For decades, inventory management for telcos has been treated as a necessary but unremarkable feature of daily operations. Operators’ need to understand where their network physically exists was crucial for planning and maintenance, but there was little thought that this data could influence investment decisions, operational efficiency, or customer experience
Today, however, this perception is beginning to shift, as operators grow increasingly aware of the strategic importance of network data, the limitations of legacy systems, and the practical realities of deploying AI at scale.
According to Luke Sullivan, Head of Global Pre-Sales, Telecommunications at VertiGIS, the transformation is about telcos finally making use of the data they have always had access to.
“Fundamentally, inventory is still boring,” he joked. “But what’s exciting is that understanding inventory on a more granular level means that you have a much better appreciation of the value of your network, how it’s used, and how you can deliver services to individual customers.”
From obligation to opportunity with AI
Historically, telco inventory systems were built purely to ensure that operators knew what infrastructure they had and where it was located, with little thought given to using this data after it was recorded. In fact, as Sullivan points out, in many cases this data was only accessed when there were issues with the network.
“It is often the case that the primary focus for operators during deployment is speed – how fast can we construct the network? And what ends up happening is they only realise their weakness in inventory when something goes wrong,” said Sullivan. “As a result, these operators can take years before they understand how valuable their inventory would have been if it had been collected and managed their data more effectively.”
The rapid advances in AI, however, has led to this process being re-evaluated, offering not only significant cost savings through operational efficiency but also competitive advantage through improved customer service.
“The change in the last years has really been understanding that the inventory data has immense value,” Sullivan explains. “We can use that data to improve the way we deploy services, to maximise the efficiency of the network, and to improve operations. We now have the tools to leverage that data in the most efficient ways possible, and companies are finding much more creative and powerful ways of taking advantage of it.”
One area seeing significant improvement is inventory validation. Previously, such validation would involve manually visiting and identifying the physical infrastructure, a process that was both time consuming and prone to error. AI can greatly accelerate these tasks.
“Insufficient checks or validations of what was installed in the field compared to what was planned can create a significant gap between inventory data and the real network,” said Sullivan. “AI can help field engineers document deployments by automatically analysing and categorising images and video. Then, it can take the results and compare them to planning documents, flag discrepancies, and adjust the network accordingly.”
“These are processes that have historically been semi-manual or needed additional validation but are now being done automatically. That saves a lot of time and hard work, so it’s enormously valuable,” he added.
Creating a single source of truth from disparate data
Of course, as with any automation process, the quality of data remains a key concern. Older networks in particular suffer from poor or missing inventory information, which can greatly delay returns from AI implementation.
“One of the fundamental issues is if the data in the inventory system is incomplete or incorrect, then any decisions an AI tool is going to make are also going to be incorrect,” he said. “Both humans and AI can only work with the information in front of them.”
While some operators struggle with incomplete data, others face a different problem: they already have high-quality data but cannot use it effectively.
“There are lots of legacy systems that have perfectly good datasets. That doesn’t actually mean that they are able to leverage it efficiently,” said Sullivan.
This disconnect reflects a broader challenge across the industry. Many inventory systems were not designed with advanced analytics, automation, or integration in mind, with even well-maintained datasets can remain siloed or inaccessible.
For Sullivan, the solution is to bring this data together into a unified Geographic Information System (GIS)-based environment that enables consistent modelling, planning, and operational insight, such as VertiGIS ConnectMaster.
“We call it our single source of truth,” he explained. “It is built on VertiGIS’ Neo framework, which focuses on cloud-first architecture and scalable deployment models.”
Crucially, it also integrates into customers’ existing systems through APIs, making it easy to customise to the operators’ individual needs.
“We’re evolving our applications to provide flexibility for deployments, flexibility for how the applications and the solutions can scale, but also to future-proof them as the customer requirements continue to change,” said Sullivan.
Unlocking value from ‘boring data’
Ultimately, for Sullivan and ConnectMaster, the future of inventory and GIS systems lies in making infrastructure data both accessible and actionable. More than a technological shift, this will involve a major mindset shift for operators.
“Operators need to understand not just how to collect the data, but how to maximise its value,” said Sullivan. “That involves a lot of analysis and a lot of modelling of future demands on the network. These are key value points that are much more at the forefront of people’s minds today.”
The rigid systems of the past are rapidly becoming malleable, able to be tailored to specific outcomes and solving real-world problems. Operators that succeed in structuring, governing, and leveraging this “boring” data will gain a measurable advantage in how they plan, operate, and evolve their networks.
“I actually wish the customers would come to us with more problems,” concluded Sullivan. “In most cases, the data is there already. They just need experts who understand their unique challenges and can provide a flexible solution to help deliver positive outcomes.”
Meet the VertiGIS team at FTTH Conference 2026
VertiGIS is attending FTTH Conference 2026, taking place 14–16 April 2026 at Excel London, where the team is discussing the evolving role of network inventory as a foundation for efficient fibre network planning, operations, and AI-enabled workflows.If you would like to explore how fibre operators are modernising network inventory management and creating a structured system of record across planning, documentation, and operations, we welcome the opportunity to connect at Booth S22. How telecom’s most boring data is becoming its most valuable asset - Total Telecom
