As enterprises add cloud services, security tools and monitoring platforms, managing the technology stack can become harder than managing the individual systems within it. 

The problem is not necessarily how many tools an organisation uses. Complexity emerges when teams must move between platforms to investigate problems, reconcile conflicting information, or maintain integrations that require constant attention. 

At Tech Show London 2026, speakers across cybersecurity and DevOps examined how fragmented environments affect operational performance. Their discussions highlighted the additional work created when tools and processes do not operate together effectively. 

1. How do technology silos create operational problems? 

Organisations often introduce applications to address individual business requirements without fully considering how those systems will operate together. This can create problems when teams need to investigate activity across multiple platforms. 

During his Tech Show London session Mission Possible: Basics That Break Zero-Trust Failures, Hariharan N, Senior Technical Evangelist at ManageEngine, highlighted the difficulties created by disconnected infrastructure. 

“The problems with the current infrastructure that we have is the siloed mode in which we operate,” said Hariharan. 

This becomes particularly difficult when security teams need to investigate activity across multiple systems. Logs may exist, but bringing them together to understand what happened can require additional work. 

The UK's National Cyber Security Centre (NCSC) advises organisations to understand their architecture, including the relationships between users, devices, services, and data. 

For enterprise teams, the first question to answer is whether they can build a reliable picture of their environment without manually assembling information from separate platforms. 

2. Why does interoperability matter as the stack grows? 

Once an organisation operates multiple systems, the ability to exchange information becomes essential. 

Interoperability means different technologies can share the information needed to complete a task without someone manually connecting the results. 

For example, a security platform might detect suspicious activity, but the investigation could also require identity records and cloud infrastructure logs. If those systems cannot exchange useful information, analysts must reconstruct the sequence themselves. 

Hariharan's discussion of unified logging and correlation highlighted this challenge. Collecting information is only part of the process; organisations must also be able to connect related events. 

The OpenTelemetry project, hosted by the Cloud Native Computing Foundation, provides open-source tools and standards for collecting observability data across different systems, reducing reliance on proprietary data collection methods. 

This can also support flexibility when changing suppliers. However, interoperability needs to be tested against real operational requirements rather than judged solely by whether products advertise integrations. 

3. Does every new capability make the enterprise more complex? 

Adding new services or integrating external platforms can create dependencies that organisations must manage long after the initial implementation. 

Lee Howard, Group Head of Cyber Security and Data Protection at Evri, addressed this challenge during his session Delivery at Scale, Risk at Speed, drawing on the operational demands of a large logistics business. 

“Every new capability creates the dependency,” said Howard. 

Each integration can create relationships that other systems or business processes depend upon. As those relationships grow, organisations need to understand what happens if one component becomes unavailable. 

Howard argued for simpler architecture and design patterns that make integration easier to manage. 

“Simplicity scales that pace,” he said. 

The challenge is particularly relevant as AI accelerates software development. More frequent changes can create additional demands on teams responsible for testing and maintaining infrastructure. 

Research from DORA on platform engineering supports the use of internal platforms to reduce the complexity developers must manage directly. 

Before adding another capability, enterprises should understand the dependencies it creates and whether existing teams can support them. 

4. Are teams dealing with too much information rather than too little? 

New research suggests that managing multiple monitoring platforms remains a challenge. According to New Relic's 2025 UK and Ireland Observability Forecast, organisations across EMEA use an average of 4.83 observability tools, while 51.6% of EMEA respondents expressed a preference for a single consolidated platform. 

The findings, based on a survey of 750 IT and engineering professionals across Europe and the Middle East, also identify complex technology stacks and siloed monitoring data among the barriers to achieving full-stack observability. 

However, consolidation does not necessarily resolve the underlying problem. Organisations still need to determine whether information from different systems can be connected and used to investigate incidents effectively. 

5. When does tooling create more cognitive load than capability? 

The effect of fragmentation is often clearest in the experience of the people operating the technology. 

During a DevOps Live keynote, Martin Reynolds, Field CTO at Harness, and Billy Mattison, Senior Solution Engineer at Harness, demonstrated how developers could work across security remediation, deployment verification, and infrastructure cost analysis within a connected platform. 

Explaining the experience during the demonstration, Mattison said: “I don't have to context switch, and I don't have to memorise. Hey, what did I work on two weeks ago?”