blogsimplified.com

Anyone interested in designing websites that engage users and grow virally would enjoy this presentation.

Discovery is the new cocaine – Presentation Transcript

1. Discovery is the new cocaine Going beyond engagement By: Ming Yeow Ng, Yu-shan Fung (Discoverio) Note: Section on “Flow” was derived from “The Science Of Designing Interactions” (SXSW 2008) by Andreas Weigend (people and data), and Ming Yeow Ng (Discoverio)
2. First off… Why should you care about Discovery?
3. From Wall Street Journal: What is it about a Web site that might make it literally irresistible? • We are programmed for scarcity and can’t dial back even when information is abundant • … (addictive sites) all presented new information that somehow needed to be interpreted… • this triggers a chemical reaction that makes us feel good… Source: WSJ, Why We’re Powerless To Resist Grazing On Endless Web Data March 12, 2008
4. In other words…. Done right, Discovery can make a website irresistible!
5. What drives Discovery? Desire by humans to be a live node in the network We are always… “Constantly scanning the peripheral, always motivated by the desire not to miss opportunities” “Desire for other people recognizing us, make us feel important.” “Services that win are those that help us protect, filter, create meaningful connections ” Linda Stone Source: Etech 2006
6. What is Discovery? Allowing users to: 3. Discover new, valuable information 5. Get Discovered by others 7. Discovering more about themselves
7. Facilitate user towards: Discovery of Information New and Richly interpretable information triggers chemical reactions
8. New information Twitter, Friendfeed: Constant stream of new information
9. Richly Interpretable Facebook Breakups “Who was she?” “Was his girlfriend pretty??” “You mean she is gay???”
10. Helping users to: Get Discovered by Others “Who is paying attention to me?” “Am I being recognized?”
11. Who is paying attention to me? MyBlogLog Twitter Who is checking me out? Who is following me?
12. Allowing users to: Discover More About Themselves “Who am I to others?” “What do I like?” Key here: People do not know what they want!
13. The Old Personality Test The New Personality Test Tickle: What am I? What am I to others? (400 million personality tests taken)
14. What do I like? Amazon : Stylefeeder: Collective Recommendation Analyzing attributes
15. Taking all 3 types together: Facilitate the Discovery Cycle for your users
16. The Discovery Cycle for Your App Engage and allow contribution of information, leaving of footprints, and giving of feedback Throughout the process: Data becomes basis for: 1) Discovery of User learning more 2) System to 3) Other Users Content and Users about himself generate new to discover him Meaningful connections (Keeps him coming back!) insights for user gets established Personalized Recommendations Interactions that lead to reciprocity and interactions Continuous Learning, Contributing, Recognition
17. Done right, this leads to “Intermittent Variable Reinforcement “ • “One of Skinner’s most important discoveries was that behavior that is reinforced intermittently is much more difficult to extinguish than behavior that is reinforced continuously.“ – via Kathy Sierra • \”You are not sure you are going to get a reward every time or how often you will, so you keep pulling that handle.“ – Patricia Wallace on slot machines
18. In other words… Discovery is less about predicting precisely right about what the user wants It is more about the entire user flow of discovery, with all the hits and misses Think window shopping down the mall or walking through an expo hall
19. Discovering Activity by Social Graph My good friend The guy with great events The babe I check out Based on collective behavior Based on live updates and attention data
20. Collective Judging Collective Behavior
21. No concept of individual Personalization
22. Helping users get discovered Clear attention/intention data Crowdvine: Updated profiles, Crowdvine: want-to-meet, fan comments are on front page
23. Connecting based on commonalities Apache Social Network: Open source projects worked on together
24. Exploitation Exploration
25. Connecting based on mutual interest
26. Randomness that shows the new and richly interpretable Flickr: Etsy: Interestingness Many ways to shop 26
27. Make it fun to explore Etsy: 10 different fun ways to discover! (boring categories are at the end)
28. How to create the right user experience? Understanding the elements of “flow” Feedback, Rewards, Lightweight, Reducing uncertainty Inspired by Mihály Csíkszentmihályi
29. Elements of flow (1) Direct and immediate feedback Successes and failures in the course of the activity are apparent, so that behavior can be adjusted as needed
30. Feedback for users to guide their future behavior Maslow Hierarchy Metrics for: Contribution to others Achievements Popularity Wrong-doings Feedback can be by system or can be by other users 30
31. How am I doing? Yelp: popularity, achievement, and contribution
32. Points as indicators of power and desirability Cash = power Value = desirability 32
33. Setting expected “norm” Linkedin Profile Completeness Pointing out “deficiencies” to induce actions
34. Elements of flow (2) Rewards Intrinsically Rewarding Clarity of Reward Recognition for contribution
35. Intrinsically Rewarding Twitter: Self Expression is its own reward
36. Intrinsically Rewarding Twitter: Validation, and Social Interaction are powerful rewards by themselves
37. Clear recognition for intended actions Crowdvine: Digg: Updated profiles, comments Made popular by… are on front page
38. Making Rewards Clear and Explicit Best Match: Invite 10 to see “best matches”
39. Suggest action and clarify reward GetSatisfaction: I have this question too!
40. Elements of Flow (3) Lightweight Interactions Make it easy for users to contribute
41. Lightweight, In the moment expressions Crowdvine: Yelp: Fan : expression of confidence Variety of lightweight Want-to-meet: initiate relationship interactions directly on reviews
42. Make user do as little as possible Automated retrieval of external data, all user needs to do is confirm Signup Time 15m -> 2m Conversion 0.1% -> 4% Lijit: Made the process entertaining and a little mysterious, with rewards at each step Source: Lijit, SXSW 2008
43. Do not make users think Woot: One Single Action
44. Elements of Flow (4) Reducing fear and uncertainty Uncertainty, even irrelevant uncertainty kills action
45. Assurance that it is up to recipients’ “mode of receiving” Twitter: Communicating with little fear of incurring cost to recipient
46. Reducing Personal Commitment Zoosk: Flirt vs risk
47. Surfacing Improbable Matches HotorNot: No risk of “rejection” 47
48. Reducing uncertainty on “correct” behavior Latest Comments Crowdvine: Showing social behavior right on frontpage
49. Elements of Addiction Day Trading MMORPG 1) Attractor: Things happening outside your control in the system, yet affecting your status 3) Motivator: You have a stake (self- esteem/emotional/financial) in changes happening to your status Source: \”Ariadne – Understanding MMORPG Addiction\” Nick Yee
50. Systematically Introducing Discovery Into Your App (uncompleted) 1. Know what data points you need (Individual / Collective / External) 3. Know what mode of discovery (Recommendations / Search / Browsing) 5. Create pathways of discovery (Filtering / Anchoring ) 7. Maximize exploration / exploitation within the pathway (Intention / Attention) 9. Providing reasons/context for discovery ( Context/Persona/Content) 11. Facilitating mediums for connections (Currency / Reputation) 13. Understand what metrics you need (Diversity, Virality, Engagement) Process by Ming Yeow Ng, Yu-shan Fung (Discoverio)
51. We LOVE Feedback!!! • Discovery is a new, unexplored and very exciting area to be innovating in, and we love to hear from anyone who is just as excited about this as us! • Touch base with us via any of these mechanisms – mingyeow@discoverio.com | @mingyeow (twitter) – yushan@discoverio.com | @ambivalence (twitter)