11 Ways Humans Kill Good Analysis (Kevin Ertell)

680 views

Published on

We’re drowning in data in the eCommerce world. We can and do measure everything. But how do we get the most out of those numbers? Those mountains of data can be full of gold if we mine them correctly, or they can just be big piles of useless dirt. All too often, we misuse the valuable data we have and end up flailing away. Many of the reasons we aren’t happy with the results of the analyses come down to fundamental disconnects in human relations between all parties involved. Groups of people with disparate backgrounds, training and experiences gather in a room to “review the numbers.” We each bring our own sets of assumptions, biases and expectations, and we generally fail to establish common sets of understanding before digging in.

Published in: Marketing
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
680
On SlideShare
0
From Embeds
0
Number of Embeds
22
Actions
Shares
0
Downloads
1
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide

11 Ways Humans Kill Good Analysis (Kevin Ertell)

  1. 1. #monetatesummit 11 Ways HumansKill GoodAnalysis KevinErtell, SurLaTable
  2. 2. FOCUSED
  3. 3. FOCUSED ACTIONABLE
  4. 4. FOCUSED ACTIONABLE MANAGEABLE
  5. 5. FOCUSED ACTIONABLE MANAGEABLE ENLIGHTENING
  6. 6. FOCUSED
  7. 7. ACTIONABLE
  8. 8. MANAGEABLE
  9. 9. ENLIGHTENING
  10. 10. FOCUSED ACTIONABLE MANAGEABLE ENLIGHTENING
  11. 11. 11 Ways Humans Kill Good Analysis
  12. 12. 1. We hire reporters not analysts
  13. 13. Logical Sequential Rational Objective
  14. 14. 2. We turn analysts into reporters
  15. 15. Why is conversion down on Google paid search?
  16. 16. Why is conversion down on Google paid search? What’s'the'op+mal' marke+ng'mix'to'use' to'launch'Brand'X?
  17. 17. Why is conversion down on Google paid search? What’s'the'op+mal' marke+ng'mix'to'use' to'launch'Brand'X? Why'are'return' rates'growing?
  18. 18. 3. We expect the data to be perfect and the 
 analysis to be flawless
  19. 19. A man with one watch knows what time it is; 
 
 a man with two watches is never quite sure.
  20. 20. 4. We fail to define objectives 
 and state our assumptions
  21. 21. 5. We want numbers for number’s sake
  22. 22. KPIs Suppor+ng'Metrics Forensic'Metrics
  23. 23. Supporting metrics
  24. 24. KPIs Supporting metrics
  25. 25. 6. We insist on simplicity
  26. 26. How likely are you to recommend this business?
  27. 27. • Large margins of error • Low precision • Low detection of movement • Interpretation problems • Not very predictive How likely are you to recommend this business?
  28. 28. All'we'did'was'quan+fy'this' common'sense'in'a'way'that' made'sense'to'business'leaders— the'target'audience'for'my'book.'
 These'prac+cal'leaders'have'liJle' interest'in'advanced'sta+s+cal' methods.'
  29. 29. All'we'did'was'quan+fy'this' common'sense'in'a'way'that' made'sense'to'business'leaders— the'target'audience'for'my'book.'
 These'prac+cal'leaders'have'liJle' interest'in'advanced'sta+s+cal' methods.' 
 These practical leaders have
 little interest in advanced
 statistical methods
  30. 30. correlations
  31. 31. 7. We just want the number
  32. 32. “Plans based on average assumptions are wrong on average.” -Sam Savage
  33. 33. 8. We aren’t multilingual in the 
 languages of business & statistics
  34. 34. Standard deviations
  35. 35. Standard deviations Variances
  36. 36. 9. We expect answers immediately
  37. 37. You’re approaching a Coast Guard security zone. … If you don’t stop your vessel, you will be fired upon. Stop your vessel immediately.
  38. 38. You’re approaching a Coast Guard security zone. … If you don’t stop your vessel, you will be fired upon. Stop your vessel immediately. bang,'bang,'bang,'bang Bang! Bang! Bang! Bang!
  39. 39. regression to the mean
  40. 40. 10. We ignore our guts
  41. 41. Prefrontal cortex
  42. 42. 11. We blow the presentation
  43. 43. Questions? Kevin'Ertell' @kevinertell Kevin.Ertell@SurLaTable.com

×