BEST PRACTICES IN DIGITAL ANALYTICS:
Optimizing For Virality
With Website And
Social Media A/B Testing
OUR STORIES OF SELF
Jim Angelica
WHAT IS VIRALITY?
• Often used, poorly understood:
 “Make me a viral video!”
 “There’s a viral campaign happening on Twitter”
 “Social media goes viral on Capitol Hill”
• Virality measures organic growth, not overall popularity
• Determined by how much your users are sharing with friends
0
100
200
300
400
500
600
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Users
Time
Sub-Viral
0
5000
10000
15000
20000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Users
Time
Viral
WHAT IS VIRALITY?
• Virality Ratio – average number of new people each user recruits
• Definition of “viral”: Virality ratio greater than 1 – explosive growth
• Virality ratio less than 1 is sub-viral – growth is limited
VIRALITY EXAMPLES
VIRALITY
RATIO
INITIAL
USERS
(GEN 1)
RECRUITE
D USERS
(GEN 2)
RECRUITE
D USERS
(GEN 3)
USERS AFTER 10
GENERATIONS
0.01 5,000 50 1 5,051
0.1 5,000 500 50 5,556
0.5 5,000 2,500 1,250 9,990
0.9 5,000 4,500 4,050 32,566
1.1 5,000 5,500 6,050 79,687
Action
Invite
Response
Action Rate x Invite Rate x Response Rate =
Virality Ratio
THE VIRAL LOOP
• Users take action
• Users invite friends
• Friends respond, become users
• Repeat
IMPROVING VIRALITY
• "Ok, I understand what virality is -- but how can I increase it?"
• Remember: Virality Ratio = Action Rate x Invite Rate x Response
Rate
• Improving rate of any step improves virality ratio by same amount
A BRIEF INTRODUCTION
TO A/B TESTING
• Q: You have two ideas for language on a web page –
which one do you choose?
• A: Both!
• Steps to run a webpage A/B test:
– Create several different versions of the web page
– For users coming to page, randomly show one version
– Record if visitor takes the action
– Calculate conversion rate (average # of actions per
visitors) for each version – highest conversion rate is best
29% increased
conversion rate
WEB PAGE A/B TESTING
+140% share
recruitment rate
SOCIAL MEDIA A/B TESTING
STATISTICAL SIGNIFICANCE
VIRALITY OPTIMIZATION
EXAMPLE #1
• New campaign has virality ratio of 0.1 – viral actions increase total
action count by 11%
• Virality loop optimizations:
 Optimize action page -> 20% higher action rate
 Use optimized share page -> 60% more sharing to friends
 Test share language -> 35% higher click rate
• New virality ratio: 0.1 x 1.20 x 1.60 x 1.35 = 0.26
• Viral actions increase total action count by 35%
• New campaign has virality ratio of 0.4 – viral actions increase total
action count by 67%
• Virality loop optimizations:
 Optimize action page -> 20% higher conversion rate
 Use optimized share page -> 60% more sharing to friends
 Test share language -> 35% higher click rate
• New virality ratio: 0.4 x 1.20 x 1.60 x 1.35 = 1.04
• VIRAL CAMPAIGN
VIRALITY OPTIMIZATION
EXAMPLE #2
CASE STUDY
CREDO ACTION PETITION:
“STOP THE MONSANTO PROTECTION ACT”
CASE STUDY
CREDO ACTION PETITION:
“STOP THE MONSANTO PROTECTION ACT”
RESULTS
• The average petition gets less than 2% of names from social
sharing
• CREDO received 13% of signatures from social sharing, resulting
in 25,000 new names on their list
• How did they do this? CREDO optimized their petition page, used
pre-optimized share pages, and optimized their share language
asks
CASE STUDY
CREDO ACTION PETITION:
“STOP THE MONSANTO PROTECTION ACT”
OPTIMIZED PETITION PAGE
• Multiple rounds of A/B testing have improved the action rate for
the petition page where people add their name
OPTIMIZED SHARE PAGE
• CREDO used a pre-optimized ShareProgress share page for the
petition to increase the sharing rates for signers
OPTIMIZING SHARE LANGUAGE
• CREDO tested share language to turn more shares into viral
actions
• One Facebook language variation performed 63% better that the
first version, increasing overall viral actions by 35%
Facebook Title
Shares Driving Action /
Total Shares
Success Rate Improvement
Monsanto shouldn’t be
above the law
341/1041
32.8% (±2.9%)
--
“One of the most outrageous
special interest provisions in
years”
322/962
33.5% (±3.0%)
2.2%
Tell the Senate: Repeal the
Monsanto Protection Act
458/972
47.1% (±3.1%)
43.8%
Sign the Petition: Repeal the
Monsanto Protection Act
555/1037 53.5% (±3.0%) 63.4%
OVERALL RESULTS
• Virality loop optimizations:
 Optimize action page -> 20% higher conversion rate
 Plug in ShareProgress share page -> 50% more sharing to friends
 Test share language -> 35% higher social conversions
• Result: Virality ratio of 0.135 (13.5% of actions from social sharing)
WHAT WOULD HAVE HAPPENED
WITHOUT OPTIMIZATION?
With Optimization:
Virality ratio: 0.135
35,000 actions from social
25,000 new-to-list
Without Optimization (estimated):
Virality Ratio: 0.056
13,000 actions from social – 22,000 fewer
9,300 new-to-list – 15,700 fewer
QUESTIONS?
VIRALITY CHALLENGE
• http://www.shareprogress.org/virality-challenge/
• You need to optimize the action page, share page, and social
sharing content to make your petition go viral.
• Only one combination of parameter choices will result in a viral
campaign!
• Check your statistical significance at
http://www.thumbtack.com/labs/abba/.

More Related Content

PDF
Mohan Sawhney - Transforming Marketing in a Digital World
PPT
What Is Brickfish?
PPT
LA event: December 11, 2008: Digital Tools for LA non profit workshop
PPTX
Mobile app ysmc
PPTX
Harshal personallearningassi.ppt
PPTX
Going Mobile: Optimizing Your Digital Communications
PDF
IncentiveWorks 2012 - Do-it-yourself Mobile Apps
PDF
The influence engine managing a brand's online ecology
Mohan Sawhney - Transforming Marketing in a Digital World
What Is Brickfish?
LA event: December 11, 2008: Digital Tools for LA non profit workshop
Mobile app ysmc
Harshal personallearningassi.ppt
Going Mobile: Optimizing Your Digital Communications
IncentiveWorks 2012 - Do-it-yourself Mobile Apps
The influence engine managing a brand's online ecology

Viewers also liked (17)

PPTX
How to Optimize Your Website to Reach Your Goals -- Netroots Nation 2016, Cin...
PPT
Presenza online delle aziende
PDF
SEO for Social Change -- Netroots 2016, Debra Cleaver and Jim Pugh
DOC
Resume-sureshkumar
PDF
Note d’orientation pour le développement d’une stratégie nationale d’apprenti...
PDF
Note d’orientation pour le développement d’une stratégie nationale d’apprenti...
DOC
Ziklo amaierako ebaluazio txostena jatorrizkoa
PPTX
C omercio electrónico
PPTX
Crisis parlamentarismo y alessandri psu
PPTX
Encuestas y Programas de Gob
PPTX
Unidad 2 Cs Sociales
PPTX
Ppt transformaciones sociales Parte II
DOC
Metal arotzeriako eta metal eraikuntzen muntaiako langilea (jatorrizkoa)
PDF
Web site goals & objectives
PDF
Melanie consoli resume_february-2013
PDF
Top 5 Reasons to Automate LA
PPTX
Ppt ibañez 1927-1931
How to Optimize Your Website to Reach Your Goals -- Netroots Nation 2016, Cin...
Presenza online delle aziende
SEO for Social Change -- Netroots 2016, Debra Cleaver and Jim Pugh
Resume-sureshkumar
Note d’orientation pour le développement d’une stratégie nationale d’apprenti...
Note d’orientation pour le développement d’une stratégie nationale d’apprenti...
Ziklo amaierako ebaluazio txostena jatorrizkoa
C omercio electrónico
Crisis parlamentarismo y alessandri psu
Encuestas y Programas de Gob
Unidad 2 Cs Sociales
Ppt transformaciones sociales Parte II
Metal arotzeriako eta metal eraikuntzen muntaiako langilea (jatorrizkoa)
Web site goals & objectives
Melanie consoli resume_february-2013
Top 5 Reasons to Automate LA
Ppt ibañez 1927-1931
Ad

Similar to Optimizing for Virality -- Netroots 2016, Jim Pugh (20)

PPT
Design Learnings From Viral Applications Presentation
PPT
Design Learnings From Viral Applications Presentation
PPT
Design Learnings From Viral Applications Presentation
PDF
Does Fear-based Marketing Work: How one company saw a $45 million increase in...
PPTX
"Rise of the Social Business" - Customer Engagement Network Forum, March 2013
PDF
Digital influence 501: What are the next trends and hot cases in social media
PPT
Prove It! Tools and Techniques to Measure the Effectiveness of Social Media
PDF
Unruly - NOAH13 London
PPTX
Commonsense Social Media Measurement
PPTX
DataScienceInnovation_ShareThis
PDF
Squeezing the Most out of a Click: Local Funnel Optimization by Evan Kramer, ...
PDF
E xtension 2011 fsa cop social media project-05-11
PPTX
Facebook: How to maximise your influence and reach
PPT
RockYou! Snap Summit 3-25-08
PPTX
Mobile Marketing Strategies for 2014 and Beyond
PDF
Google Analytics 101
PDF
Communicating Sustainability using Social Media - Agrion Webinar by @JoeyShepp
PDF
Lithuania Krds How Facebook Is Changing The Way We Interact With Brands -...
PDF
PDF
Social Media Boot Camp Boston 2
Design Learnings From Viral Applications Presentation
Design Learnings From Viral Applications Presentation
Design Learnings From Viral Applications Presentation
Does Fear-based Marketing Work: How one company saw a $45 million increase in...
"Rise of the Social Business" - Customer Engagement Network Forum, March 2013
Digital influence 501: What are the next trends and hot cases in social media
Prove It! Tools and Techniques to Measure the Effectiveness of Social Media
Unruly - NOAH13 London
Commonsense Social Media Measurement
DataScienceInnovation_ShareThis
Squeezing the Most out of a Click: Local Funnel Optimization by Evan Kramer, ...
E xtension 2011 fsa cop social media project-05-11
Facebook: How to maximise your influence and reach
RockYou! Snap Summit 3-25-08
Mobile Marketing Strategies for 2014 and Beyond
Google Analytics 101
Communicating Sustainability using Social Media - Agrion Webinar by @JoeyShepp
Lithuania Krds How Facebook Is Changing The Way We Interact With Brands -...
Social Media Boot Camp Boston 2
Ad

Recently uploaded (20)

PDF
sbt 2.0: go big (Scala Days 2025 edition)
PDF
Consumable AI The What, Why & How for Small Teams.pdf
PPTX
The various Industrial Revolutions .pptx
PDF
Developing a website for English-speaking practice to English as a foreign la...
PDF
STKI Israel Market Study 2025 version august
PDF
UiPath Agentic Automation session 1: RPA to Agents
PPTX
Chapter 5: Probability Theory and Statistics
PDF
A comparative study of natural language inference in Swahili using monolingua...
PDF
Zenith AI: Advanced Artificial Intelligence
PDF
Abstractive summarization using multilingual text-to-text transfer transforme...
PDF
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
PDF
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
PDF
Convolutional neural network based encoder-decoder for efficient real-time ob...
PDF
Enhancing emotion recognition model for a student engagement use case through...
PDF
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
PDF
Architecture types and enterprise applications.pdf
PDF
OpenACC and Open Hackathons Monthly Highlights July 2025
PDF
Five Habits of High-Impact Board Members
PPT
Galois Field Theory of Risk: A Perspective, Protocol, and Mathematical Backgr...
PPTX
2018-HIPAA-Renewal-Training for executives
sbt 2.0: go big (Scala Days 2025 edition)
Consumable AI The What, Why & How for Small Teams.pdf
The various Industrial Revolutions .pptx
Developing a website for English-speaking practice to English as a foreign la...
STKI Israel Market Study 2025 version august
UiPath Agentic Automation session 1: RPA to Agents
Chapter 5: Probability Theory and Statistics
A comparative study of natural language inference in Swahili using monolingua...
Zenith AI: Advanced Artificial Intelligence
Abstractive summarization using multilingual text-to-text transfer transforme...
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
Convolutional neural network based encoder-decoder for efficient real-time ob...
Enhancing emotion recognition model for a student engagement use case through...
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
Architecture types and enterprise applications.pdf
OpenACC and Open Hackathons Monthly Highlights July 2025
Five Habits of High-Impact Board Members
Galois Field Theory of Risk: A Perspective, Protocol, and Mathematical Backgr...
2018-HIPAA-Renewal-Training for executives

Optimizing for Virality -- Netroots 2016, Jim Pugh

  • 1. BEST PRACTICES IN DIGITAL ANALYTICS: Optimizing For Virality With Website And Social Media A/B Testing
  • 2. OUR STORIES OF SELF Jim Angelica
  • 3. WHAT IS VIRALITY? • Often used, poorly understood:  “Make me a viral video!”  “There’s a viral campaign happening on Twitter”  “Social media goes viral on Capitol Hill” • Virality measures organic growth, not overall popularity • Determined by how much your users are sharing with friends
  • 4. 0 100 200 300 400 500 600 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Users Time Sub-Viral 0 5000 10000 15000 20000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Users Time Viral WHAT IS VIRALITY? • Virality Ratio – average number of new people each user recruits • Definition of “viral”: Virality ratio greater than 1 – explosive growth • Virality ratio less than 1 is sub-viral – growth is limited
  • 5. VIRALITY EXAMPLES VIRALITY RATIO INITIAL USERS (GEN 1) RECRUITE D USERS (GEN 2) RECRUITE D USERS (GEN 3) USERS AFTER 10 GENERATIONS 0.01 5,000 50 1 5,051 0.1 5,000 500 50 5,556 0.5 5,000 2,500 1,250 9,990 0.9 5,000 4,500 4,050 32,566 1.1 5,000 5,500 6,050 79,687
  • 6. Action Invite Response Action Rate x Invite Rate x Response Rate = Virality Ratio THE VIRAL LOOP • Users take action • Users invite friends • Friends respond, become users • Repeat
  • 7. IMPROVING VIRALITY • "Ok, I understand what virality is -- but how can I increase it?" • Remember: Virality Ratio = Action Rate x Invite Rate x Response Rate • Improving rate of any step improves virality ratio by same amount
  • 8. A BRIEF INTRODUCTION TO A/B TESTING • Q: You have two ideas for language on a web page – which one do you choose? • A: Both! • Steps to run a webpage A/B test: – Create several different versions of the web page – For users coming to page, randomly show one version – Record if visitor takes the action – Calculate conversion rate (average # of actions per visitors) for each version – highest conversion rate is best
  • 12. VIRALITY OPTIMIZATION EXAMPLE #1 • New campaign has virality ratio of 0.1 – viral actions increase total action count by 11% • Virality loop optimizations:  Optimize action page -> 20% higher action rate  Use optimized share page -> 60% more sharing to friends  Test share language -> 35% higher click rate • New virality ratio: 0.1 x 1.20 x 1.60 x 1.35 = 0.26 • Viral actions increase total action count by 35%
  • 13. • New campaign has virality ratio of 0.4 – viral actions increase total action count by 67% • Virality loop optimizations:  Optimize action page -> 20% higher conversion rate  Use optimized share page -> 60% more sharing to friends  Test share language -> 35% higher click rate • New virality ratio: 0.4 x 1.20 x 1.60 x 1.35 = 1.04 • VIRAL CAMPAIGN VIRALITY OPTIMIZATION EXAMPLE #2
  • 14. CASE STUDY CREDO ACTION PETITION: “STOP THE MONSANTO PROTECTION ACT”
  • 15. CASE STUDY CREDO ACTION PETITION: “STOP THE MONSANTO PROTECTION ACT”
  • 16. RESULTS • The average petition gets less than 2% of names from social sharing • CREDO received 13% of signatures from social sharing, resulting in 25,000 new names on their list • How did they do this? CREDO optimized their petition page, used pre-optimized share pages, and optimized their share language asks
  • 17. CASE STUDY CREDO ACTION PETITION: “STOP THE MONSANTO PROTECTION ACT” OPTIMIZED PETITION PAGE • Multiple rounds of A/B testing have improved the action rate for the petition page where people add their name
  • 18. OPTIMIZED SHARE PAGE • CREDO used a pre-optimized ShareProgress share page for the petition to increase the sharing rates for signers
  • 19. OPTIMIZING SHARE LANGUAGE • CREDO tested share language to turn more shares into viral actions • One Facebook language variation performed 63% better that the first version, increasing overall viral actions by 35% Facebook Title Shares Driving Action / Total Shares Success Rate Improvement Monsanto shouldn’t be above the law 341/1041 32.8% (±2.9%) -- “One of the most outrageous special interest provisions in years” 322/962 33.5% (±3.0%) 2.2% Tell the Senate: Repeal the Monsanto Protection Act 458/972 47.1% (±3.1%) 43.8% Sign the Petition: Repeal the Monsanto Protection Act 555/1037 53.5% (±3.0%) 63.4%
  • 20. OVERALL RESULTS • Virality loop optimizations:  Optimize action page -> 20% higher conversion rate  Plug in ShareProgress share page -> 50% more sharing to friends  Test share language -> 35% higher social conversions • Result: Virality ratio of 0.135 (13.5% of actions from social sharing)
  • 21. WHAT WOULD HAVE HAPPENED WITHOUT OPTIMIZATION? With Optimization: Virality ratio: 0.135 35,000 actions from social 25,000 new-to-list Without Optimization (estimated): Virality Ratio: 0.056 13,000 actions from social – 22,000 fewer 9,300 new-to-list – 15,700 fewer
  • 23. VIRALITY CHALLENGE • http://www.shareprogress.org/virality-challenge/ • You need to optimize the action page, share page, and social sharing content to make your petition go viral. • Only one combination of parameter choices will result in a viral campaign! • Check your statistical significance at http://www.thumbtack.com/labs/abba/.

Editor's Notes

  • #15: Help progressive organizations to take advantage of powerful, new social sharing tools and get a much wider reach for their actions and content Build a profitable and scalable company Help support and grow the space of progressive for-profit companies
  • #16: Help progressive organizations to take advantage of powerful, new social sharing tools and get a much wider reach for their actions and content Build a profitable and scalable company Help support and grow the space of progressive for-profit companies
  • #17: - Typical Petition does 1-2% from social actions versus 13% on this one.
  • #21: - Typical Petition does 1-2% from social actions versus 13% on this one.
  • #22: - Typical Petition does 1-2% from social actions versus 13% on this one.
  • #24: - Typical Petition does 1-2% from social actions versus 13% on this one.