re-finding
Analysing Search Behaviours
Locating Verifying Monitoring Comparing
Comprehending Exploring
Scenarios
15UXLabs - User Experience Research and Design - www.uxlabs.co.uk
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 16
Search Modes: Lookup•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
Locating
•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
Verifying
•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
Monitoring
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 17
Search Modes: Learn•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
Comparing
•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
Comprehending
•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
Exploring
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 18
Search Modes: Investigate •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
Analyzing
•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
Evaluating
•To generate or communicate insight by integrating diverse inputs to create a novel artefact or composite view•e.g. I
need to prepare a weekly report for my boss (sales mgr) of how things are going - Account Rep
Synthesizing
Search Modes
Lookup• Locating• Verifying• Monitoring
Learn• Comparing• Comprehendi
ng• Exploring
Investigate• Analyzing• Evaluating• Synthesizing
19UXLabs - User Experience Research and Design - www.uxlabs.co.uk
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 20
Patterns of search behaviourMode chains and sequences
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 21
Comparison–driven Search
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
Cost Estimators: Analyze & understand gaps between current costs of commodity versus best in class manufacturing costs
EvaluatingComparingAnalyzing
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 22
Exploration-driven Search
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?
EvaluatingAnalyzingExploring
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 23
Strategic Insight
Engineering: Track module cost versus functionality over time to determine trends.
Portfolio Manager: Understand a lead's underlying positions so that I can assess the quality of the investment opportunity
Procurement: Understand the cost drivers for a commodity so I can negotiate better terms with my suppliers and forecast business risk based on market indices
EvaluatingComprehendingAnalyzing
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 24
Strategic Oversight
SVP Sales: Monitor how well we are tracking to revenue and margin targets by division
Core Engineer: Monitor global commodity use in relation to plan/guidelines to identify gaps that require corrective action
Financial Analyst: Monitor & assess commodity status against strategy/plan/target
EvaluatingAnalyzingMonitoring
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 25
Comparison–driven Synthesis
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
SynthesizingComparingAnalyzing
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 26
Mode networks
Analyzing Comprehending
Comparing
Verifying
Synthesizing
Evaluating
Monitoring Analyzing
Exploring
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 27
Design implicationsApplying the insights
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 28
Levels of Abstraction
Single element patterns• Comparison views (Comparing)• Contextual views (Analyzing)
Whole screen patterns• Dashboard (Monitoring, Synthesizing)• Visual discovery (Exploring, Evaluating,
Verifying)Application-level patterns• Differentiated (Analysing, Comparing,
Evaluating, Monitoring Strategic Oversight)
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 29
Design Principles
Map search modes to screen components Avoid “one size fits all”
Create views by combining components Communicate conceptual relationships through
Gestalt principles of organization (e.g. similarity, closure, proximity, etc.)
Compose applications by combining views
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 30
View: multi-purpose
Purpose Supports exploration,
comparison and visualization
Modes Locating Verifying Evaluating Analyzing Exploring
Components Breadcrumb Faceted Navigation Metrics Bar Results Table Chart Cross Tab
GuidedNavigation
Breadcrumb
Range Filter
Results Table
Chart Chart
Page 4.A | Page 4.B
Metrics Bar
Page 1 Page 2 Page 3
View 1 View 2 View 3Search
Cross Tab
View 1 View 2 View 3
Record Detail
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 31
View: status & alerts dashboard
Purpose Present global
overview / summary of key metrics
Modes Monitoring Exploring
Components Metrics Bar Alerts Chart Cross Tab
Metrics Bar Metrics Bar
Chart
Alerts
Cross Tab
Page 1
Metrics Bar
Metrics Bar
Metrics Bar
Metrics Bar
Page 2 Page 3
Metrics Bar
Metrics Bar
Metrics Bar
Alerts
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 32
View: unstructured data discovery
Purpose Explore unstructured data
Modes Comprehending Exploring Synthesizing
Components Search Breadcrumb Faceted Navigation Results List Compare Record Detail Find Similar
Compare
Record Detail
Search
GuidedNavigation
Breadcrumb Results List
Page 1 Page 2 Page 3
Record Detail Find Similar
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 33
Application structure: classic
DataSource
A
DataSource
B
DataSource
C
analyze
compareevaluate
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 34
Application structure: hub & spoke
DataSource
A
DataSource
B
DataSource
CMetrics Bar
Metrics Bar
Metrics BarData
CData
BData
A
analyze
compareevaluate
monitor
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 35
Application structure: comprehensive
Monitor
Explore
Analyze
Compare
Comprehend
Evaluate
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 36
ConclusionsSummary and future work
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 37
Summary & Future Work
A Taxonomy of Enterprise Search• A perspective for understanding information-seeking
behaviour• A framework for designing discovery applications
Search Modes• Combine academic and practitioner perspectives• Atomic units and composite patterns
Further Work• Larger data sample, more principled sampling strategy• Acquire data directly & consistently (from primary
sources)• Refine and validate process
UXLabs - User Experience Research and Design - www.uxlabs.co.uk 38
Thank you
Tony Russell-Rose, PhDDirector, UXLabs
Web: http://uxlabs.co.uk Email: [email protected] Blog: http://isquared.wordpress.com LinkedIn:
http://uk.linkedin.com/in/tonyrussellrose Twitter: @tonygrr