SlideShare uma empresa Scribd logo
1 de 25
Measuring For Impact: Knowing
What, and How to A/B Test
@mike_greenfield
CEO/Co-Founder, Laserlike
2014-08-07
@mike_greenfield
You know you should A/B test.
@mike_greenfield
You also know you should
exercise more
eat less sugar
spend less on coffee
wear sunscreen
etc., etc.
@mike_greenfield
(Don’t worry, I’m not going to
say anything else about sugar
or sunscreen.)
@mike_greenfield
So, how do you create a culture
in which people will
constructively A/B test?
Do six things.
@mike_greenfield
1. Embrace “I don’t know”
We have 2+ ideas.
I don’t know which one will be more
effective.
@mike_greenfield
@mike_greenfield
2. Have Data, Choose
Metrics
To test, you need:
• People using your product
• (Approximate) agreement on the metrics
that matter
@mike_greenfield
Not Many Users? Don’t A/B test!
• Laserlike, has ~60 users and has never
run an A/B test
• We will run many, many tests when we
have enough users
• A test should have at least a few hundred
instances (and a lot more if effect sizes
are likely to be small)
• Test iff you can have “business
significance”
@mike_greenfield
Know What You Want to Optimize
• If it’s important, you should be running
tests to improve it
• If it’s not important, spend time on other
things
• Most tests should be aimed at improving
1-2 specific variables
@mike_greenfield
3. Have Clear Process, Tech
for Testing
@mike_greenfield
A/B Testing Process
• New feature: if possible, roll out to a small
test subset first (10s or 100s of thousands)
• Version change: always test things that
could (cumulatively) have business impact
• Everyone on the product team should be
running and resolving tests
@mike_greenfield
A/B Testing Tech
• Using a third party testing service is akin to
building your site on Wordpress: great at
some scales/competency levels
• No matter how you’re testing, a new test
should be at most a few lines of code
• It should be easy to see how each side of a
test compares across many variables
@mike_greenfield
4. Understand the Math of
What to Test
@mike_greenfield
Process: Same vs. New Tweak
• What’s the probability your tweak will have
a positive effect?
• What kind of effect might that have, and
how might that effect change the
company’s prospects?
• Will you be able to measure the change?
• Optimize on one variable, but look at
others
@mike_greenfield
Process: Same vs. Big Change
• What’s the probability that your change will
have a negative impact?
• How big an impact might there be?
• Will you be able to measure the change?
• Holistic approach
@mike_greenfield
A/B Test for Quality
• Circle of Moms: test “warning” users when
questions seemed short, low quality
• Resulting questions were graded for quality,
without grader knowing test bucket
• End result: warning yielded ~5% fewer questions,
but much higher quality
@mike_greenfield
5. Understand the Math of
Picking Winners
@mike_greenfield
Resolving Too Soon vs. Resolving
Too Late
• How big is the potential audience for this
test?
• Example 1: end of year “most popular
baby names” email that will never be sent
again
• Example 2: Facebook signup flow
@mike_greenfield
Longitudinal Tests vs. Immediate
Tests
• Longitudinal: change home page, email
frequency, product framing
• Need to examine effect over a long period
• Immediate: change button color, email
subject
• Likely that long-term effects will be
minimal
@mike_greenfield
Automatically Resolve Tests?
• Longitudinal tests should not be
automatically resolved
• Example: new home page design
• Immediate tests can be automatically
resolved when speed is important and
there is one clear objective function
• Example: Circle of Moms email subject
optimization
@mike_greenfield
Choose robust statistics
• Bad: # of page views
• Good: % of users viewing at least [5, 25,
100] pages
• Potentially bad: # of sales (when small)
• Potentially good: # of people getting
through the second step of a sales funnel
@mike_greenfield
6. Celebrate A/B Testing
Successes
@mike_greenfield
@mike_greenfield
Thanks.
mike@laserlike.com
@mike_greenfield
@mike_greenfield

Mais conteúdo relacionado

Mais procurados

The Hard Thing About Hard Things
The Hard Thing About Hard ThingsThe Hard Thing About Hard Things
The Hard Thing About Hard ThingsAmazon Web Services
 
Always be Testing: How eROI + Taco Bell Built a Testing Agenda (and What Happ...
Always be Testing: How eROI + Taco Bell Built a Testing Agenda (and What Happ...Always be Testing: How eROI + Taco Bell Built a Testing Agenda (and What Happ...
Always be Testing: How eROI + Taco Bell Built a Testing Agenda (and What Happ...eROI
 
[WMD 2015] Qualaroo >> Sean Ellis, "Rhythm Is A Hacker: The Advantages & Chal...
[WMD 2015] Qualaroo >> Sean Ellis, "Rhythm Is A Hacker: The Advantages & Chal...[WMD 2015] Qualaroo >> Sean Ellis, "Rhythm Is A Hacker: The Advantages & Chal...
[WMD 2015] Qualaroo >> Sean Ellis, "Rhythm Is A Hacker: The Advantages & Chal...500 Startups
 
10NTC - Data Superheroes - DiJulio
10NTC - Data Superheroes - DiJulio10NTC - Data Superheroes - DiJulio
10NTC - Data Superheroes - DiJuliosarahdijulio
 
[CXL Live 16] Growth Hacking BS: Fixing Marketing One Truth at a Time by Morg...
[CXL Live 16] Growth Hacking BS: Fixing Marketing One Truth at a Time by Morg...[CXL Live 16] Growth Hacking BS: Fixing Marketing One Truth at a Time by Morg...
[CXL Live 16] Growth Hacking BS: Fixing Marketing One Truth at a Time by Morg...CXL
 
My question tool 2
My question tool 2My question tool 2
My question tool 2Jenny Kang
 
5 whys - The Path to Resolution
5 whys - The Path to Resolution5 whys - The Path to Resolution
5 whys - The Path to ResolutionTor Ivry
 
Pavlo Pedenko. Measuring Product-Market Fit Quantitatively. Setapp and Slack ...
Pavlo Pedenko. Measuring Product-Market Fit Quantitatively. Setapp and Slack ...Pavlo Pedenko. Measuring Product-Market Fit Quantitatively. Setapp and Slack ...
Pavlo Pedenko. Measuring Product-Market Fit Quantitatively. Setapp and Slack ...IT Arena
 
Data Analysis Goes Wrong by Microsoft Sr PM
Data Analysis Goes Wrong by Microsoft Sr PMData Analysis Goes Wrong by Microsoft Sr PM
Data Analysis Goes Wrong by Microsoft Sr PMProduct School
 
How can we understand the problem?
How can we understand the problem?How can we understand the problem?
How can we understand the problem?Frank Calberg
 
[CXL Live 16] Beyond Test-by-Test Results: CRO Metrics for Performance & Insi...
[CXL Live 16] Beyond Test-by-Test Results: CRO Metrics for Performance & Insi...[CXL Live 16] Beyond Test-by-Test Results: CRO Metrics for Performance & Insi...
[CXL Live 16] Beyond Test-by-Test Results: CRO Metrics for Performance & Insi...CXL
 
Five Whys Lessons Learned
Five Whys Lessons LearnedFive Whys Lessons Learned
Five Whys Lessons LearnedTony Ford
 
Get comfortable breaking your product - Mind the Product 2018 conference talk
Get comfortable breaking your product - Mind the Product 2018 conference talkGet comfortable breaking your product - Mind the Product 2018 conference talk
Get comfortable breaking your product - Mind the Product 2018 conference talkRik Higham
 
Stop multiplying by 4: Practical Software Estimation
Stop multiplying by 4: Practical Software EstimationStop multiplying by 4: Practical Software Estimation
Stop multiplying by 4: Practical Software EstimationChuck Reeves
 
Machine Learning Vital Signs
Machine Learning Vital SignsMachine Learning Vital Signs
Machine Learning Vital SignsDonald Miner
 
What people cant capture
What people cant captureWhat people cant capture
What people cant captureHimanshu Sharma
 
3 types of unhappy customers
3 types of unhappy customers3 types of unhappy customers
3 types of unhappy customersGlyn Davies
 

Mais procurados (20)

The Hard Thing About Hard Things
The Hard Thing About Hard ThingsThe Hard Thing About Hard Things
The Hard Thing About Hard Things
 
Always be Testing: How eROI + Taco Bell Built a Testing Agenda (and What Happ...
Always be Testing: How eROI + Taco Bell Built a Testing Agenda (and What Happ...Always be Testing: How eROI + Taco Bell Built a Testing Agenda (and What Happ...
Always be Testing: How eROI + Taco Bell Built a Testing Agenda (and What Happ...
 
[WMD 2015] Qualaroo >> Sean Ellis, "Rhythm Is A Hacker: The Advantages & Chal...
[WMD 2015] Qualaroo >> Sean Ellis, "Rhythm Is A Hacker: The Advantages & Chal...[WMD 2015] Qualaroo >> Sean Ellis, "Rhythm Is A Hacker: The Advantages & Chal...
[WMD 2015] Qualaroo >> Sean Ellis, "Rhythm Is A Hacker: The Advantages & Chal...
 
10NTC - Data Superheroes - DiJulio
10NTC - Data Superheroes - DiJulio10NTC - Data Superheroes - DiJulio
10NTC - Data Superheroes - DiJulio
 
[CXL Live 16] Growth Hacking BS: Fixing Marketing One Truth at a Time by Morg...
[CXL Live 16] Growth Hacking BS: Fixing Marketing One Truth at a Time by Morg...[CXL Live 16] Growth Hacking BS: Fixing Marketing One Truth at a Time by Morg...
[CXL Live 16] Growth Hacking BS: Fixing Marketing One Truth at a Time by Morg...
 
My question tool 2
My question tool 2My question tool 2
My question tool 2
 
5 whys - The Path to Resolution
5 whys - The Path to Resolution5 whys - The Path to Resolution
5 whys - The Path to Resolution
 
Pavlo Pedenko. Measuring Product-Market Fit Quantitatively. Setapp and Slack ...
Pavlo Pedenko. Measuring Product-Market Fit Quantitatively. Setapp and Slack ...Pavlo Pedenko. Measuring Product-Market Fit Quantitatively. Setapp and Slack ...
Pavlo Pedenko. Measuring Product-Market Fit Quantitatively. Setapp and Slack ...
 
Data Analysis Goes Wrong by Microsoft Sr PM
Data Analysis Goes Wrong by Microsoft Sr PMData Analysis Goes Wrong by Microsoft Sr PM
Data Analysis Goes Wrong by Microsoft Sr PM
 
How can we understand the problem?
How can we understand the problem?How can we understand the problem?
How can we understand the problem?
 
[CXL Live 16] Beyond Test-by-Test Results: CRO Metrics for Performance & Insi...
[CXL Live 16] Beyond Test-by-Test Results: CRO Metrics for Performance & Insi...[CXL Live 16] Beyond Test-by-Test Results: CRO Metrics for Performance & Insi...
[CXL Live 16] Beyond Test-by-Test Results: CRO Metrics for Performance & Insi...
 
Five Whys Lessons Learned
Five Whys Lessons LearnedFive Whys Lessons Learned
Five Whys Lessons Learned
 
Get comfortable breaking your product - Mind the Product 2018 conference talk
Get comfortable breaking your product - Mind the Product 2018 conference talkGet comfortable breaking your product - Mind the Product 2018 conference talk
Get comfortable breaking your product - Mind the Product 2018 conference talk
 
5 whys
5 whys5 whys
5 whys
 
Software testing
Software testingSoftware testing
Software testing
 
Stop multiplying by 4: Practical Software Estimation
Stop multiplying by 4: Practical Software EstimationStop multiplying by 4: Practical Software Estimation
Stop multiplying by 4: Practical Software Estimation
 
Machine Learning Vital Signs
Machine Learning Vital SignsMachine Learning Vital Signs
Machine Learning Vital Signs
 
What people cant capture
What people cant captureWhat people cant capture
What people cant capture
 
3 types of unhappy customers
3 types of unhappy customers3 types of unhappy customers
3 types of unhappy customers
 
Digital 201: Email Marketing
Digital 201: Email MarketingDigital 201: Email Marketing
Digital 201: Email Marketing
 

Destaque

Google Partner Agency Digital Breakfast with Geary LSF
Google Partner Agency Digital Breakfast with Geary LSFGoogle Partner Agency Digital Breakfast with Geary LSF
Google Partner Agency Digital Breakfast with Geary LSFKatie Fellenz
 
University of Nebraska Think Tank: Google Analytics as your Digital Marketing...
University of Nebraska Think Tank: Google Analytics as your Digital Marketing...University of Nebraska Think Tank: Google Analytics as your Digital Marketing...
University of Nebraska Think Tank: Google Analytics as your Digital Marketing...Caleb Whitmore
 
A/B Testing With Real Examples And Results
A/B Testing With Real Examples And ResultsA/B Testing With Real Examples And Results
A/B Testing With Real Examples And ResultsWilson Pang
 
How Hospital Administrators Make Purchase Decision - Think with Google Report
How Hospital Administrators Make Purchase Decision - Think with Google Report How Hospital Administrators Make Purchase Decision - Think with Google Report
How Hospital Administrators Make Purchase Decision - Think with Google Report Medico Mart
 
comScore's Presentation Google's Think Digital Vietnam 11-Dec-2012
comScore's Presentation Google's Think Digital Vietnam 11-Dec-2012comScore's Presentation Google's Think Digital Vietnam 11-Dec-2012
comScore's Presentation Google's Think Digital Vietnam 11-Dec-2012action.vn
 
Evolution of TV Measuring Across All Screens | Think With Google
Evolution of TV Measuring Across All Screens | Think With GoogleEvolution of TV Measuring Across All Screens | Think With Google
Evolution of TV Measuring Across All Screens | Think With GoogleArgent media agency
 
Keynote Wheel of Persuasion - Bart Schutz of Online Dialogue on how to persua...
Keynote Wheel of Persuasion - Bart Schutz of Online Dialogue on how to persua...Keynote Wheel of Persuasion - Bart Schutz of Online Dialogue on how to persua...
Keynote Wheel of Persuasion - Bart Schutz of Online Dialogue on how to persua...Online Dialogue
 
The Best A/B Test Idea You Haven’t Thought Of
The Best A/B Test Idea You Haven’t Thought OfThe Best A/B Test Idea You Haven’t Thought Of
The Best A/B Test Idea You Haven’t Thought OfKissmetrics on SlideShare
 
Keynote Digital Data Tips Tuesday - Amsterdam - June 24th 2014 - A/B-testing ...
Keynote Digital Data Tips Tuesday - Amsterdam - June 24th 2014 - A/B-testing ...Keynote Digital Data Tips Tuesday - Amsterdam - June 24th 2014 - A/B-testing ...
Keynote Digital Data Tips Tuesday - Amsterdam - June 24th 2014 - A/B-testing ...Online Dialogue
 
Auto aftermarket digital driver research studies
Auto aftermarket digital driver research studiesAuto aftermarket digital driver research studies
Auto aftermarket digital driver research studiesHouseLead Italia
 
The anatomy of an A/B Test - JSConf Colombia Workshop
The anatomy of an A/B Test - JSConf Colombia WorkshopThe anatomy of an A/B Test - JSConf Colombia Workshop
The anatomy of an A/B Test - JSConf Colombia WorkshopEduardo Shiota Yasuda
 
Why everything is an A/B Test at Pinterest
Why everything is an A/B Test at PinterestWhy everything is an A/B Test at Pinterest
Why everything is an A/B Test at PinterestKrishna Gade
 
How to Successfully Run Your First Website A/B Test
How to Successfully Run Your First Website A/B TestHow to Successfully Run Your First Website A/B Test
How to Successfully Run Your First Website A/B TestKissmetrics on SlideShare
 
A/B Testing at Pinterest: Building a Culture of Experimentation
A/B Testing at Pinterest: Building a Culture of Experimentation A/B Testing at Pinterest: Building a Culture of Experimentation
A/B Testing at Pinterest: Building a Culture of Experimentation WrangleConf
 
Millward brown vermeer organizational design think brand 2014 snf
Millward brown vermeer organizational design think brand 2014 snfMillward brown vermeer organizational design think brand 2014 snf
Millward brown vermeer organizational design think brand 2014 snfVermeer
 
Google Automotive Think Insights - Digital Drives Car Researches and Auto Pur...
Google Automotive Think Insights - Digital Drives Car Researches and Auto Pur...Google Automotive Think Insights - Digital Drives Car Researches and Auto Pur...
Google Automotive Think Insights - Digital Drives Car Researches and Auto Pur...SL Ecommerce and ReviewsReputation.com
 
The Operating Model That is Eating the World
The Operating Model That is Eating the WorldThe Operating Model That is Eating the World
The Operating Model That is Eating the Worldundercurrent
 
Always Be Testing - Learn from Every A/B Test (Hiten Shah)
Always Be Testing - Learn from Every A/B Test (Hiten Shah)Always Be Testing - Learn from Every A/B Test (Hiten Shah)
Always Be Testing - Learn from Every A/B Test (Hiten Shah)Future Insights
 

Destaque (20)

Google Partner Agency Digital Breakfast with Geary LSF
Google Partner Agency Digital Breakfast with Geary LSFGoogle Partner Agency Digital Breakfast with Geary LSF
Google Partner Agency Digital Breakfast with Geary LSF
 
University of Nebraska Think Tank: Google Analytics as your Digital Marketing...
University of Nebraska Think Tank: Google Analytics as your Digital Marketing...University of Nebraska Think Tank: Google Analytics as your Digital Marketing...
University of Nebraska Think Tank: Google Analytics as your Digital Marketing...
 
A/B Testing With Real Examples And Results
A/B Testing With Real Examples And ResultsA/B Testing With Real Examples And Results
A/B Testing With Real Examples And Results
 
Mobile A/B Test
Mobile A/B TestMobile A/B Test
Mobile A/B Test
 
How Hospital Administrators Make Purchase Decision - Think with Google Report
How Hospital Administrators Make Purchase Decision - Think with Google Report How Hospital Administrators Make Purchase Decision - Think with Google Report
How Hospital Administrators Make Purchase Decision - Think with Google Report
 
comScore's Presentation Google's Think Digital Vietnam 11-Dec-2012
comScore's Presentation Google's Think Digital Vietnam 11-Dec-2012comScore's Presentation Google's Think Digital Vietnam 11-Dec-2012
comScore's Presentation Google's Think Digital Vietnam 11-Dec-2012
 
Evolution of TV Measuring Across All Screens | Think With Google
Evolution of TV Measuring Across All Screens | Think With GoogleEvolution of TV Measuring Across All Screens | Think With Google
Evolution of TV Measuring Across All Screens | Think With Google
 
Keynote Wheel of Persuasion - Bart Schutz of Online Dialogue on how to persua...
Keynote Wheel of Persuasion - Bart Schutz of Online Dialogue on how to persua...Keynote Wheel of Persuasion - Bart Schutz of Online Dialogue on how to persua...
Keynote Wheel of Persuasion - Bart Schutz of Online Dialogue on how to persua...
 
The Best A/B Test Idea You Haven’t Thought Of
The Best A/B Test Idea You Haven’t Thought OfThe Best A/B Test Idea You Haven’t Thought Of
The Best A/B Test Idea You Haven’t Thought Of
 
Keynote Digital Data Tips Tuesday - Amsterdam - June 24th 2014 - A/B-testing ...
Keynote Digital Data Tips Tuesday - Amsterdam - June 24th 2014 - A/B-testing ...Keynote Digital Data Tips Tuesday - Amsterdam - June 24th 2014 - A/B-testing ...
Keynote Digital Data Tips Tuesday - Amsterdam - June 24th 2014 - A/B-testing ...
 
Auto aftermarket digital driver research studies
Auto aftermarket digital driver research studiesAuto aftermarket digital driver research studies
Auto aftermarket digital driver research studies
 
The anatomy of an A/B Test - JSConf Colombia Workshop
The anatomy of an A/B Test - JSConf Colombia WorkshopThe anatomy of an A/B Test - JSConf Colombia Workshop
The anatomy of an A/B Test - JSConf Colombia Workshop
 
Why everything is an A/B Test at Pinterest
Why everything is an A/B Test at PinterestWhy everything is an A/B Test at Pinterest
Why everything is an A/B Test at Pinterest
 
How to Successfully Run Your First Website A/B Test
How to Successfully Run Your First Website A/B TestHow to Successfully Run Your First Website A/B Test
How to Successfully Run Your First Website A/B Test
 
A/B Testing at Pinterest: Building a Culture of Experimentation
A/B Testing at Pinterest: Building a Culture of Experimentation A/B Testing at Pinterest: Building a Culture of Experimentation
A/B Testing at Pinterest: Building a Culture of Experimentation
 
Millward brown vermeer organizational design think brand 2014 snf
Millward brown vermeer organizational design think brand 2014 snfMillward brown vermeer organizational design think brand 2014 snf
Millward brown vermeer organizational design think brand 2014 snf
 
A Crash Course on A/B Testing
A Crash Course on A/B TestingA Crash Course on A/B Testing
A Crash Course on A/B Testing
 
Google Automotive Think Insights - Digital Drives Car Researches and Auto Pur...
Google Automotive Think Insights - Digital Drives Car Researches and Auto Pur...Google Automotive Think Insights - Digital Drives Car Researches and Auto Pur...
Google Automotive Think Insights - Digital Drives Car Researches and Auto Pur...
 
The Operating Model That is Eating the World
The Operating Model That is Eating the WorldThe Operating Model That is Eating the World
The Operating Model That is Eating the World
 
Always Be Testing - Learn from Every A/B Test (Hiten Shah)
Always Be Testing - Learn from Every A/B Test (Hiten Shah)Always Be Testing - Learn from Every A/B Test (Hiten Shah)
Always Be Testing - Learn from Every A/B Test (Hiten Shah)
 

Semelhante a [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test

Getting Strategic with Digital: Intro to Analytics
Getting Strategic with Digital: Intro to AnalyticsGetting Strategic with Digital: Intro to Analytics
Getting Strategic with Digital: Intro to AnalyticsnForm User Experience
 
Culture of Optimization
Culture of OptimizationCulture of Optimization
Culture of OptimizationNikki Johnson
 
How to Increase Your Testing Success by Combining Qualitative and Quantitativ...
How to Increase Your Testing Success by Combining Qualitative and Quantitativ...How to Increase Your Testing Success by Combining Qualitative and Quantitativ...
How to Increase Your Testing Success by Combining Qualitative and Quantitativ...Optimizely
 
Losing is the New Winning
Losing is the New WinningLosing is the New Winning
Losing is the New WinningOptimizely
 
Online Survey Creation Guide
Online Survey Creation GuideOnline Survey Creation Guide
Online Survey Creation GuideSimplyCast
 
Message-Market Fit VWO Workshop by Daphne tideman
Message-Market Fit VWO Workshop by Daphne tidemanMessage-Market Fit VWO Workshop by Daphne tideman
Message-Market Fit VWO Workshop by Daphne tidemanVWO
 
EiR & Engineering Continuing Ed: Understanding the Web Site Feedback Loop
EiR & Engineering Continuing Ed: Understanding the Web Site Feedback LoopEiR & Engineering Continuing Ed: Understanding the Web Site Feedback Loop
EiR & Engineering Continuing Ed: Understanding the Web Site Feedback LoopLaunch Angels
 
Unlock Your Data's Potential By Integrating Qualtrics & Tableau
Unlock Your Data's Potential By Integrating Qualtrics & TableauUnlock Your Data's Potential By Integrating Qualtrics & Tableau
Unlock Your Data's Potential By Integrating Qualtrics & TableauQualtrics
 
VWO Webinar: How To Plan Your Optimisation Roadmap
VWO Webinar: How To Plan Your Optimisation RoadmapVWO Webinar: How To Plan Your Optimisation Roadmap
VWO Webinar: How To Plan Your Optimisation RoadmapVWO
 
Prsa social influence and measurement
Prsa social influence and measurementPrsa social influence and measurement
Prsa social influence and measurementwedu, inc.
 
PRSA - Social Influence and Measurement
PRSA - Social Influence and MeasurementPRSA - Social Influence and Measurement
PRSA - Social Influence and Measurementwedu, Inc
 
Meta-Analyses in Experimentation: The Whats and Hows
Meta-Analyses in Experimentation: The Whats and HowsMeta-Analyses in Experimentation: The Whats and Hows
Meta-Analyses in Experimentation: The Whats and HowsVWO
 
Testing the unknown: the art and science of working with hypothesis
Testing the unknown: the art and science of working with hypothesisTesting the unknown: the art and science of working with hypothesis
Testing the unknown: the art and science of working with hypothesisArdita Karaj
 
How a year of SEO split testing changed how I thought SEO worked
How a year of SEO split testing changed how I thought SEO workedHow a year of SEO split testing changed how I thought SEO worked
How a year of SEO split testing changed how I thought SEO workedDominic Woodman
 
GA_LeanResearch
GA_LeanResearchGA_LeanResearch
GA_LeanResearchLiya James
 
Brighton CRO Meetup #1 - Oh Boy These AB tests Sure Look Like Bullshit to Me
Brighton CRO Meetup #1 - Oh Boy These AB tests Sure Look Like Bullshit to MeBrighton CRO Meetup #1 - Oh Boy These AB tests Sure Look Like Bullshit to Me
Brighton CRO Meetup #1 - Oh Boy These AB tests Sure Look Like Bullshit to MeCraig Sullivan
 
Lessons learned from growing LinkedIn to 400m members - Growth Hackers Confer...
Lessons learned from growing LinkedIn to 400m members - Growth Hackers Confer...Lessons learned from growing LinkedIn to 400m members - Growth Hackers Confer...
Lessons learned from growing LinkedIn to 400m members - Growth Hackers Confer...Aatif Awan
 
Castillo high quality program evaluation in nonprofits
Castillo   high quality program evaluation in nonprofitsCastillo   high quality program evaluation in nonprofits
Castillo high quality program evaluation in nonprofitsIsaac Castillo
 
Accelerating Product Delivery with Design Sprints
Accelerating Product Delivery with Design SprintsAccelerating Product Delivery with Design Sprints
Accelerating Product Delivery with Design SprintsAnthony Broad-Crawford
 
Personalization Strategy Workshop - Los Angeles
Personalization Strategy Workshop - Los AngelesPersonalization Strategy Workshop - Los Angeles
Personalization Strategy Workshop - Los AngelesOptimizely
 

Semelhante a [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test (20)

Getting Strategic with Digital: Intro to Analytics
Getting Strategic with Digital: Intro to AnalyticsGetting Strategic with Digital: Intro to Analytics
Getting Strategic with Digital: Intro to Analytics
 
Culture of Optimization
Culture of OptimizationCulture of Optimization
Culture of Optimization
 
How to Increase Your Testing Success by Combining Qualitative and Quantitativ...
How to Increase Your Testing Success by Combining Qualitative and Quantitativ...How to Increase Your Testing Success by Combining Qualitative and Quantitativ...
How to Increase Your Testing Success by Combining Qualitative and Quantitativ...
 
Losing is the New Winning
Losing is the New WinningLosing is the New Winning
Losing is the New Winning
 
Online Survey Creation Guide
Online Survey Creation GuideOnline Survey Creation Guide
Online Survey Creation Guide
 
Message-Market Fit VWO Workshop by Daphne tideman
Message-Market Fit VWO Workshop by Daphne tidemanMessage-Market Fit VWO Workshop by Daphne tideman
Message-Market Fit VWO Workshop by Daphne tideman
 
EiR & Engineering Continuing Ed: Understanding the Web Site Feedback Loop
EiR & Engineering Continuing Ed: Understanding the Web Site Feedback LoopEiR & Engineering Continuing Ed: Understanding the Web Site Feedback Loop
EiR & Engineering Continuing Ed: Understanding the Web Site Feedback Loop
 
Unlock Your Data's Potential By Integrating Qualtrics & Tableau
Unlock Your Data's Potential By Integrating Qualtrics & TableauUnlock Your Data's Potential By Integrating Qualtrics & Tableau
Unlock Your Data's Potential By Integrating Qualtrics & Tableau
 
VWO Webinar: How To Plan Your Optimisation Roadmap
VWO Webinar: How To Plan Your Optimisation RoadmapVWO Webinar: How To Plan Your Optimisation Roadmap
VWO Webinar: How To Plan Your Optimisation Roadmap
 
Prsa social influence and measurement
Prsa social influence and measurementPrsa social influence and measurement
Prsa social influence and measurement
 
PRSA - Social Influence and Measurement
PRSA - Social Influence and MeasurementPRSA - Social Influence and Measurement
PRSA - Social Influence and Measurement
 
Meta-Analyses in Experimentation: The Whats and Hows
Meta-Analyses in Experimentation: The Whats and HowsMeta-Analyses in Experimentation: The Whats and Hows
Meta-Analyses in Experimentation: The Whats and Hows
 
Testing the unknown: the art and science of working with hypothesis
Testing the unknown: the art and science of working with hypothesisTesting the unknown: the art and science of working with hypothesis
Testing the unknown: the art and science of working with hypothesis
 
How a year of SEO split testing changed how I thought SEO worked
How a year of SEO split testing changed how I thought SEO workedHow a year of SEO split testing changed how I thought SEO worked
How a year of SEO split testing changed how I thought SEO worked
 
GA_LeanResearch
GA_LeanResearchGA_LeanResearch
GA_LeanResearch
 
Brighton CRO Meetup #1 - Oh Boy These AB tests Sure Look Like Bullshit to Me
Brighton CRO Meetup #1 - Oh Boy These AB tests Sure Look Like Bullshit to MeBrighton CRO Meetup #1 - Oh Boy These AB tests Sure Look Like Bullshit to Me
Brighton CRO Meetup #1 - Oh Boy These AB tests Sure Look Like Bullshit to Me
 
Lessons learned from growing LinkedIn to 400m members - Growth Hackers Confer...
Lessons learned from growing LinkedIn to 400m members - Growth Hackers Confer...Lessons learned from growing LinkedIn to 400m members - Growth Hackers Confer...
Lessons learned from growing LinkedIn to 400m members - Growth Hackers Confer...
 
Castillo high quality program evaluation in nonprofits
Castillo   high quality program evaluation in nonprofitsCastillo   high quality program evaluation in nonprofits
Castillo high quality program evaluation in nonprofits
 
Accelerating Product Delivery with Design Sprints
Accelerating Product Delivery with Design SprintsAccelerating Product Delivery with Design Sprints
Accelerating Product Delivery with Design Sprints
 
Personalization Strategy Workshop - Los Angeles
Personalization Strategy Workshop - Los AngelesPersonalization Strategy Workshop - Los Angeles
Personalization Strategy Workshop - Los Angeles
 

Mais de 500 Startups (20)

Get on Board
Get on BoardGet on Board
Get on Board
 
Connected Analytics
Connected AnalyticsConnected Analytics
Connected Analytics
 
Sira Medical
Sira MedicalSira Medical
Sira Medical
 
The Atlas
The AtlasThe Atlas
The Atlas
 
Trash Warrior
Trash WarriorTrash Warrior
Trash Warrior
 
Thematic
ThematicThematic
Thematic
 
Shiplyst
ShiplystShiplyst
Shiplyst
 
Renetec
RenetecRenetec
Renetec
 
Predina
PredinaPredina
Predina
 
Pluto
PlutoPluto
Pluto
 
Plant an App
Plant an AppPlant an App
Plant an App
 
Pilota
PilotaPilota
Pilota
 
Mero Technologies
Mero TechnologiesMero Technologies
Mero Technologies
 
Omnitron Sensors
Omnitron SensorsOmnitron Sensors
Omnitron Sensors
 
Juked
JukedJuked
Juked
 
GamerzClass
GamerzClassGamerzClass
GamerzClass
 
eino
einoeino
eino
 
Cenos
CenosCenos
Cenos
 
Bliinx
BliinxBliinx
Bliinx
 
Butlr
ButlrButlr
Butlr
 

Último

『澳洲文凭』买拉筹伯大学毕业证书成绩单办理澳洲LTU文凭学位证书
『澳洲文凭』买拉筹伯大学毕业证书成绩单办理澳洲LTU文凭学位证书『澳洲文凭』买拉筹伯大学毕业证书成绩单办理澳洲LTU文凭学位证书
『澳洲文凭』买拉筹伯大学毕业证书成绩单办理澳洲LTU文凭学位证书rnrncn29
 
ETHICAL HACKING dddddddddddddddfnandni.pptx
ETHICAL HACKING dddddddddddddddfnandni.pptxETHICAL HACKING dddddddddddddddfnandni.pptx
ETHICAL HACKING dddddddddddddddfnandni.pptxNIMMANAGANTI RAMAKRISHNA
 
Company Snapshot Theme for Business by Slidesgo.pptx
Company Snapshot Theme for Business by Slidesgo.pptxCompany Snapshot Theme for Business by Slidesgo.pptx
Company Snapshot Theme for Business by Slidesgo.pptxMario
 
Unidad 4 – Redes de ordenadores (en inglés).pptx
Unidad 4 – Redes de ordenadores (en inglés).pptxUnidad 4 – Redes de ordenadores (en inglés).pptx
Unidad 4 – Redes de ordenadores (en inglés).pptxmibuzondetrabajo
 
TRENDS Enabling and inhibiting dimensions.pptx
TRENDS Enabling and inhibiting dimensions.pptxTRENDS Enabling and inhibiting dimensions.pptx
TRENDS Enabling and inhibiting dimensions.pptxAndrieCagasanAkio
 
IP addressing and IPv6, presented by Paul Wilson at IETF 119
IP addressing and IPv6, presented by Paul Wilson at IETF 119IP addressing and IPv6, presented by Paul Wilson at IETF 119
IP addressing and IPv6, presented by Paul Wilson at IETF 119APNIC
 
Cybersecurity Threats and Cybersecurity Best Practices
Cybersecurity Threats and Cybersecurity Best PracticesCybersecurity Threats and Cybersecurity Best Practices
Cybersecurity Threats and Cybersecurity Best PracticesLumiverse Solutions Pvt Ltd
 
『澳洲文凭』买詹姆士库克大学毕业证书成绩单办理澳洲JCU文凭学位证书
『澳洲文凭』买詹姆士库克大学毕业证书成绩单办理澳洲JCU文凭学位证书『澳洲文凭』买詹姆士库克大学毕业证书成绩单办理澳洲JCU文凭学位证书
『澳洲文凭』买詹姆士库克大学毕业证书成绩单办理澳洲JCU文凭学位证书rnrncn29
 
SCM Symposium PPT Format Customer loyalty is predi
SCM Symposium PPT Format Customer loyalty is prediSCM Symposium PPT Format Customer loyalty is predi
SCM Symposium PPT Format Customer loyalty is predieusebiomeyer
 

Último (9)

『澳洲文凭』买拉筹伯大学毕业证书成绩单办理澳洲LTU文凭学位证书
『澳洲文凭』买拉筹伯大学毕业证书成绩单办理澳洲LTU文凭学位证书『澳洲文凭』买拉筹伯大学毕业证书成绩单办理澳洲LTU文凭学位证书
『澳洲文凭』买拉筹伯大学毕业证书成绩单办理澳洲LTU文凭学位证书
 
ETHICAL HACKING dddddddddddddddfnandni.pptx
ETHICAL HACKING dddddddddddddddfnandni.pptxETHICAL HACKING dddddddddddddddfnandni.pptx
ETHICAL HACKING dddddddddddddddfnandni.pptx
 
Company Snapshot Theme for Business by Slidesgo.pptx
Company Snapshot Theme for Business by Slidesgo.pptxCompany Snapshot Theme for Business by Slidesgo.pptx
Company Snapshot Theme for Business by Slidesgo.pptx
 
Unidad 4 – Redes de ordenadores (en inglés).pptx
Unidad 4 – Redes de ordenadores (en inglés).pptxUnidad 4 – Redes de ordenadores (en inglés).pptx
Unidad 4 – Redes de ordenadores (en inglés).pptx
 
TRENDS Enabling and inhibiting dimensions.pptx
TRENDS Enabling and inhibiting dimensions.pptxTRENDS Enabling and inhibiting dimensions.pptx
TRENDS Enabling and inhibiting dimensions.pptx
 
IP addressing and IPv6, presented by Paul Wilson at IETF 119
IP addressing and IPv6, presented by Paul Wilson at IETF 119IP addressing and IPv6, presented by Paul Wilson at IETF 119
IP addressing and IPv6, presented by Paul Wilson at IETF 119
 
Cybersecurity Threats and Cybersecurity Best Practices
Cybersecurity Threats and Cybersecurity Best PracticesCybersecurity Threats and Cybersecurity Best Practices
Cybersecurity Threats and Cybersecurity Best Practices
 
『澳洲文凭』买詹姆士库克大学毕业证书成绩单办理澳洲JCU文凭学位证书
『澳洲文凭』买詹姆士库克大学毕业证书成绩单办理澳洲JCU文凭学位证书『澳洲文凭』买詹姆士库克大学毕业证书成绩单办理澳洲JCU文凭学位证书
『澳洲文凭』买詹姆士库克大学毕业证书成绩单办理澳洲JCU文凭学位证书
 
SCM Symposium PPT Format Customer loyalty is predi
SCM Symposium PPT Format Customer loyalty is prediSCM Symposium PPT Format Customer loyalty is predi
SCM Symposium PPT Format Customer loyalty is predi
 

[#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test

  • 1. Measuring For Impact: Knowing What, and How to A/B Test @mike_greenfield CEO/Co-Founder, Laserlike 2014-08-07 @mike_greenfield
  • 2. You know you should A/B test. @mike_greenfield
  • 3. You also know you should exercise more eat less sugar spend less on coffee wear sunscreen etc., etc. @mike_greenfield
  • 4. (Don’t worry, I’m not going to say anything else about sugar or sunscreen.) @mike_greenfield
  • 5. So, how do you create a culture in which people will constructively A/B test? Do six things. @mike_greenfield
  • 6. 1. Embrace “I don’t know” We have 2+ ideas. I don’t know which one will be more effective. @mike_greenfield
  • 8. 2. Have Data, Choose Metrics To test, you need: • People using your product • (Approximate) agreement on the metrics that matter @mike_greenfield
  • 9. Not Many Users? Don’t A/B test! • Laserlike, has ~60 users and has never run an A/B test • We will run many, many tests when we have enough users • A test should have at least a few hundred instances (and a lot more if effect sizes are likely to be small) • Test iff you can have “business significance” @mike_greenfield
  • 10. Know What You Want to Optimize • If it’s important, you should be running tests to improve it • If it’s not important, spend time on other things • Most tests should be aimed at improving 1-2 specific variables @mike_greenfield
  • 11. 3. Have Clear Process, Tech for Testing @mike_greenfield
  • 12. A/B Testing Process • New feature: if possible, roll out to a small test subset first (10s or 100s of thousands) • Version change: always test things that could (cumulatively) have business impact • Everyone on the product team should be running and resolving tests @mike_greenfield
  • 13. A/B Testing Tech • Using a third party testing service is akin to building your site on Wordpress: great at some scales/competency levels • No matter how you’re testing, a new test should be at most a few lines of code • It should be easy to see how each side of a test compares across many variables @mike_greenfield
  • 14. 4. Understand the Math of What to Test @mike_greenfield
  • 15. Process: Same vs. New Tweak • What’s the probability your tweak will have a positive effect? • What kind of effect might that have, and how might that effect change the company’s prospects? • Will you be able to measure the change? • Optimize on one variable, but look at others @mike_greenfield
  • 16. Process: Same vs. Big Change • What’s the probability that your change will have a negative impact? • How big an impact might there be? • Will you be able to measure the change? • Holistic approach @mike_greenfield
  • 17. A/B Test for Quality • Circle of Moms: test “warning” users when questions seemed short, low quality • Resulting questions were graded for quality, without grader knowing test bucket • End result: warning yielded ~5% fewer questions, but much higher quality @mike_greenfield
  • 18. 5. Understand the Math of Picking Winners @mike_greenfield
  • 19. Resolving Too Soon vs. Resolving Too Late • How big is the potential audience for this test? • Example 1: end of year “most popular baby names” email that will never be sent again • Example 2: Facebook signup flow @mike_greenfield
  • 20. Longitudinal Tests vs. Immediate Tests • Longitudinal: change home page, email frequency, product framing • Need to examine effect over a long period • Immediate: change button color, email subject • Likely that long-term effects will be minimal @mike_greenfield
  • 21. Automatically Resolve Tests? • Longitudinal tests should not be automatically resolved • Example: new home page design • Immediate tests can be automatically resolved when speed is important and there is one clear objective function • Example: Circle of Moms email subject optimization @mike_greenfield
  • 22. Choose robust statistics • Bad: # of page views • Good: % of users viewing at least [5, 25, 100] pages • Potentially bad: # of sales (when small) • Potentially good: # of people getting through the second step of a sales funnel @mike_greenfield
  • 23. 6. Celebrate A/B Testing Successes @mike_greenfield