From Network Blame to Platform Teams
Rethinking Infrastructure Support
In the world of IT operations, there’s a metric that network teams know all too well: Mean Time to Innocence (MTTI). It’s the average time it takes for a network team to prove they’re not responsible for an outage or performance issue. While this might sound amusing, it points to a deeper problem in how we structure and organise our infrastructure teams.
The Network-First Blame Game
When applications slow down or services fail, the network team is often the first to receive the ticket. “It must be the network” has become such a common refrain that it’s both a running joke and a source of frustration for network engineers worldwide. This default assumption leads to wasted time and resources as network teams repeatedly prove their innocence before issues can be properly diagnosed.
The irony? Network engineers are often excellent troubleshooters, frequently identifying and solving problems that have nothing to do with the network. While this demonstrates their versatility, it reinforces the cycle of network-first blame and creates inefficient support patterns.
The Silo Trap
As organisations grow, infrastructure teams typically fragment into specialised silos: LAN, WAN, security, cloud networking, application delivery, and more. While this specialisation can drive deeper expertise, it often results in a costly game of ticket ping-pong. Issues bounce between teams until every silo has proved its innocence, while users wait for resolution.
This problem has only intensified with cloud adoption. While cloud providers promise to simplify infrastructure, they’ve actually increased network complexity and the need for sophisticated connectivity expertise.
The Platform Team Solution
The answer lies in reimagining our approach to infrastructure support. Instead of technology-focused silos, we need platform teams with broad skills and comprehensive visibility. Think “full stack infrastructure” — similar to how development has embraced full stack engineers.
These platform teams should embody a new way of working. They need to bring together networking, security, cloud, and systems expertise under one umbrella. Critical to their success is maintaining end-to-end visibility across all connectivity paths, while focusing on service delivery rather than individual components. Most importantly, they must operate with a holistic view of the infrastructure landscape.
Making the Transition
The transformation from siloed specialists to platform teams requires several key changes. First, we need tools that support the vision. This means implementing end-to-end synthetic monitoring and comprehensive observability platforms. But critically, we need to rethink how we present this information.
Traditional dashboards showing the status of individual services or technology components miss the point entirely. Instead, dashboards should be structured around business outcomes and customer journeys. A retail dashboard shouldn’t tell you if the payment gateway is online — it should answer the fundamental question “Can your customers buy things?” This single view should show the health of every component and data flow required for a customer to scan their goods and complete their purchase.
Consider a banking platform. Rather than showing database uptime or network latency, the dashboard should answer “Can customers check their balance?” This immediately contextualises the technology stack in terms of customer experience. For an e-commerce operation, the key questions might be “Can customers fill their basket? Can they pay for it? Can we pick the order? Can we ship it?” Each of these represents a complex chain of interconnected services and data flows, but the dashboard reduces this complexity to what matters: business outcomes.
This business-centric approach to monitoring transforms how teams understand and respond to issues. When a problem occurs, everyone immediately sees its impact on customer experience, regardless of which technical component is at fault. This shared context helps break down silos and focuses everyone on what truly matters.
The evolution of skills is equally crucial. We should be developing “T-shaped” engineers with broad knowledge and deep expertise in specific areas. This means encouraging cross-training across traditional boundaries and focusing on service-oriented problem solving rather than technology-specific fixes.
The Path Forward
The future of infrastructure support isn’t in proving innocence — it’s in collaborative problem-solving focused on service delivery. By breaking down silos and building platform teams with comprehensive skills and visibility, we can move past the blame game and focus on what really matters: delivering reliable, high-performance services to our users.
The technology industry has already embraced “full stack” as a development paradigm. It’s time we applied the same thinking to infrastructure teams. The result will be faster resolution times, better service quality, and happier teams who can focus on innovation rather than defending their innocence.