Hello Reader—
“We’re shutting down the platform team.”
The Slack channel went silent. 15 engineers. £4.2M annual budget. Gone.
But here’s the twist: It was their idea.
Six months later, deployment frequency increased 400%. Developer satisfaction hit 9.2/10. Platform costs dropped £3M annually.
The platform team didn’t get fired. They got promoted to “Product Engineering” and became the most valuable team in the company.
Here’s how they did it - and why your platform team should consider the same radical move.
THE PLATFORM BOTTLENECK EPIDEMIC
After analysing 73 platform teams, I discovered a shocking pattern:
The platform team becomes the biggest bottleneck to platform adoption.
The death spiral looks like this:
Week 1: “We need a platform team to help developers”
Week 12: “All requests must go through the platform team”
Week 24: “Platform team is overwhelmed, 6-week backlog”
Week 48: “Developers are building shadow infrastructure”
Week 96: “Platform adoption is 12%, we’re spending £4M on nothing”
Sound familiar?
That fintech client had 47 engineers supporting a platform used by… 23 developers. The math was brutal: £182K per active user annually.
THE SELF-SERVICE REVOLUTION
The breakthrough came when their platform lead asked one question:
“What if we made ourselves unnecessary?”
Instead of being the gatekeepers, they became the product builders:
Before (Gatekeeper Model):
- Developers submit tickets for deployments
- Platform team manually provisions resources
- 6-week average turnaround time
- 47 people managing infrastructure
- 12% platform adoption
After (Self-Service Model):
- Developers deploy directly via automated pipelines
- Platform team builds tools and guardrails
- 6-minute average deployment time
- 8 people building platform products
- 94% platform adoption
The secret: They stopped doing work FOR developers and started building tools that let developers do work THEMSELVES.
THE PLATFORM AUTONOMY SCORECARD™
Rate your platform team (1-5 scale):
Developer Independence
1: Developers need tickets for everything
5: Developers self-serve 90% of needs
Request Turnaround Time
1: Weeks for simple requests
5: Minutes via automation
Team Focus
1: Reactive firefighting and manual work
5: Proactive product development
Adoption Rate
1: <30% voluntary adoption
5: >80% voluntary adoption
Knowledge Distribution
1: Platform team holds all knowledge
5: Developers understand and own their deployments
Score 15+: You’re building a product
Score 10-14: You’re a service team (risky territory)
Score <10: You’re a bottleneck (time for radical change)
REAL STORY: THE £8M SELF-SERVICE TRANSFORMATION
Media company. 200 developers. Platform team of 52 people.
The breaking point: Developer survey showed 73% were “actively seeking alternatives” to the internal platform.
The transformation:
- Fired 44 platform team members (with generous packages)
- Kept 8 to build self-service tools
- Gave developers direct access with guardrails
- Built automated compliance and security
Results after 6 months:
- Platform costs: £8M → £2.1M annually
- Deployment time: 4 hours → 8 minutes
- Developer satisfaction: 2.1/10 → 8.7/10
- Platform adoption: 23% → 89%
- Incidents: 47/month → 12/month
The kicker: Those 8 remaining platform engineers became the highest-paid team in the company. They weren’t managing infrastructure - they were building products that generated millions in developer productivity.
THE 4-PHASE SELF-SERVICE TRANSFORMATION
Phase 1: The Audit (Week 1-2)
- Track every platform team request for 2 weeks
- Categorise: Automatable vs. Requires human judgment
- Survey developers: What do you actually need?
Phase 2: The Quick Wins (Week 3-6)
- Automate the top 5 most common requests
- Build self-service dashboards for status/metrics
- Create “break glass” procedures for emergencies
Phase 3: The Product Pivot (Week 7-12)
- Redefine platform team as product team
- Focus on building tools, not doing tasks
- Measure success by developer adoption, not tickets closed
Phase 4: The Scale Test (Week 13-24)
- Gradually reduce manual interventions
- Monitor for gaps in self-service capabilities
- Continuously improve based on developer feedback
THIS WEEK’S PLATFORM PSYCHOLOGY INSIGHT
From last week’s Metric Makeover responses:
87% of you are tracking vanity metrics that hide real problems.
But here’s the connection to self-service: Teams with high self-service scores had the BEST business metrics.
Why? When developers own their deployments, they care about the metrics that actually matter. When platform teams do everything, developers disconnect from the consequences.
The uncomfortable truth: Your platform team might be shielding developers from the feedback loops that create great software.
THE SELF-SERVICE SUCCESS FORMULA
Instead of: “Submit a ticket for deployment”
Build: One-click deployment with automated rollback
Instead of: “Platform team will investigate”
Build: Self-service debugging tools and runbooks
Instead of: “We’ll provision your resources”
Build: Resource templates with cost visibility
Instead of: “Platform team owns production”
Build: Developer ownership with platform guardrails
Instead of: “Come to us for help”
Build: Documentation so good help isn’t needed
READER SUCCESS: THE TEAM THAT MADE THEMSELVES OBSOLETE
“Steve, we followed your self-service framework. In 3 months, we went from 67 tickets/week to 3. Our platform team shrunk from 12 to 4 people, but those 4 are now building tools that the entire company uses. My CEO asked how we ‘magically’ became so efficient. The secret: We stopped being the bottleneck and became the highway.”
- Platform Director, SaaS (Edinburgh)
YOUR SELF-SERVICE CHALLENGE
This week, pick ONE manual process and automate it:
- Deployment approvals → Automated compliance checks
- Resource provisioning → Self-service templates
- Environment setup → One-click creation
- Monitoring setup → Automatic instrumentation
- Access requests → Role-based automation
Track the time saved. Multiply by 52 weeks. That’s your annual ROI from ONE automation.
WHAT’S COMING NEXT WEEK
Issue #010: “The Platform Migration That Saved Christmas” - When everything goes wrong at the worst possible time - Crisis management for platform leaders - Your disaster recovery playbook
Plus: Results from this week’s Self-Service Challenges!
📢 FROM THE NEWSLETTER TO THE STAGE
I'll be at BitSummit Hamburg (Sept 4th) sharing the full story behind our biggest platform transformation - the one that started with pink post-its and ended with GitOps clarity.
"From Console Chaos to GitOps Clarity: A FinTech Transformation Tale"
Newsletter readers get 15% off with code: STEVE_BITSUMMIT
Register: https://bitsummitapp.eventify.io/t2/tickets
See you in Hamburg? Reply and let me know!
Build products, not processes.
Steve
P.S. That platform team that “fired themselves”? They’re now the most requested team for new projects. Turns out, when you solve problems instead of managing them, everyone wants to work with you.
P.P.S. In the masterclass, I’ll show you the exact automation that saved one team 847 hours per month. Spoiler: It wasn’t complex.