Five best mistakes we've made
Download as PPTX, PDF7 likes3,419 views
The document discusses key learnings from social media practices, emphasizing that the best ideas are open to interpretation and not forced. It highlights issues with statistics and self-reported behaviors in social media usage, indicating that many common metrics can be misleading. Additionally, it notes that poor-quality content can still be effective, as tribalism and identity often influence engagement more than content quality.
Five best mistakes we've made
- 1. #OgilvyDS Getting it wrong Rob Blackie, Director of Social Leo Ryan, Global Head of Social Innovations & Partnerships Follow us @robblackie_oo and @Leotwit
- 3. #OgilvyDS ‘CAN SQUIRRELS GET ADDICTED TO CRACK?’
- 4. #OgilvyDS
- 5. #OgilvyDS
- 6. #OgilvyDS
- 7. #OgilvyDS
- 8. #OgilvyDS
- 9. #OgilvyDS WE’VE LEARNT • The best ideas are open to interpretation – not forced or bribed collaboration
- 10. #OgilvyDS WE’VE LEARNT • Accidents happen – so volume helps you get a hit
- 11. #OgilvyDS WE’VE LEARNT • Huge ethical issues around jokes that might be interpreted as facts
- 12. #OgilvyDS Mary Meeker’s stats Don’t trust numbers unless you understand them
- 14. #OgilvyDS THIS IS STRANGE…. • Messages = • 700m photos uploaded & sent PLUS • 500m Stories consumed
- 15. #OgilvyDS THIS IS STRANGE…. • Equivalent Facebook figures: • 400m photos PLUS • 50bn impressions Facebook Messages / Day = 50.4bn
- 16. #OgilvyDS THE GOOD NEWS We started taking Snapchat seriously, even if the numbers were flawed.
- 17. #OgilvyDS OTHER STATISTICS PROBLEMS • Small samples sizes – less than 1,000 is rarely a good idea.
- 18. #OgilvyDS OTHER STATISTICS PROBLEMS • Direction of causation – did people watch your video because they liked the brand or vice versa? Source: Making an impact: Measuring the value of Facebook video
- 19. #OgilvyDS OTHER STATISTICS PROBLEMS • Self reported behaviours – do people know what they do? Will they be honest? Students report 145 minutes per day on Facebook, actually spending 26 minutes per day Source: Comparing actual and self-reported measures of Facebook use, Computers in Human Behaviour 29
- 20. #OgilvyDS OTHER STATISTICS PROBLEMS • Figures from social networks – self-serving, misleading, rarely useful Source: qz.com / Twitter
- 21. #OgilvyDS OTHER STATISTICS PROBLEMS • Cross-country engagement scores – huge differences in behaviours driven by things as simple as the most common devices used • Yet virtually all studies cited in the Social industry are based on the US
- 23. #OgilvyDS THIS IMAGE IS TERRIBLE
- 25. #OgilvyDS SO I REPEATED THE TEST…
- 26. #OgilvyDS WE’VE LEARNT • Tribalism & identity often trump quality
- 27. #OgilvyDS Thank You Rob Blackie, Director of Social Leo Ryan, Global Head of Social Innovations & Partnerships Follow us @robblackie_oo and @Leotwit
Editor's Notes
- #9: https://www.youtube.com/watch?v=E11epT9z1Eg&index=2&list=RDCO0yAuce9Hk