[Craft Conf 2024] Occupy the Space: A grassroots guide to engineering (and product) strategy when none exists

The typical problem in product engineering is not bad strategy, so much as “no strategy”. This leads to confusion, lack of motivation, and incoherent action. The next time you look for a strategy and find an empty space, instead of waiting for it to be filled, I will show you how to fill it in yourself. If you’re wrong, it forces a correction. If you’re right, it helps create focus. I’ll share how I’ve approached this in the past, both what works and lessons for what didn’t work so well.

1. Occupy the Space: A grassroots guide to engineering (and product) strategy when none exists Jason Yip Principal at Reinvention Cycle LLC

2. Here’s the scenario…

3. Low morale “No strategy” Why? Leadership response? Just repeating the vision

4. Someone needs to step up!

5. What do we mean when we say “strategy”?

6. TL;DR Strategy is problem-solving for a high stakes challenge.

7. Good strategy has 3 parts.

8. Diagnosis Not just “this is a problem” but also “why is the problem happening?” Guiding Policy Guiding principles for how we believe we can address the problem. Coherent Actions Specific actions we take that are coherent with the guiding policy.

9. Diagnosis Guiding Policy Coherent Actions Our ML models are not performing well because our legacy system does not capture the data we need to train effectively. Migrate off the legacy system to new services. Focus on migration of traffic over migration of capabilities. Direct all new traffic to new services Explicit migration timeline with target outcomes (not just best efforts) New capabilities built on new services first – even if exception made to also add to legacy.

10. Therefore, “no strategy” means…

11. Diagnosis Guiding Policy Coherent Actions ? ? ?

12. Diagnosis Guiding Policy Coherent Actions ? ? Our ML models are not performing well because our legacy system does not capture the data we need to train effectively. Either incoherent action due to lack of guiding policy OR incoherent action that ignores guiding policy

13. Diagnosis Guiding Policy Coherent Actions Our ML models are not performing well because our legacy system does not capture the data we need to train effectively. Migrate off the legacy system to new services. Focus on migration of traffic over migration of capabilities. Direct all new traffic to new services No explicit migration targets. Explicit short-term business outcome targets. New capabilities built on legacy system first because most of the traffic is there.

14. So what? What’s the problem with “no strategy”?

15. Confusion Lack of motivation Incoherent and competing actions

16. Countermeasures for “no strategy”

17. Occupy the Space Don’t Sweat Ownership Overall concept Inner challenge Draft Share Decide Get Staff+ Engineers Involved Recruit dedicated liaisons How

18. Occupy the Space

19. Steal underpants Profit Empty space where there should be a strategy

20. There is an empty space where there should be a strategy Complain about leaders not stepping up You’re capable, why not step up yourself? Someone else fills the space with a bad strategy Only one of these choices is useful.

21. One of two outcomes Your strategy is accepted… because it’s competent and has buy-in. Your strategy is not accepted… but leaders are forced to correct it. This one is better This one is acceptable

22. Diagnosis Not just “this is a problem” but also “why is the problem happening?” If you want your strategy to be “good”... Coherent Actions Specific actions we take that are coherent with the guiding policy. Guiding Policy Guiding principles for how we believe we can address the problem.

23. I’ve yet to see this be a problem… but just in case… You get a “+1” for addressing the strategy gap. The leader who was supposed to provide the strategy gets a “-1”

24. Recap: Occupy the Space When there is a strategy gap, fill it yourself. This will lead to either: Acceptance of your strategy Leaders correct it. Either way, you no longer have a strategy gap.

25. Your strategy is accepted… because it’s competent and has buy-in. How do you increase the likelihood of this? Draft Share Decide Get Staff+ Engineers Involved Recruit dedicated liaisons

26. Draft Share Decide Draft, share, decide. Starting principles | by Jason Yip | Medium

27. You can’t get to depth quickly with too many people involved. Why you can’t have a useful conversation with a lot of people | by Jason Yip | Medium Strategies created by committee tend to be shallow. Which is why

28. “Without involvement, there is no commitment. Mark it down, asterisk it, circle it, underline it. No involvement, no commitment. ” Stephen R. Covey

29. NOTE: Sharing can sometimes be more about buy-in than it is about getting additional insight… Don’t make it more complicated than it needs to be.

30. Recap: Draft Share Decide Draft strategy with a small group Share it with a broader group for feedback Decide on the “final” version

31. Get Staff+ Engineers involved

32. Target characteristics for “Staff+ Engineers” Been around for a while. Knows a lot about how the systems work. Implicit respect and authority independent of reporting lines This could be you. If not, get that person (or people) involved

33. Why get Staff+ Engineers involved? They probably have a perspective that should be incorporated in the strategy. Their involvement can sometimes automatically create buy-in.

34. Recap: Get Staff+ Engineers Involved The most senior technical ICs tend to have influence. Exploit this and get them involved.

35. Recruit Dedicated Liaisons

36. “This is our proposed strategy” “This is our proposed strategy… AND all the dependent groups are bought-in” This one is stronger

37. Get dependent groups bought-in. Establish good relationships with dependent groups. Recruit dedicated liaisons to establish good relationships.

38. “Bureaucratic pressure alone is an inherently flawed method for ensuring genuine, long-term cooperation — as no one appreciates being told to play nice with others. When the pressure is removed for whatever reason, cooperation is likely to diminish.” “liaisons offered a completely different channel for connectivity, a human-to-human personal network that was accountable for the creation of cross-boundary, dotted-line relationships among independent entities.”

39. “The liaison seems well-mannered and intelligent. We must assume the rest of the team is the same.”

40. Recap: Recruit Dedicated Liaisons A strategy with buy-in from other groups is stronger. If you want buy-in from other groups it’s useful to have someone dedicated to building relationships with them.

41. Don’t Sweat Ownership

42. You’ve addressed the strategy gap. Your area is now under control and doing well. This makes your area attractive for someone to takeover and/or steal credit. Empire builder

43. No strategy… with associated confusion, lack of motivation, incoherent action, etc. Clear strategy… but someone steals credit for it. This one is less bad.

44. Recap: Don’t Sweat Ownership When you come up with a good strategy, it’s possible someone else with more power will show up and take credit. This is still better than having a strategy gap.

45. Summary

46. Low morale “No strategy” Why? Leadership response? Just repeating the vision Occupy the space!

47. Occupy the Space Don’t Sweat Ownership Overall concept Inner challenge Draft Share Decide Get Staff+ Engineers Involved Recruit dedicated liaisons How

48. Questions?