_ Designing with data Lucinda Burtt - Fairfax Media @LucindaBurtt
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
_Designing with data
Lucinda Burtt - Fairfax Media
@LucindaBurtt
_Intro
What makes great product design?
why what
want
Experience Design
Discern how the user will interact with the product, allow them to do what they are trying to do
Visual Design
Make the interface communicate what it can do so the user wants to interact with it
Research & Optimisation
To discover who our users are and ask: why would they use our product?
Follow up: will they use it?
_Why UX design and data?
Definitions & Lean UX
_
Facts or information used to calculate, analyse or plan something
Merriam-Webster
data
“
Research is formalised curiosity. It is poking and prying with a purpose
Zora Neale Hurston
A set of principles (not a process) for creating digital experiences that are valued by end users
The three foundations of Lean UX
● Design thinkingInnovation through observation
● Agile software developmentContinuous delivery
● Build-Measure-LearnLean Startup feedback loop
Source: Lean UX (2013), Jeff Gothelf
Further reading: Stanford d.school
Source: Lean UX (2013), Jeff Gothelf
Principles of Lean UX
Aiming for
Early customer validation
Outcomes, not outputs
Collaborative design
Measuring KPIs
Applying appropriate tools
Further reading: ‘The Lean UX Manifesto’ - smashingmagazine.com, Anthony Viviano
Instead of
Releasing unknown end-user value
Working on the next ‘cool’ feature
Siloed product design & development
Undefined success metrics
Following a rigid plan
Early customer validation
● Learning about real user needs
● 360° view of the customer
● Seeking customer validation or early failure
Surveys
User forms
Call centre
Usability tests
Card sorting
Support emails
Review sites
Analytics
Diary studies
Focus groups
Outbound calls
Outcomes, not output
● What problem are we trying to solve?
● Hypothesis statements: we believe that...
● User stories/JTBD
Collaborative design
● Design studio method: create, pitch, critique
● Style guides and pattern libraries
● Cross-functional: product, business analyst, design & dev teams
Measuring KPIs
● Acquisitions
● Activation
● Retention
● Referral
● Revenue
Applying appropriate tools
● Done, not perfect
● User personas
● Identifying assumptions
● Prototyping
_Making meaning from data
What data to use?
Source: ‘When to use which UX methods’ - nngroup.com, Christian Rohrer
What data to collect?
Qualitative
The “why” and “how”
Understanding user logic
Provides perspective, direct
Looks like: usability testing, personas
Quantitative
The “how many/much” and “will they…”
Understanding user patterns
Provides tactical insight, indirect
Looks like: A/B testing, analytics (heatmaps)
Phases of product development ● Strategize
(Qualitative/Quantitative)New ideas and opportunities
● Execute(Qualitative)“Go/No-Go” decisions
● Assess(Quantitative)Measurement of success
Source: ‘When to use which UX methods’ - nngroup.com, Christian Rohrer
Lean user testing
● Less users, more often
● Sketches validate value, visual mockups test brand effectiveness, prototypes provide direct insight
● Results fed straight into development
User personas
● Fictional yet realistic, data-based description of a typical/target user
● Each persona represents a cluster of user behaviours/needs
● A tool to create empathy between us and the user
Heatmaps
● Graphical/visual representation of data
● Insights into attention and distraction metrics
● Identifying problem areas
● Behavioural
A/B testing
● Testing intuitions and assumptions
● Validating our qual insights
● Understanding the wide impact of design decisions
● Tactical
What to A/B test first?
The PIE framework
● PotentialHow much improvement can be made?
● ImportanceHow valuable is the traffic to those pages?
● EaseHow complex is the test to implement?
Source: ‘How to prioritise [CRO] tests using PIE’ - widerfunnel.com, Chris Goward
Case StudyNewswell WOFFs
_Moving to data informed design
Conclusion
“
If I had asked people what they wanted, they would have said faster horses
Henry Ford
Data informed design● Focus on key decisions
Testable, tactical
● Define your hypothesis
● Choose your KPIs wiselyBreadth of data
● BenchmarkSet a control, compare like-for-like
● Analyse & ApplyDesign, collect, analyse, iterate
Further reading: ‘Data informed design’ - uxforthemasses.com, Neil Turner
_Thank you!
16 June 2016
@LucindaBurtt
Related Documents
