SHARE
TWEET

Untitled

a guest Aug 13th, 2017 46 Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
  1. evanhallhtp: given the premises that posting every other weekday and both on saturday and sunday is the best strategy to avoid saturation
  2. evanhallhtp: how would you determine the performance difference between mon/wed and tue/thurs?
  3. sfowles.htpcompany@ymail.com: that one doesn't need a split
  4. sfowles.htpcompany@ymail.com: this is where it gets more complicated
  5. sfowles.htpcompany@ymail.com: because you will never post the same thing twice
  6. sfowles.htpcompany@ymail.com: it's hard to claim that one post was equally as engaging as the last
  7. evanhallhtp: ok, fast testing is impossible, so dont do it?
  8. sfowles.htpcompany@ymail.com: no, do what you can but minimize outside influence
  9. sfowles.htpcompany@ymail.com: test smart, not fast
  10. sfowles.htpcompany@ymail.com: 5 lame tests will not give you as much data as 1 good test
  11. sfowles.htpcompany@ymail.com: so theoretically the 1 good test can take 5x longer
  12. sfowles.htpcompany@ymail.com: and be more useful
  13. evanhallhtp: scott my movie is coming out in 6 months, i need this early time to get the best posting schedule on facebook, give me results in two months or i'm giving my $6000 a month to someone else
  14. sfowles.htpcompany@ymail.com: you should have given yourself more time
  15. sfowles.htpcompany@ymail.com: you're fucked
  16. evanhallhtp: you need this $6000 to pay rent
  17. evanhallhtp: and cure your own cancer
  18. evanhallhtp: what do you do
  19. sfowles.htpcompany@ymail.com: testing is testing, i'm not going to guarantee an outcome. That negates the point of using the term 'testing'
  20. sfowles.htpcompany@ymail.com: I'll do what I can in 6 months lol
  21. sfowles.htpcompany@ymail.com: give me your money
  22. evanhallhtp: you end up telling the client "post on tuesdays at 8am, post on thursdays at 6pm, post twice on saturday, three times on sunday. weekend times dont matter to shares or traffic"
  23. evanhallhtp: how did you get there?
  24. sfowles.htpcompany@ymail.com: right, that will give you what worked best that week, assuming that all your posts were equally as engaging as the last
  25. sfowles.htpcompany@ymail.com: which is also impossible to say
  26. sfowles.htpcompany@ymail.com: it's a very non-scientific test
  27. sfowles.htpcompany@ymail.com: results will vary, and impact of results will vary
  28. evanhallhtp: scott i dont understand what you're telling me. you gave me specific instructions but it doesn't matter because everything is unpredictable and objective value doesn't exist? derek said you were the man for the job, i'm charging back
  29. sfowles.htpcompany@ymail.com: it does exist you're just not enabling us to perform that level of testing for you
  30. sfowles.htpcompany@ymail.com: which is why I told you you were fucked in the beginning
  31. sfowles.htpcompany@ymail.com: there needs to be at least 1 controlled aspect to make the test valid
  32. evanhallhtp: scott, are these instructions worth the $12000 i paid or not? i dont need perfection, will it be better than random, or what zarrella says?
  33. sfowles.htpcompany@ymail.com: you need any one of the following:
  34. sfowles.htpcompany@ymail.com: controlled dates and times
  35. sfowles.htpcompany@ymail.com: controlled user group
  36. sfowles.htpcompany@ymail.com: controlled posts
  37. evanhallhtp: scott i cant create a controlled replica of the internet, are these instructions going to work better than the zarrella baseline?
  38. sfowles.htpcompany@ymail.com: no
  39. evanhallhtp: ok im cancelling. warm regards, Lizard Brain CEO of Dumb Company
  40. sfowles.htpcompany@ymail.com: lol
  41. sfowles.htpcompany@ymail.com: lizard braiiiiiiinnnnnnnnn
  42. evanhallhtp: most clients are lizard brains
  43. evanhallhtp: give them something simple and useful
  44. sfowles.htpcompany@ymail.com: thank god im not gonna be doing this for small businesses lol
  45. sfowles.htpcompany@ymail.com: its a great idea though, just needs a true split test
  46. sfowles.htpcompany@ymail.com: otherwise it's all just the same as the original hypotheses from zarrella
  47. evanhallhtp: yeah i think you are better off working for people with high budgets and a long view of their business
RAW Paste Data
Pastebin PRO Autumn Special!
Get 40% OFF on Pastebin PRO accounts!
Top