Simo Ahava slides
•
0 likes•1,777 views
The document discusses various topics related to data measurement and analytics, including the importance of data quality, communication structures, data pipelines, integrating multiple data sources, and fundamentals. It notes that data quality is an investment that requires maturity over time. Organization communication structures greatly influence system designs. Elaborate data pipelines are still susceptible to issues like conflicts. No single data source exists in isolation from others. Even fundamentals of analytics like sessions can be complex and interesting. Customization must be balanced with maintaining data quality.
Report
Share
Simo Ahava slides
- 1. You can’t spell MEASURE …without CUSTOMIZATION @SimoAhava from @8_bit_sheep
- 2. You can’t spell MEASURE …without CUSTOMIZATION @SimoAhava from @8_bit_sheep
- 3. Simo Ahava Partner, co-founder, 8-bit-sheep Google Developer Expert, Google Analytics Blogger, developer, www.simoahava.com Twitter-er, @SimoAhava
- 4. @SimoAhava from @8_bit_sheep processes are clay
- 5. @SimoAhava from @8_bit_sheep processes are clay and you are the sculptor
- 6. How can data be(come) meaningful? Tool expertise Quick wins. Low-hanging fruit. Parlor tricks. Tactics.
- 7. How can data be(come) meaningful? Biz expertise Measurement plans. Solutionism. Vacuous hand-waving. Strategies.
- 8. How can data be(come) meaningful? Tool expertise Biz expertise Purposeful execution. Tactics and goals aligned. Territorialism. Finger-pointing.
- 9. How can data be(come) meaningful? Tool expertise Biz expertise Maturity Holy grail Communication. Participation. Transparency. Shared goals. Agility. Purpose.
- 10. Any organization that designs a system (defined broadly) will produce a design whose structure is a copy of the organization's communication structure. - (Mel) Conway’s law Good data quality is an investment.
- 11. IT UX Data Biz Marketing The Project WORST: Lack of involvement, lack of participation, lack of collaboration, lack of trust. Encourage multi-disciplinary teams.
- 12. IT UX Marketing BETTER: Good communication, lack of shared goals, lack of involvement. Data Biz The Project Encourage multi-disciplinary teams.
- 13. BEST: Transparency, involvement, shared knowledge and skills, constant learning. Encourage multi-disciplinary teams. The Project
- 14. A consultant should work in a manner that inevitably makes them redundant. https://goo.gl/aPvVjb
- 15. BAD: Experts are hired as extra pairs of hands and as reporting tools. The Project Hire for education, not for delegation.
- 16. BEST: Experts work with the team, teaching, facilitating, building processes. The Project Hire for education, not for delegation.
- 17. @SimoAhava from @8_bit_sheep complexity is sea water
- 18. @SimoAhava from @8_bit_sheep complexity is sea water and you are the desalination plant
- 19. communication breakdown results in rusty pipelines and debt
- 21. Naming conventions, unified IDs, authentication, privacy
- 22. Batch or stream, misaligned schemata, chronology issues, hit validation (+ previous)
- 23. Separate ETL processes, patch or discard invalid data (+ previous)
- 24. Aggregation of pipelines (+ previous)
- 25. Visualizing what, when, and how (+ previous)
- 26. each individual component is susceptible to conflict, too
- 27. What is purposeful? What is responsible? What is relevant? What is actionable? What is discardable? What is customizable?
- 28. What is purposeful? What is responsible? What is relevant? What is actionable? What is discardable? What is customizable?
- 29. What can be customized? Visibility Interaction Dwell time Short-term goals Long-term goals Qualitative
- 30. Page visibility Content visibility What can be customized? Visibility Interaction Dwell time Short-term goals Long-term goals Qualitative
- 31. I love you Click Scroll Keypress Long press Drag Resize Highlight Copy What can be customized? Visibility Interaction Dwell time Short-term goals Long-term goals Qualitative
- 32. I love you SERP bounce Idle time Active time Read time What can be customized? Visibility Interaction Dwell time Short-term goals Long-term goals Qualitative
- 33. I love you Mailing list Favorite Print Comment Share What can be customized? Visibility Interaction Dwell time Short-term goals Long-term goals Qualitative
- 34. I love you $ Followers More RPU Recruitment Better rankings What can be customized? Visibility Interaction Dwell time Short-term goals Long-term goals Qualitative
- 35. I love you $ Buzz Feedback Session rec. Heatmaps Polls What can be customized? Visibility Interaction Dwell time Short-term goals Long-term goals Qualitative
- 36. I love you $ What can be customized? Visibility Interaction Dwell time Short-term goals Long-term goals Qualitative
- 38. Google Analytics Session Group of interactions that takes place on a website
- 39. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction
- 40. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone
- 41. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone reacts to campaign changes
- 42. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone reacts to campaign changes referral exclusion override
- 43. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone reacts to campaign changes referral exclusion override manual override with session control
- 44. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone reacts to campaign changes referral exclusion override manual override with session control interactive vs. non-interactive hits
- 45. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone reacts to campaign changes referral exclusion override manual override with session control interactive vs. non-interactive hits requires persistent storage
- 46. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone reacts to campaign changes referral exclusion override manual override with session control interactive vs. non-interactive hits requires persistent storage cross-domain tracking
- 47. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone reacts to campaign changes referral exclusion override manual override with session control interactive vs. non-interactive hits requires persistent storage cross-domain tracking campaign timeout resets session
- 48. Google Analytics Session Group of interactions that takes place on a website expiration after 30 minutes of non-interaction midnight per configured timezone reacts to campaign changes referral exclusion override manual override with session control interactive vs. non-interactive hits requires persistent storage cross-domain tracking campaign timeout resets session
- 49. the schema is inflexible but your application thereof is not
- 50. sessionized the schema is inflexible but your application thereof is not
- 51. @SimoAhava from @8_bit_sheep spare the customization
- 52. @SimoAhava from @8_bit_sheep spare the customization spoil the data quality
- 54. Dev UX Marketing Data Biz There is a flaw in the current model. Impressions are tracked for views that did not happen. Identify a flaw in the status quo.
- 55. Dev UX Marketing Data Biz Identify a flaw in the status quo. Measuring qualified conversions of individual creatives has always been a pain point. Introduce the business problem.
- 56. Dev UX Marketing Data Biz Identify a flaw in the status quo. We can use more accurate CTR reports to better design the creatives and their context in our campaigns. Introduce the business problem. Explore the impact of the change.
- 57. Dev UX Marketing Data Biz Identify a flaw in the status quo. It’s doable with our current Ecommerce integration combined with scroll and visibility tracking. Introduce the business problem. Explore the impact of the change. Produce a technological solution.
- 58. Dev UX Marketing Data Biz Identify a flaw in the status quo. Introduce the business problem. Explore the impact of the change. Produce a technological solution. This will help us choose between lazy/infinite load and pagination of impressions, too! Uncover fringe benefits.
- 60. Align measurement across data sources. Adjust enrichment and ETL to fit data into schema. Design the pipeline flow with queries, reports, and visualizations as the drivers.
- 62. Triggers
- 70. Views Clicks CTR Cart adds Purchases Revenue Quick Buy Retargeting Footer Offers Promo_Jan X-sell 62,065 46,281 33,237 26,412 21,745 21,045 1,472 6,162 581 3,181 520 1,270 2.37% 13.31% 1.75% 12.04% 2.39% 5.85% 1,546 58 0 12 306 1,259 594 17 0 3 137 297 €297,638.42 €5,605.92 €0.00 €1,485.92 €11,234.30 €157,169.30
- 71. Views Clicks CTR Cart adds Purchases Revenue Quick Buy Retargeting Footer Offers Promo_Jan X-sell 62,065 46,281 33,237 26,412 21,745 21,045 1,472 6,162 581 3,181 520 1,270 2.37% 13.31% 1.75% 12.04% 2.39% 5.85% 1,546 58 0 12 306 1,259 594 17 0 3 137 297 €297,638.42 €5,605.92 €0.00 €1,485.92 €11,234.30 €157,169.30 True Qualified Qualified Qualified Qualified Qualified
- 72. @SimoAhava from @8_bit_sheep data is difficult
- 73. @SimoAhava from @8_bit_sheep data is difficult data quality is earned, not acquired
- 74. 1. Data quality is an investment Tool expertise Biz expertise Maturity Holy grail Key takeaways:
- 75. 2. Communication structures are everything Any organization that designs a system (defined broadly) will produce a design whose structure is a copy of the organization's communication structure. - (Mel) Conway’s law Key takeaways:
- 76. 3. Elaborate pipelines are not rust-proof Key takeaways:
- 77. 4. No data source is an island Key takeaways:
- 78. 5. Fundamentals can be interesting and challenging Key takeaways: