DevOps and End to End Visibility with Ed Gaile
- 1. Gaining Organizational Visibility Christopher Pepe Head of Solution Architecture Appfire Ed Gaile Principal Solution Architect Appfire
- 2. The system is completely down. Your code sucks! Ops 10:30am It was working fine during testing. Must be you. Devs 10:32am
- 3. ● Development Teams: Focused on delivering new features quickly. ● Operations Teams: Focused on stability, uptime, and risk reduction. ● The Result: Silos, conflicting priorities, and poor communication. Classic problems: Dev vs Ops Fragile process ● Manual deployment processes ● Infrequent, big-bang releases ● High risk of failure, with long rollback times ● Lack of test automation No visibility, no accountability ● Devs lacked insight into production performance ● Ops had limited context around code changes ● Customer issues took too long to reach the right team
- 4. ● Breaks down silos between Dev and Ops; eliminate blame ● Shared ownership, collaboration, automation ● Promotes continuous integration, testing, and delivery DevOps: people and process
- 5. ● Beware: automation increases complexity and does not ensure higher performance ● DevOps != tools and tech ● Start simple, achieve something ● Change often so that you get good at it Incremental improvement You cannot automate what you don’t understand. –Pepe
- 6. DevOps improves execution at the team level but only addresses a fraction of the visibility and alignment challenges that organizations face. Leaders still struggle to answer: ● Are we investing in the right initiatives? ● How do we balance short-term delivery with long-term business goals? ● How do we create visibility across departments and functions, not just dev teams? What about everyone else? CEO Operations Revenue Sales Marketing Channel Product Product Owners Support Engineering R&D Development Ops / Hosting Dev Ops
- 7. By 2025, 70% of digital investments will fail to deliver the expected business outcomes due to the absence of an SPM approach. Gartner, 2024 Strategic Roadmap for the EPMO, published 11 January 2024
- 8. Delivery Strategy Delivery Coordination Strategy Delivery Coordination Strategy Problem: The larger an organization gets, the more difficult it is to connect their strategic vision to their actual deliverable outputs. Size of organization
- 9. Middle Managers fail because: ● Self-taught, No formal training provided ● Frustrated at inability to affect change How? ● Alignment through KPIs ● Strategy lost in translation ● Empire Builders distract, delay, and deflect ● Bureaucracy leads to apathy Why Coordination Fails Middle management is critical in translating Strategy to Execution
- 10. Warning: Don’t optimize for your KPIs
- 11. Do KPIs ensure success?
- 12. Do KPIs ensure success?
- 13. Business Challenges Align to work goals Align budget, talent, & strategy Plan & Track Work Developer Productivity & Efficiency Scattered Metrics No visibility into cycle times & handoffs Maximize ROI by delivering the right features — on time
- 14. ● Aligns investments with business strategy and outcomes ● Enables data-driven decisions across teams and leadership ● Optimizes resource allocation and funding for highest-value initiatives ● Reduces redundancy and improves efficiency of execution ● Creates a single source of truth for initiatives, budgets, and timelines ● Bridges silos across IT, Finance, Marketing, and Product Strategic Portfolio Management
- 15. Program management Strategic portfolio management Goals and objectives Strategy Strategy layer ● CPO, CTO, Finance, R&D Portfolio layer ● Product Directors, Group PMs, PMO, ePMO Execution layer ● Programs, Products, Portfolios, Roadmaps, Projects 7pace Project management Work management Time Tracking
- 16. Use cases Objectives and key results (OKR) Financial management Strategic prioritization Risk management Define measurable outcomes aligned to long-term goals Track budgets and actuals in real-time, integrated into project workflows Prioritize initiatives using custom or built-in frameworks like RICE, ICE, and WSJF Identify, assess, and mitigate risks across strategy and execution
- 17. 0. Non-strategic goals Example walkthrough The Board wants to see the ELT stabilize revenue, and grow gross billings, while improving margin % Business Engine
- 18. 1. Planning and Strategy Example walkthrough ● Initiative: Open New Market ● OKR: Convert 20% of Jira users to Confluence in 1H ● Budget, timelines, and milestone tracking are centralized and tracked.
- 19. 2. Time tracking and effort management Example walkthrough As engineers begin sprint work: ● Actual effort versus estimated effort. ● Issues like under-scoping or hidden complexity early. ● Monitor/manage effort distribution
- 20. 3. Developer productivity and workflow insights Example walkthrough Throughout execution: ● Flow ingests data from GitHub, GitLab, Bitbucket, and Jira to show velocity trends, cycle time, and collaboration metrics. ● It highlights bottlenecks in PR reviews, long lead times, or teams not collaborating effectively. ● Leadership can intervene proactively instead of post-mortem analysis.
- 21. 4. Feedback loop and accountability Example walkthrough ● BigPicture pulls in real-time data from 7pace (actuals vs planned) and Flow (productivity, throughput). ● Stakeholders see a holistic, data-driven view of how strategic goals are progressing—backed by delivery and developer activity. ● This informs reallocation, reprioritization, or executive escalations if needed.
- 23. Consider the DVCS Connector
- 24. Flow uses Git and Ticket data for visibility across your SDLC to maximize efficiency and improve predictability Jira Github Bitbucket GitLab Azure DevOps Code Commits Engineering work items / tickets Pull Requests
- 25. Imagine Invest Plan Create Release Measure Winning Aspiration Idea to Production Outcome Plan & Track Illustrative Questions Answered ● How accurate is our planning? ● Do we have overburdened engineers? ● Are we able to really focus on our sprint goals? Code & Collaborate Illustrative Questions Answered ● What friction do engineers experience? ● Do teams review and collaborate effectively? ● Do we have knowledge silos? Build & Release Illustrative Questions Answered ● How often do we deploy? ● What waiting states take the longest? Invest Illustrative Questions Answered ● Is my team working on the right things? ● How am I spending my engineering investment? Observe and reduce friction through the SDLC
- 26. *Success metrics derived from 2M+ Flow data points between 2018 and 2022 and actual customer results. Unlock your people ● Improve collaboration; Reduce time to merge by 14 hours ● Reduce engineer ramp time by 4 months Upgrade your process ● Reduce cycle time by 35% ● Optimize investment dollars and save millions annually on reporting costs
- 27. A successful change journey starts with a solid foundation and needs services to support the long term path Demonstrate impact Build on your foundations with small, iterative improvements of your “mechanical” metrics progressing towards goals. ● Improve delivery metrics ● Eliminate bottlenecks ● Improve quality through collaboration ● Improve tooling and the delivery pipeline ● Measure improvements and create tight feedback loops across your system of delivery Establish a foundation Baseline your organization with Flow data shared across your teams and begin to understand the cadence of your business and identify patterns. ● Standardize practices, processes, & tooling ● Set and communicate expectations ● Establish stable teams ● Identify areas of improvement ● Set targets relative to your baselines Strategically grow Harness the improvements being made within your system of delivery and improve the commercial results of your business. ● Demonstrate product/service improvements ● Improve strategic investments ● Regularly deliver on Objectives and Key Results ● Increase revenue and customer satisfaction
- 28. A successful change journey starts with a solid foundation and needs services to support the long term path Establish a foundation Baseline your organization with Flow data shared across your teams and begin to understand the cadence of your business and identif patterns. ● Standardize practices, processes, & tooling ● Set and communicate expectations ● Establish stable teams ● Identify areas of improvement ● Set targets relative to your baselines
- 29. A successful change journey starts with a solid foundation and needs services to support the long term path Demonstrate impact Build on your foundations with small, iterative improvements of your “mechanical” metrics progressing towards goals. ● Improve delivery metrics ● Eliminate bottlenecks ● Improve quality through collaboration ● Improve tooling and the delivery pipeline ● Measure improvements and create tight feedback loops across your system of delivery
- 30. A successful change journey starts with a solid foundation and needs services to support the long term path Strategically grow Harness the improvements being made within your system of delivery and improve the commercial results of your business. ● Demonstrate product/service improvements ● Improve strategic investments ● Regularly deliver on Objectives and Key Results ● Increase revenue and customer satisfaction
- 31. How do I find bottlenecks? Engineering signals are Leading Indicators in a workstream. This objectively gives a view of what is happening on the ground in near real time. ● Impact score - effort != progress ● Sprint Movement - sprint over sprint completion rate. Based on what's delivered in a sprint you can see if on/off track ● Dev Weeks - engineer effort distribution ● Time to Merge - part of cycle time, early indicator of a problem
- 32. Change set 1 1 file ++100 lines Change set 2 3 files, 4 insertion points ++16 lines -24 lines On the left we have a large commit of 100 new lines in a single file. On the right, we have a much smaller commit, where only 16 lines were added, 24 were removed and happened over three files at multiple insertion points. Both commits receive an Impact score, but the right one scores slightly higher. Smaller, frequent commits focused on iteration, cleanup, and fixes happen more often, leading to higher overall scores. We’ve observed that teams working this way deliver more for their customers. Impact score
- 33. Sprint movement
- 34. Sprint movement Ticket Delivery Sprint 6 Dec 3, 2021 - Jan 6, 2022 ● Completed a lot of work ● Most of it was planned ● We had some scope creep
- 35. Dev weeks
- 36. Why do you spend 40% of your time on Defects? Could that be improved? ● Indicates poor quality control in the SDLC ● Impacts planning in the SPM ● Solution: Adopt or advance TDD capability/automate testing as part of the SDLC ● Solution: Improve Pull Request Review quality, adopt common coding standards, create more collaboration between developers ● Solution: Create clearer story definitions and acceptance criteria Dev weeks
- 37. Time to merge: PR resolution chart ● On average Pull Requests are merged within the 24 hr SLA ● A few outliers but generally all are done within 75 hrs Max Allowable ● What is the reason for the outliers? It is a process or behavior/individual that can be addressed?
- 38. Questions?