Download as PDF, PPTX
Uploaded byAtlassian
A Product Manager's Place in a DevOps World
The document discusses the evolving role of product managers in a DevOps environment, emphasizing their responsibility for managing features throughout the entire lifecycle, including incident management and reliability as a critical aspect of performance. It highlights the need for collaboration between product teams and site reliability engineering (SRE) to ensure operational excellence and customer satisfaction. The importance of clear communication during incidents and post-release processes, as well as utilizing metrics like Apdex to measure performance and reliability, is also outlined.
More Related Content
Similar to A Product Manager's Place in a DevOps World
A Product Manager's Place in a DevOps World
- 1. A Product Manager’sPlace In a DevOps World DAVE MEYER | SENIOR PRODUCT MANAGER | ATLASSIAN | @D_MEYER
- 2. What does aproduct manager do? BIG HAIRY QUESTION
- 4. Tech. Sgt. BradFallin, U.S. Air Force; Source (https://commons.wikimedia.org/wiki/File:F-15C.jpg)
- 5. National Aeronautics andSpace Administration (NASA); Source (https://commons.wikimedia.org/wiki/File:STSCPanel.jpg)
- 6. Gus Pasquarella; Source(https://commons.wikimedia.org/wiki/File:Hindenburg_burning,_1937.jpg)
- 7. What does aproduct manager care about? BIG HAIRY QUESTION #2
- 8. Looking good BIG HAIRYQUESTION #2 Customers
- 9.
- 10. Lasse Fuss; Source(https://commons.wikimedia.org/wiki/File:Lufthansa_A380_D-AIMA-1.jpg)
- 11. DevOps Collaboration between dev andIT SRE Site Reliability Engineering DevOps vs SRE
- 12. BEN TRAYNOR, GOOGLE SREis what happens when a software engineer is tasked with operations “Keys to SRE”
- 13. All product teamsare ops teams now
- 14. Keep the thing running SHIP FEATURES PMDEV TEAM GET FEEDBACK SHIP MORE FEATURES (ship features)
- 15. (Re)align your priorities Source(https://commons.wikimedia.org/wiki/File:KC-97G_Illinois_ANG_refueling_F-100C_DC_ANG_1964.jpg)
- 16. Your tactics Reliability asa feature Support the culture Managing incidents
- 17. You own featuresall the way to release, then past release, then forever. Nessy-Pic; Source (https://commons.wikimedia.org/wiki/File:Caernarfon_Castle_Walls.jpg)
- 18. If you can’tmeasure it, you can’t improve it it’s probably broken and your customers are tweeting about it.
- 19. Error rates Is therea new problem in the system? What dev teams measure in production Performance Is the system running at the same speed? Behavior Are the things we expect users to do happening?
- 20. Behavior Are the thingswe expect users to do happening? Error rates Is there a new problem in the system? Performance Is the system running at the same speed? Alerting Do you know about it? What dev teams measure in production
- 24. Something bad 403 errorsin project creation have spiked!
- 27.
- 28. React in minutes MTTRis the key to happy customers Wait for support Support response time is the key to happy customers
- 29. RELIABILITY AS AFEATURE: PERFORMANCE How Apdex changed JIRA Cloud
- 30. What is Apdex? GOODREQUESTS 0.5(OK REQUESTS) TOTAL REQUESTS 0 - 1
- 31. RUF breakdown 🐶 0 0.125 0.25 0.375 0.5 January15 Feb Mar Apr May Jun July Aug Sep Oct Nov Dec Jan 16 33.8% 31.3% 31.1% 32.8% 30.3% 31.5% 34.5% 33.5% 32.8% 28.8% 19.2% 16.7% 16.8% 41.7% 41% 41.8% 43.7% 40% 43.9% 43.1% 42.1% 43.7% 39.7% 43.1% 43.9% 42.8% 20.3% 21.8% 21% 19.5% 20.4% 19.5% 18% 20% 15.8% 15.4% 9.5% 8.1% 7.9% Reliability Usability Functionality
- 32. Performance Mentions vsApdex 0 0.175 0.35 0.525 0.7 4.5% 9% 13.5% 18% January 15 Feb Mar Apr May Jun July Aug Sep Oct Nov Dec Jan 16 % Mention Apdex
- 33. JIRA Cloud performanceis now reasonable instead of feeling like grandmother’s old laptop. JIRA CLOUD CUSTOMER
- 34. NPS ++ RELIABILITY ASA FEATURE: PERFORMANCE
- 35. RELIABILITY AS AFEATURE Balancing your roadmap
- 36. Your tactics Reliability asa feature Support the culture Managing incidents
- 37.
- 38. Have a checklist Processesbuild confidence Pair with an expert Discuss impact of features on performance and error rates with SRE Document your decisions Make sure nobody is in the dark Are you ready to release?
- 39. Operational Readiness: Credo We automaticallyclone this template for every new service at Atlassian.
- 41. We made PMsbecome service owners, otherwise they just pay lip service to operational excellence. PAUL SLADE, ATLASSIAN IDENTITY HEAD OF ENGINEERING
- 42.
- 43. Feature flags Separating “deployingthe code” from “reeleasing the change” Allow quick iteration Encourage PMs to plan iterative rollouts
- 47. Have a checklist Processesbuild confidence Pair with an expert Discuss impact of features on performance and error rates with SRE Document your decisions Make sure nobody is in the dark Are you ready to release?
- 48. Have a checklist Processesbuild confidence Pair with an expert Discuss impact of features on performance and error rates with SRE Document your decisions Make sure nobody is in the dark Are you ready to release?
- 50.
- 51. Your tactics Reliability asa feature Support the culture Managing incidents
- 52. An event whichcauses disruption to or a reduction in the quality of services and customer productivity. WHAT IS AN INCIDENT?
- 53. If it affectscustomers, it’s your problem. WHY DO YOU CARE?
- 55. Know your role. INCIDENTMANAGEMENT FOR PRODUCT MANAGERS
- 58. Your job is notto fix it. You own the customer relationship.
- 59. Assess Help dev andSRE understand impact Communicate Own customer communications A PM during an incident
- 65. Responsible Even when it’snot your fault, it’s your fault. Clear No BS Active What are you doing about it? Empathetic Respect the customer Great customer communication
- 66.
- 68. What a successful PIRlooks like
- 69. Summary Tell me whathappened. Root cause Tell me exactly why it happened. Actions Tell me how you’re going to make sure it doesn’t happen again. What a successful PIR looks like
- 70. “Business value” is somuch more than features THE PM IN THE DEVOPS WORLD
- 71. DAVE MEYER |SENIOR PRODUCT MANAGER | ATLASSIAN | @D_MEYER 🎊Thanks🎊