Data Driven, Agile like it is supposed to be

It is long since Agile has become a commonality. Everyone is working in Scrum, Kanban or even Scrumban!
However, in our competitive and super-fast moving world this is not enough. We cannot assume that we know what our users really need, only the users know. Therefore, we need more tools and methods that will help us to learn faster and build the right product.
In this talk we’ll see how Lean development principles, analytics, A/B tests, hypotheses and other methods can support this process.

Transcript

1. Data Driven Agile, like it supposed to be
2. We all love agile!
3. Scrum
4. Kanban
6. Quick question...
8. We our process
10. Lean
11. It works great!
13. How many times... We’ve developed a feature that we didn’t understand why it was needed?
14. How many times... R&D and business were not aligned on the requirements?
15. How many times... Did we hurt our quality to get something done on time?
16. How many times... Have we developed features that no one have ever used?
18. Well...That’s also not so true
19. https://medium.com/dennis-nerush @DennisNerush nerush.dennis@gmail.com Team Leader @SEARS ISRAEL Dennis Nerush
21. # Data and analytics # Lean development principles
22. What is Software Development? LEAN
23. Lean
24. Lean
25. MVP
26. Making sense of MVP
28. How can we move fast?
29. Option #1: Reduce Quality
30. It is hard to Move Fast with Tech debts, low infrastructure, bad design And no Tests
31. I don’t have time to fill gaz I’m late
32. How many developers are here that were slowed down by bad code that was generated by “fast moving” scrum teams?
33. We can lower the feature quality
34. We can lower the feature quality Or we can decrease the feature scope
35. We can lower the feature quality Or we can decrease the feature scope And even not implement the feature at all
36. You cannot have speed without quality Robert C. Martin - The Land that Scrum Forgot
37. Option #2: Reduce Scope
38. Build the right thing right
39. MVP != Low Quality MVP == Smallest scope
40. https://themission.co/towards-a-minimum-buyable-product-why-your-buy-button-should-appear-sooner-f192eb5e91be Google’s PM interview question
44. Build the right thing right
45. What is the wrong product?
50. They all send emails...
52. How do we decide what should be build next?
53. The “Kano Model” https://foldingburritos.com/kano-model/ http://www.kanomodel.com/discovering-the-kano-model/ http://uxi.org.il/pages/19821
55. Must Have
57. Performance
59. Attractive
62. Balance
63. Recap
64. Build the right thing right MVP != Low quality MVP == Small scope
65. Lean
66. What do we measure?
69. Assumption Hypothesis KPI’s
70. There are ~4000 vegetarians in SF
71. Stop!
72. Is it enough?
74. KPI
75. KPI Range 30% - 40%
78. Lean
79. Know your numbers
80. #1 Visualize
83. 1. Business KPI’s 2. R&D KPI’s 3. Performance metrics 4. More...
84. #2 Numbers meeting
86. 1. Go over open experiments 2. Check the metrics, the KPI’s 3. Assess the hypothesis 4. Decide
88. Create a culture of numbers
89. 1. Build the right thing right 2. Hypotheses -> Tests -> Decision 3. Culture of numbers Let’s Recap...
90. One more thing ;)
91. Let’s say we do all that… Is it really enough?
92. NO
96. Research & Product discovery
98. It is not only the PM’s job
99. *Brainstorm sessions *Research sessions *Context sessions You have smart people in your company...
100. One vision
101. *Full alignment *Ownership *Commitment
102. Final Recap
103. Product Discovery & Research
104. So is it a new process? Maybe...
105. I call it Agile and sometimes Lean and sometimes... simply work :)
106. https://medium.com/dennis-nerush @DennisNerush nerush.dennis@gmail.com Team Leader @SEARS ISRAEL Dennis Nerush