Designing for Context
The chance to design for 43 different languages and 200+ countries comes with the challenge of balancing a lot of information per pixel. Booking.com gets millions of users everyday with different needs and goals, and designers need to understand how to best approach them without bringing more complexity to the product. In this talk we will go through how Booking.com deals with that problem using design to best leverage Machine Learning with personas, user journeys and AI to build a personalized experience to our customers. The difference between designing for everyone and designing for context is in the understanding of who your customers are, their intent with your product and what can we do to anticipate their needs without disrupting the experience. Data Science is a powerful tool to get to that goal, but without a well thought of design behind it the intended experience will probably not be fulfilled.
Designing for Context
- 2. oi
- 3. oi
- 4. hi.
- 8. languages 43
- 10. empathy how the machine learns where is machine learning UX + ML UX + ML at booking.com
- 11. empathy
- 12. Empathy is patiently and sincerely seeing the world through the other person’s eyes. It is not learned in school; it is cultivated over a lifetime." Albert Einstein
- 13. machine learning
- 14. Machine learning is an application of arti?cial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed.
- 17. machine learning google’s self driving car
- 18. machine learning skin cancer detection
- 19. ?Any sufficiently advanced technology is indistinguishable from magic " Arthur C. Clarke
- 21. machine learning what is an apple? apple
- 22. machine learning not an apple what is an apple?
- 23. machine learning apple what is an apple?
- 24. machine learning what’s an apple? not an apple
- 25. machine learning
- 27. machine learning
- 35. training set features label apple color red/green texture Shape round taste
- 36. training set features label
- 37. training set features label dates platform time of day ip …
- 38. training set features label date flexibility dates platform time of day ip …
- 39. yes is this an apple? models
- 40. yes is this an apple? models are you flexible on dates? yes
- 41. apple is this an apple, peach or banana? models
- 42. apple is this an apple, peach or banana? models are you more price, location or quality sensitive? price
- 43. based on this apple, here are others you might like models
- 44. based on this apple, here are others you might like models based on the destination you searched for, here are others you might like London paris ?
- 45. based on this apple, here are others you might like models based on the destination you searched for, here are others you might like London paris Amsterdam
- 46. based on this apple, here are others you might like models based on the destination you searched for, here are others you might like nyc ?
- 47. based on this apple, here are others you might like models based on the destination you searched for, here are others you might like nyc Brooklyn
- 48. travel world Predict a fact about the future Where do customers travel after going to rio de janeiro? Guess a fact about the present What dates are current visitors likely to travel on? Automate human tasks Fraud detection, photo tagging
- 49. user experience
- 52. patterns
- 56. booking.com
- 57. empathy
- 58. ML can’t learn empathy because humans are too chaotic, and that’s where UX can help, by bringing the human empathy and chaos to the organized, patterned thinking of a machine.