William Hudson - Agile User Experience and UCD (workshop)
Post on 18-Oct-2014
474 Views
Preview:
DESCRIPTION
Transcript
syntagm
Agile User Experience and UCD
William Hudson User Experience Strategist william.hudson@syntagm.co.uk
syntagm syntagm
Guerrilla UCD – The Series • Getting Started
– 0. Free Overview – 1. Guerrilla UCD Boot Camp
• Strategy Webinars – 2. Visual Design for Usability – 3. Navigation & Menu Design – 4. Designing for SEO & Accessibility – 5. Human Error, Messages & Feedback – 6. Usability Evaluation
• Tactics Webinars – 7. User-Centred IA with Card Sorting – 8. Dynamic Web Pages: Effective Use of Ajax – 9. Writing Effective Web & Intranet Content – 10. Designing for Advanced Users – 11. Persuasion, Trust and Seduction
William Hudson 2
syntagm syntagm
Guerrilla UCD – The Series • Lean & Agile Webinars
– 12. Making the Case for UCD in Agile – 13. Integrating UCD & Agile – 14. Agile UX: Users, Personas & Design Maps – 15. Agile UX: Use Cases, Stories & Scenarios – 16. Agile UX: Conceptual Models
• This course touches on many topics in GUCD
sessions 12-16 (each of those is 90 minutes)
• See www.guerrillaucd.com for more details
William Hudson 3
syntagm syntagm
Topics • Basic characteristics of Agile development • Primary differences between waterfall and Agile user
experience • Empathy gap, balanced teams and embedded user
experience roles • Empathetic design • Design decision styles and minimum viable products (MVPs) • Personas and the persona myths • User and persona stories • Integrating usability evaluation • Persuasive facilitation • Design maps
Agile UX & UCD 4
syntagm
Understanding Agile UCD
syntagm Agile UX & UCD 6
syntagm syntagm
The Need for User-centred Design • Experience has shown that we cannot establish
user requirements just by asking users what they need. Users… – are not designers – may not see the big picture – may not be aware of what it possible – may not be aware of what is changing – are not necessarily familiar with organizational needs
• Instead, what is required for interactive systems is a design approach focussed on users
7 Agile UX & UCD
syntagm syntagm
The Case for User-Centred Design • User-Centred Design aims to research and
understand the real needs of users and to produce designs that meet those needs, with the following benefits – Less frustration – Improved user experience – Better data quality – Fewer support calls – Lower development costs – Fewer legal concerns (Disability Discrimination Act) – Bridging the mass-market chasm...
8 Agile UX & UCD
syntagm syntagm
The Case for User-Centred Design
Innovators Early Adopters
Early Majority
Late Majority Laggards
The technology chasm
(Adapted from Geoffrey Moore’s Crossing the Chasm and Everett Roger’s Diffusion of Innovation)
9 Agile UX & UCD
syntagm syntagm
Key Features of User-Centred Design
• Direct engagement with users: observation, research and evaluation
• Investigation and understanding of contexts of use, for example – Point of sale system for a pub versus a supermarket – Returns system for a dusty, noisy warehouse rather
than a clean and comfortable office
• Main focus of UCD is suitability of solution for real users
Agile UX & UCD 10
syntagm syntagm Agile UX & UCD 11
syntagm syntagm
Origins of Agile
• Based on the working practices at the Lockheed Skunk Works™ (see bit.ly/agile-skunkworks)
• Small teams • Close quarters • Low process overheads • No ‘big design up front’ • Focus on team motivation and working code Agile UX & UCD 12
syntagm syntagm http://bit.ly/big-pizza
Small Teams?
syntagm
p communication paths for n nodes:
p = n∙(n – 1) / 2 (or p = (n2 – n) / 2)
Small Teams
Agile UX & UCD 14
0
20
40
60
80
100
120
140
160
180
200
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Path
s
Nodes
For a team of 5 there are 10 paths of communication
But there are 105 for a team of 15!
Agile UX & UCD 15
syntagm syntagm
The Agile Manifesto
• Individuals and interactions over processes and tools
• Working software over comprehensive documentation
• Customer collaboration over contract negotiation
• Responding to change over following a plan
Agile UX & UCD
(The manifesto and its related 12 principles can be found at www.agilemanifesto.org)
16
syntagm syntagm
The Agile Team
Customer / Owner / User Rep
Users ?
Agile UX & UCD 17
syntagm syntagm
The Agile-to-User Relationship • eXtreme Programming (XP) was the first new
approach to be called Agile • Various suggestions were made about how to
address user requirements – ‘Expert’ user on team (this was done in practice for
the Chrysler Comprehensive Compensation System, the first XP project, but one team member wrote of the burnout of this particular user)
– In Extreme Programming eXplained a product owner or customer would make all user-related decisions
Agile UX & UCD 18
syntagm syntagm
Typical Agile Approach (Scrum)
Agile UX & UCD
(www.mountaingoatsoftware.com)
19
(Image from www.buzzle.com/articles/waterfall-model-diagram.html)
Software development method proposed by Winston Royce – projects would pass through most phases at least twice
The origins of the ‘waterfall model’
Agile UX & UCD 20
Software development method as desired by project managers, coined ‘waterfall’
The origins of the ‘waterfall model’
Agile UX & UCD 21
Origins of Scrum
syntagm syntagm
Origins of Scrum • A 1986 article in the influential Harvard Business
Revue talked about product development in terms of Rugby and ‘moving the scrum downfield’ – The New New Product Development Game
• Keys to success were seen as – moving from sequential (waterfall) working to
overlapping (iterative) approaches – giving interdisciplinary (cross/multi-functional) teams
autonomy: this is the scrum
Agile UX & UCD 23
syntagm
The New New Product Development Game
Agile UX & UCD 24
syntagm
Overlapping phases promoted in the New New Product Development G (the authors were Japanese)
syntagm syntagm
Overlapping Phases
• Overlapping phases can be seen in the iterative software development approach in use since the 1960’s and which was becoming more popular during the 1990’s
Agile UX & UCD 26
Iterative development model from the Rational Unified Process, developed in the 1990’s (now owned by IBM)
syntagm syntagm
Scrum
• Scrum and Scrum-like approaches are now the most popular in Agile working
• But the Agile emphasis on ‘no big design up front’ has frustrated UX & UC design since they have traditionally been front-heavy
• ‘Up-front UXD’ is a natural, but unfortunate consequence
Agile UX & UCD 27
Up-front UXD as often implemented
Up-front UXD
UX
Agile Stuff
User Testing
Agile UX & UCD 28
syntagm syntagm
Shortcomings of UF-UXD • Assumption of ‘right-first-time’
– Prototyping, mock-ups and user testing can help to address this, but limited relative to live system
• Little or no engagement with Agile team – No learning (in either direction) – No appreciation for UX activities – No involvement of UX during development
• The UX step can look like waste from a lean perspective (especially if organization has low UX maturity – customer value needs to be proven)
Agile UX & UCD 29
syntagm syntagm
Activity 1
• Organize into small groups with someone to take notes
• Await further instructions!
Agile UX & UCD 30
syntagm
The Empathy Gap
syntagm syntagm
Scrum Roles
• Roles in Agile vary according to the approach, but in Scrum there are only three: – Scrum master – Product owner – Team
• Scrum master provides project management and oversight
• Product owner represents the stakeholders • Team does the work! Agile UX & UCD 32
syntagm syntagm
Product Owner Role
Product Owner
Stakeholders
33 Agile UX & UCD
Team Scrum Master
syntagm syntagm
The Agile-to-User Relationship
• Two main problems: 1. No one person in isolation can represent or fully
understand the needs of users • Consider this actual user quotation:
“What do you want to talk about, what we really do or what we’re supposed to do?”
2. Some team members can find it hard to
appreciate – and sometimes to understand – problems that users have (the empathy gap)
Agile UX & UCD 34
syntagm syntagm
The Empathy Gap: Empathizing/Systemizing Theory • Large-scale study of 450 IT workers (see
Hudson, William, 2009, Reduced Empathizing Skills Increase Challenges for User-Centered Design, CHI 2009 Conference, Boston and British HCI Group Conference, Cambridge)
Agile UX & UCD 35
syntagm
Empathizers
syntagm syntagm
Systemizers
RTFM Agile UX & UCD 38
(A symptom of the empathy gap)
syntagm syntagm
Technologists can have strong systemizing but reduced empathizing skills
Agile UX & UCD 39
syntagm syntagm Agile UX & UCD
(There is a small effect for the few female technologists, but no ‘crossover’)
40
syntagm syntagm
Men Women
Systemizing
Empathizing Empathizing
Systemizing
syntagm syntagm
Impediments to UX Maturity
• Low empathy • Technology focus • Cost concerns • Featuritis • Time scales • Narrow scope
Agile UX & UCD
“I don’t want it good, I want it Tuesday”
- Jack Warner
42
syntagm syntagm
Agile UCD Challenges
• UCD needs to be adapted to fit into an Agile framework – Fewer ‘up-front’ activities – Short cycles – Reactive (particularly in teams with low UX
maturity) • UCD practitioners also need to represent and
explain users’ needs effectively – Empathetic design – Persuasive facilitation
Agile UX & UCD 43
syntagm syntagm
Empathetic Design
• Research has found that empathy is related to specific components of the brain called mirror neurons
• They are activated when performing a task but also when seeing the same task performed by others
• To promote empathy we should rely on the adage ‘seeing is believing’
Agile UX & UCD 44
syntagm syntagm
Field Research
(Image from Copenhagen Institute of Interaction Design)
syntagm syntagm
Personas
Agile UX & UCD 46
syntagm syntagm
Usability Testing
Agile UX & UCD
Techsmith’s Morae (and similar) allow remote observation of usability tests
47
syntagm syntagm
Empathy-Assistive Technology
syntagm syntagm
Persuasive Facilitation
• Beware that empathetic techniques are not always going to be the most persuasive – Quantitative evidence will be more effective in
some cases – Persuasive facilitation requires knowing what
evidence to present to whom and when (more on this later)
Agile UX & UCD 49
syntagm syntagm
Activity 2
• Organize into small groups with someone to take notes
• Await further instructions!
Agile UX & UCD 50
syntagm
Integrating Agile and UXD
51
syntagm syntagm
UCD Versus Agile Methods • There are some fundamental differences in the nature
of traditional versus Agile methods • Traditional methods tend to be problem-oriented while
Agile methods are solution-oriented – Problem-oriented approaches to design focus on
understanding the problem in the abstract – Solution-oriented approaches focus on concrete solutions
Agile UX & UCD 52
Problem Solution Solution-oriented
approaches
Problem-oriented approaches
syntagm syntagm
The Problem-Solution Continuum
• In traditional Object-Oriented development much effort is devoted to software design and modeling
Agile UX & UCD 53
Traditional OO Development
Prob
lem
Solu
tion
Use Case Models
Object Models
Other UML Models Implementation
Functional Testing
syntagm syntagm
The Problem-Solution Continuum
• In Agile methods, the approach to design can be better described as ‘just in time’
Agile UX & UCD 54
Prob
lem
Solu
tion
Agile Development
User Stories
Design, Implementation &
Unit Testing
Functional Testing
This gap can spell trouble for users
syntagm syntagm
User Interface Design Issues • Focus is primarily on architectural modeling
and/or programming • Very little (if any) conceptual design – how the
service or product will be understood by users • No explicit user interface design • No specific user involvement • Evaluation is mostly in terms of functional testing
(does the system do what we said is was going to?)
• Usability is not a consideration
Agile UX & UCD 55
syntagm syntagm
Introducing User-Centred Design • The principles of ISO 9241-210 require:
– Investigation of the contexts of use – Active involvement of users – Multidisciplinary design – Prototyping – Evaluation of solutions
• For good user interface/interaction design we also need: – Early UI design – Focus (what the product or service is and is not) – A clear conceptual model, explained through the UI – Consistency – Empathy (developers → users)
Agile UX & UCD 56
syntagm syntagm
UCD Techniques vs Deliverables
Agile UX & UCD 57
User Research X X X X
Personas X X X X Paper Prototyping X X X X
Usability Evaluation X X X X
Conceptual & UI Design X X X
Style Guide X X
UCD Technique
Deliverables
ISO 9241-210 UI Design
syntagm syntagm
Development Activities
Agile UX & UCD 58
User Stories
Design, Implementation &
Unit Testing
Functional Testing
Prob
lem
Use Case Models
Object Models
Other UML Models Programming
Functional Testing
OO
Agile
User Research
Personas
Paper Prototyping
Conceptual & UI Design
(not to scale)
Usability Evaluation
Style Guide
Goal-oriented User Stories
UCD
Prob
lem
Solu
tion
Solu
tion
Traditionally, these activities are performed up front
syntagm syntagm
‘Up Front’ UX Approach User
Research
Personas
User Stories*
UX
Cycle 1 Dev Cycle n ...
Usability Testing
Agile UX & UCD
?
?
* Including wireframes and prototypes if used
59
syntagm syntagm
‘Up Front’ UX Approach
• The up front approach presents several challenges – Excessive lead time – Big design up front is not Agile – Potentially wasted developer resources – No developer engagement with UX
Agile UX & UCD 60
syntagm syntagm
Usability Testing
Agile UX Approach
Agile UX & UCD 61
0
0
1
UX
Dev ... Cycle 0
2
1 n 2 3
3
1 2
4
UX is always once cycle ahead and one cycle behind development
Pre-release
...
...
User Research
Scenarios, Wireframes & Prototypes
Personas & User Stories
syntagm syntagm
Agile UX Approach (Close Up)
Agile UX & UCD 62
1 2 3
3
1 UX
Dev
Functionality for usability testing
Designs for Implementation
Evaluation
Design
Implementation
syntagm syntagm
Agile UX Approach
• Cycle 0 much shorter than ‘up front’ UX since some personas and user stories developed later
• Developers involved with UX – Observers in user research – Collaborators in personas and user stories
• Embedded UX role within development team
Agile UX & UCD 63
syntagm
Balanced Teams
syntagm
Typical Agile Approach
Original diagram from Leffingwell, Dean (2011), Agile Software Requirements, Addison-Wesley
UI/UX
Agile UX & UCD 65
syntagm syntagm
Embedded UX Role • At least one full-time UX role should be included
in every team – Adapting/tailoring organizational UI/UX guidelines for
project at hand – Part of the inception/elaboration/construction
processes – Primary collaborator on user/persona stories – Responsible for usability evaluation (possibly through
central or external resource) – Focal point for UX questions/issues/training/events
(possibly through central or external resource)
Agile UX & UCD 66
syntagm
Central UX
Resource
Project 1
Project 2
Project 3
Project 4
Each project has at least one embedded UX resource
Agile UX & UCD 67
syntagm syntagm
Central UX Resource
• Shared by project teams, provides services to embedded UX roles: – User research – User recruitment – Usability evaluation – UX training & events
• Also responsible for – Organization style guides and UX standards – UX recruitment
Agile UX & UCD 68
syntagm syntagm
User Research • In UCD we are primarily interested in what real users
do and the contexts in which they do it: – Goals and tasks – Artefacts (things they work with) – Environment (physical and organizational)
• Primary means of recording is note-taking, but… – Photocopy or photograph all artefacts – Consider audio or audio/video
recording of observation sessions
– Get prior permission to take screenshots if necessary
(See Beyer and Holtzblatt’s Contextual Design)
69 Agile UX & UCD
syntagm syntagm
Real Users – Interviews and Focus Groups
• We are also interested in what users have to say but beware of potential problems
• People… – Describe what they should do rather than what they actually do – Omit important detail, especially where it has become second
nature or if they believe it is not relevant – Provide too much detail, obscuring larger issues – Are reluctant to raise unpopular issues – Sometimes want to promote controversy – Over-generalize – Focus too narrowly on the specific issues being raised by the
interviewer’s questions • In focus groups, group dynamics also play a factor
Agile UX & UCD 70
syntagm syntagm
Users in Design • We need to identify who we are designing for
(and who we will be doing usability testing with) • While we could build profiles, we also need to get
our team thinking like and identifying with users (empathy)
• We must also design for what users really need and do (descriptive models) rather than what we think they should need and do (normative models)
• Otherwise, our user stories are works of fiction
71 Agile UX & UCD
syntagm syntagm
User Research Outcomes
• Doing user research gives us data for... – Personas – Terminology – Conceptual models
Agile UX & UCD 72
syntagm syntagm
Conceptual Models Basket Account
Departments Search
Amazon.co.uk Homepage (c. 2002)
73 Agile UX & UCD
syntagm syntagm
Conceptual Models
Amazon.co.uk Homepage (c. 2009)
Basket Account
Departments
Search
74 Agile UX & UCD
syntagm syntagm
Conceptual Models
Conceptual model for Amazon.co.uk (UML class model notation)
Department
Products
Basket Checkout
Order HistorySelected Products
Ordered Products
Order
Account
Delivery Address
Address Book
Payment Settings
Payment DetailsSearch
Product Descriptions
75 Agile UX & UCD
syntagm
Design Styles & Minimum Viable Products
syntagm syntagm
UIE’s 5 Design Decision Styles
1. Unintended – No design: can result in very poor usability and
UX 2. Self design
– Designed by team for team – Difficult for other audiences to understand/use
3. Genius design – Works well with experience teams – Does require solid practical experience of users
Agile UX & UCD 77
syntagm Genius design can work – but not very often
syntagm syntagm
UIE’s 5 Design Decision Styles
4. Activity-focussed – Research based on users’ activities – Effective but narrow
5. User-focussed – Broad user-based research – Appropriate for green-field development and
major redesigns – Needed for excellent UX
Agile UX & UCD 79
(See www.uie.com/articles/five_design_decision_styles)
syntagm syntagm
MVP – The Lean Startup
• The goal of a startup is to figure out the right thing to build − the thing customers want and will pay for as quickly as possible – Eric Ries
The Lean Startup
Agile UX & UCD 80
syntagm Agile UX & UCD 81
(radoff.com)
Agile UX & UCD 82
syntagm syntagm
Minimum Viable Product
• Enough features to allow the product to be deployed and no more
• An alternative to extensive market research / user testing
• Popularized by Eric Reis in The Lean Startup
• But beware of the technology chasm
Agile UX & UCD 83
Agile UX & UCD 84
Innovators Early Adopters
Early Majority
Late Majority Laggards
The technology chasm
(Adapted from Geoffrey Moore’s Crossing the Chasm and Everett Roger’s Diffusion of Innovation)
Minimum Viable Product
syntagm
Personas and Persona Myths
syntagm syntagm
Users in Design • We need to identify who we are designing for
(and who we will be doing usability testing with) • While we could build profiles, we also need to get
our team thinking like and identifying with users (empathy)
• We must also design for what users really need and do (descriptive models) rather than what we think they should need and do (normative models)
• Otherwise, our user stories are works of fiction
Agile UX & UCD 86
syntagm syntagm
Personas • People are much more positive towards
individuals than groups • Personas are fictitious (but credible) individuals
who represent the main users of a solution, based on user research
• Must be developed and agreed by the team – you cannot promote empathy by forcing solutions on key players
• Personas form the basis of all discussions about features and user stories
Agile UX & UCD 87
syntagm syntagm
Psychology of Personas
• The Person-Positivity Bias (Sears, 1983) – David Sears found that attitudes towards
individuals were significantly more positive than groups with the same characteristics
• The Scope-Severity Paradox (Nordgren & McDonnell, 2010) – Nordgren and McDonnell established that
reactions to a fraud were more severe when it was presented with 3 victims rather than 30
Agile UX & UCD 88
syntagm syntagm Agile UX & UCD Photo by James Cridland 89
syntagm syntagm
Personas
Jane Soames is a 28-year-old London graphics designer who has moved into special effects. Her experience with graphics packages has helped her a lot with the 2-D modeling but she still struggles a bit with 3-D. Her current job role has her moving between 2-D and 3-D work, so having a lot of similarities between the two kinds of software makes her life a lot simpler. …
Agile UX & UCD 90
syntagm syntagm
Persona Guidelines
• Based on user research • Typically one primary persona for a product or
service - secondary personas have minor differences from primary
• Personas must be believable and liked by team members – They should have specific characteristics (for
example, age 28, not 25-35) – Names and photos agreed by all
Agile UX & UCD 91
syntagm syntagm
8 Persona Misconceptions
1. Personas are made up 2. Personas are about demographics 3. We can use the same personas across all of our
solutions 4. Personas come from market research 5. Personas are unscientific 6. Personas are for the user experience people 7. We only need to research our personas once 8. Personas will guarantee good UX Agile UX & UCD 92
syntagm syntagm
1. Personas are made up
• The name and backstory for a persona are made up, but the behaviours and needs a persona represents have been researched in the field
Agile UX & UCD 93
syntagm syntagm
2. Personas are about demographics
• No, personas are about important behaviours and needs you have identified in your user communities
• Demographics given in the backstory are just to make the persona real
• Demographics are used in the recruitment process but these are part of the user profile behind the persona
Agile UX & UCD 94
syntagm syntagm
Persona (Front)
• A persona is the face of a user community
• But make sure that designing for this persona will not disadvantage other users
Agile UX & UCD
Age, education, experience Products/services used (how often?)
Name, specific age, realistic photo
Backstory, interests,
motivations
Goals in using your solution
95
syntagm syntagm
• Behind every persona is a user profile
• These are used for – Recruitment (research and
usability testing) – Sales and marketing
• They include the demographics of the user group a persona represents
Agile UX & UCD
Age, education, experience (ranges)
Products/services used (how often?)
Behaviours and
recruitment questions
96
Persona (Reverse)
syntagm syntagm
3. We can use the same personas across all our solutions
• Personas represent the behaviours and needs of users as they relate to the problem(s) you’re solving
• Different behaviours and needs will be relevant to different problem domains – For example many users might be interested in a
cloud music service – But their needs may be substantially different
depending on whether they prefer classical music or hard rock
Agile UX & UCD 97
syntagm syntagm Agile UX & UCD 98
Handel and Jimi Hendrix shared an address (23 Brook Street, London) but would probably have had different needs for a music service
syntagm syntagm
4. Personas come from market research
• Marketing personas can be a good starting point for design personas if they are behaviour based
• Marketing personas based primarily on demographics will not be helpful
Agile UX & UCD 99
syntagm syntagm
5. Personas are unscientific
• Personas make use of well-established principles from cognitive psychology – The Person-Positivity Bias (Sears, 1983) – The Scope-Severity Paradox (Nordgren &
McDonnell, 2010) – Decisions for Others Are More Creative Than
Decisions for the Self (Polman and Emich, 2011)
Agile UX & UCD 100
syntagm syntagm
6. Personas are for the user experience people
• The primary role of personas is to make users more real to the development team
• They will be researched and created in collaboration with UX practitioners but they need to be accepted by the whole team
Agile UX & UCD 101
syntagm syntagm
7. We only need to research our personas once
• In fast-moving domains, personas should be updated before a major refresh or new projects
• For example, the popularity of tablets significantly changed users’ needs for web site interactivity (and partially drove the rush for apps)
Agile UX & UCD 102
syntagm syntagm
8. Personas will guarantee good UX
• Not on their own. Well-researched personas, collaboratively developed (see Guerrilla UCD session 14), will help keep the focus on real users, but there are many more components – Early and frequent usability evaluations (session 6) – Persona-driven user stories (session 15) – Clear goals & conceptual model (session 16)
Agile UX & UCD 103
syntagm syntagm
Activity 3
• Organize into small groups with someone to take notes
• Await further instructions!
Agile UX & UCD 104
syntagm
User Stories vs Persona Stories
105
syntagm syntagm
User in Design
Agile UX & UCD
User stories written without user research are just wishful thinking
106
syntagm syntagm
User Stories
• In Agile projects user stories replace earlier forms of requirement specification (such as use cases)
• They are really placeholders – detailed design is done much closer to implementation (just-in-time design)
• For UX design, user stories need to be accompanied by prototypes (at implementation)
Agile UX & UCD 107
syntagm syntagm
User Stories
• User stories are very brief – typically a single side of a index or post card
• Large stories (called ‘epics’) should be split into more manageable chunks
• Do not get too specific on how the interaction should work in early stages (avoid premature design)
Agile UX & UCD 108
syntagm syntagm
Persona Stories
• User stories are not ideal for user experience • They’re not really about users at all; the Agile-
recommended form is – “As a <role>, I want <goal/desire> [so that…]”
• For user-centred design, stories should be about personas and avoid the first person – “Jane updates contact preferences” – (<persona> <goal/desire>)
Agile UX & UCD 109
syntagm syntagm
Persona Stories
• This form is much easier to write and read. It also emphasizes that the development team is not the user, which has benefits including – Decisions for Others Are More Creative Than
Decisions for the Self (Polman and Emich, 2011)
• See my article User Stories Don’t Help Users: Introducing Persona Stories, ACM Interactions, Nov+Dec 2013, www.personastories.com
Agile UX & UCD 110
syntagm
Agile Usability Testing
syntagm syntagm
Agile Usability Testing • Early and often
– Part of each cycle, ideally weekly – Test whatever you have ready
• Don’t be afraid of remote (supervised) testing – Desktop sharing, audio & video – Recording usually easy (LiveMeeting, WebEx, GotoMeeting…)
• Consider user surrogates for brief ‘revolving door’ sessions – Colleagues not involved/familiar with the project but similar to users
in other respects • Make use of central UX or external resource for recruitment
– Time-consuming and specialist: don’t rely on team UX role • Engage the rest of the team
– Encourage live viewing (in-person or streaming video) – Make videos readily available – Organize usability/UX events around test sessions – Involve developers directly with the RITE method
Agile UX & UCD 112
syntagm syntagm
Agile Usability Testing • RITE (Rapid Iterative Testing and Evaluation)
– Michael Medlock and Dennis Wixon, Microsoft – Developers involved in usability testing and make
repairs in situ • Four categories of issue
1. Obvious cause, obvious solution that can be implemented quickly
2. Obvious cause, obvious solution but not quick 3. No obvious cause, therefore no obvious solution 4. Other factors
Agile UX & UCD 113
syntagm syntagm
Agile Usability Testing • Reported advantages of RITE
– The usability issues were “believed”. The decision-makers had often pre-defined what tasks participants should be able to accomplish. In addition, through their constant involvement the decision-makers “believed” issues for which there were no previous tasks (issues they or the usability engineer had not anticipated).
– Fixing the discovered issues was planned for and agreed upon prior to testing.
– The usability feedback was delivered as soon as it possibly could be –right after the issues occurred.
– The team had measurable assurance that the solutions were successfully fixing the problems because the fixes were tested by the subsequent participants. In addition the team caught “poor” fixes for problems and corrected them.
Agile UX & UCD
(See http://tinyurl.com/RITEmethodology)
114
syntagm syntagm
Agile Usability Testing
• Other quick approaches – Revolving door testing with on-site proxies – Remote testing (supervised) – Paper prototyping – Card sorting & tree sorting – A/B testing
Agile UX & UCD 115
syntagm
Persuasive Facilitation
116
syntagm syntagm
Persuasive Facilitation
• Persuasive facilitation requires knowing what evidence to present and when – Qualitative data often not seen as scientific – Technologists may not appreciate or understand
problems that users have – Different learning/thinking styles may favour
quantitative evidence (budgets, timings...)
117 Agile UX & UCD
syntagm syntagm
Activity 4
• Organize into small groups with someone to take notes
• Await further instructions!
Agile UX & UCD 118
syntagm syntagm
Persuasive Facilitation
• A staged approach to selling UCD may be needed, particularly in teams having low UX maturity 1. Stimulate 2. Explain 3. Engage (The ‘hook, line and sinker’ approach)
Agile UX & UCD 119
syntagm syntagm
Persuasive Facilitation
1. Stimulate – Stimulate interest in UCD through quantitative
evidence such as server traffic analysis, support centre statistics, customer satisfaction surveys
Agile UX & UCD 120
syntagm syntagm
Persuasive Facilitation
2. Explain – Use qualitative research to explain how to
improve usability and the user experience (usability tests, card sorting, support transcripts)
– Use ‘root cause analysis’ (or similar) to explain how identified usability issues can be addressed (RCA is a key component of quality assurance)
Agile UX & UCD 121
syntagm syntagm Agile UX & UCD 122
(Ishikawa diagram created with free Excel tool at www.freequality.org)
syntagm syntagm
Persuasive Facilitation
3. Engage – Engage the team in UCD techniques that build
UX into the development process (personas, paper prototypes, usability testing, design maps)
– More on design maps later
Agile UX & UCD 123
syntagm syntagm
Persuasive Facilitation
• The extent to which persuasive facilitation is necessary will depend on the UX maturity of the team (and in some cases, individuals)
Agile UX & UCD 124
syntagm syntagm
UX Maturity Model
Agile UX & UCD 125
(from http://johnnyholland.org/2010/04/16/planning-your-ux-strategy/)
syntagm syntagm
The Agile Team
Customer / Owner / User Rep
Users ?
126 Agile UX & UCD
syntagm syntagm
UX/UCD-Engaged Agile Team
Agile UX & UCD 127
UCD Practitioner
Users
syntagm syntagm
Outsourcing UX/UCD Activities
Agile UX & UCD 128
External Supplier
Users
(Can work if supplier involves team members as observers)
syntagm
Design Maps
syntagm syntagm
Design Maps
• In their book The Persona Lifecycle, Pruitt and Adlin describe a tool called ‘design maps’
• Design maps connect personas and user stories to detailed design issues and artefacts such as wireframes, sketches and prototypes
Agile UX & UCD 130
syntagm syntagm
Design Maps
Agile UX & UCD 131
Who • Personas • Persona-weighted feature matrix
What
• User stories (including features and constraints)
• Scenarios (elaborated stories)
How • Design maps • Design artefacts (sketches, wireframes...)
syntagm syntagm
Scenarios
• Scenarios are elaborated descriptions of user stories, usually focussed on a persona’s goal
• The can be written to represent a concept (evocative) or to describe detailed interaction (prescriptive)
• Prescriptive scenarios can be accompanied by design artefacts such as wireframes or prototypes
Agile UX & UCD 132
(See The Persona Lifecycle for more on scenarios)
syntagm syntagm
Anatomy of a Design Map • Design Map: Megan Delivers the Presentation
Megan logs on to the presentation system
Megan sees the presenter page
Megan sees that her slides are ready and she does a last-minute flip through
Megan sees the audience members are starting to arrive
Should we let Megan log on if Ivan hasn’t set everything up yet?
The presenter screen should reassure her that the streams are started and the preso is ready
What if an audience member tries to connect before Megan, or even before Ivan?
Megan has already uploaded all of her slides
Let’s create a way for her to flip through her slides without any audience members seeing
Map Title
Step
Question
Comment Design Idea
Key step comment
question
idea
Design Map: MaryAnn finds a place to work for a few hours MaryAnn arrives at the corporate headquarters at 10:30 am.
MaryAnn knows she needs a space to work until 1 pm, and maybe after that as well.
Are there particularly busy times for hoteling check-in today?
MaryAnn has not signed up for hoteling space for today.
Let’s create a way for her to book a space ‘tentatively’ and then ask her to confirm with a mobile app!
What happens if there are absolutely no spaces left for her?
MaryAnn sees several ‘places you can work today’ options.
MaryAnn can preview the spaces using photos...and maybe see if colleagues are working nearby!
MaryAnn arrives at the space and ‘badges in...’
Slide courtesy of Tamara Adlin (see Pruitt & Adlin, The Persona Lifecycle)
133 Agile UX & UCD
syntagm syntagm
MaryAnn schedules a catered breakfast meeting
Slide courtesy of Tamara Adlin (see Pruitt & Adlin, The Persona Lifecycle)
Agile UX & UCD 134
syntagm
Logical steps and actions from Design Maps go together to form single pages in a wireframe
135 Agile UX & UCD
Slide courtesy of Tamara Adlin (see Pruitt & Adlin, The Persona Lifecycle)
syntagm syntagm
Collaboration Tools • Code/Defects/Project
– Atlassian: broad range including Jira & Bitbucket – Microsoft: Application Lifecycle Management – Huddle: enterprise PM collaboration – Wrike: All-in-one collaborative PM
• Visual collaboration – Conceptboard – GroupZap – Mural.ly – Stormboard
Agile UX & UCD 136
syntagm syntagm
Design map created using online visual collaboration tool (conceptboard.com)
Design Map: MaryAnn finds a place to work for a few hours
Agile UX & UCD 137
syntagm syntagm
Design map created using online visual collaboration tool (mural.ly)
Agile UX & UCD 138
syntagm syntagm
Collaboration Tools
• General – Scribblar – GotoMeeting/GotoWebinar – Adobe Connect – Webex – Microsoft Live Meeting
Agile UX & UCD 139
syntagm syntagm
Design Maps Activity
• Organize into teams of 3-5 • Await further instructions!
Agile UX & UCD 140
syntagm
Appendix
syntagm
Number of Times Methods Used Per Year
Survey of 217 participants in Usability Week 2008 (Courtesy of Kara Pernice, NN/g)
15.2 Server traffic log analysis 11.0 Search log analysis
4.1 User testing 3.3 Surveys 3.3 Heuristic evaluation / expert review 2.9 Low-fidelity/paper prototyping 2.5 Participatory design 2.2 Measurement studies 2.1 Remote testing 2.1 Field studies 2.0 Card sorting 1.5 Competitive studies 1.4 A/B testing 1.0 Eyetracking 0.8 Diary study
142 Agile UX & UCD
syntagm
Bibliography and Further Reading
syntagm syntagm
Bibliography and Further Reading • Agile Software Development
– Cockburn, A. (2002). Agile software development: software through people, Addison Wesley.
– Cohn, M. (2009). Succeeding with Agile: Software Development Using Scrum, Addison Wesley.
• Agile UCD – Nodder, C. N., Jakob (2009). Agile Usability: Best
Practices for User Experience on Agile Development Projects, NN/g (www.nngroup.com/reports/agile/)
– Sy, D. (2007). "Adapting usability investigations for agile user-centered design." Journal of usability Studies, 2(3): 112–132.
Agile UX & UCD 144
syntagm syntagm
Bibliography and Further Reading
• Contextual Design – Beyer, H. and K. Holtzblatt (1998). Contextual Design:
Defining Customer-Centered Systems. San Francisco, CA, Morgan Kaufmann.
– Holtzblatt, K., J. Wendell, et al. (2005). Rapid contextual design: a how-to guide to key techniques for user-centered design, Morgan Kaufmann.
• Design Maps – Pruitt, J. and T. Adlin (2006). The persona lifecycle:
keeping people in mind throughout product design, Morgan Kaufmann.
Agile UX & UCD 145
syntagm syntagm
Bibliography and Further Reading
• Empathizing/Systemizing – Baron-Cohen, S. (2008). Autism and Asperger
Syndrome (The Facts), Oxford University Press. – Hudson, W. (2009). Reduced empathizing skills
increase challenges for user-centered design, ACM.
– Wray, S. (2007). SQ Minus EQ can Predict Programming Aptitude, PPIG.
Agile UX & UCD 146
syntagm syntagm
Bibliography and Further Reading
• Personas – Pruitt, J. and T. Adlin (2006). The persona lifecycle:
keeping people in mind throughout product design, Morgan Kaufmann.
– Goodwin, K. (2009). Designing for the Digital Age: How to Create Human-Centered Products and Services, John Wiley & Sons.
• Qualitative Research – Kvale, S. (1994). "Ten standard objections to
qualitative research interviews." Journal of Phenomenological Psychology, 25(2): 147-173.
Agile UX & UCD 147
syntagm syntagm
Bibliography and Further Reading
• User Stories – Cohn, M. (2004). User Stories Applied: For Agile
Software Development, Addison Wesley.
Agile UX & UCD 148
syntagm 149
www.syntagm.co.uk/design
Agile UX & UCD
top related