February 10, 2026

When you don’t notice the data centre, the acoustics are working.
In densely populated cities, space is scarce and expectations are high. Residential buildings often sit shoulder to shoulder with hotels, cafes, offices, and shops. Thousands of people walk past critical infrastructure every day, and don’t even give it a second thought. In many ways, that is the benchmark of successful urban infrastructure.
We are involved in a long-running acoustic engagement at an urban data centre, which is located in one of the UK’s largest cities/most desirable places to live. It is surrounded by residential properties, heavy pedestrian footfall, hospitality venues, and commercial spaces. Most people passing the site would never guess that behind the façade of the building sits a 24/7 operational data centre. That is not an accident.
This site in particular has been in operation for many years. In planning terms, it is a product of a different era. Today, housing has been built extremely close to the operational plant, and in some cases, within 20 meters of a bank of chillers running continuously. If this site were being proposed today, it would almost certainly never receive consent. A modern baseline noise survey, if it was possible to measure the original pre-construction sound climate, would suggest plant noise levels exceeding that background noise climate by 25-30 dB at the nearest residential receptor. By current standards, that is simply unworkable. Yet, for this operator, it works.
As cities evolve, infrastructure that once sat comfortably in industrial or less densely-populated areas often finds itself absorbed into the urban fabric. The challenge then becomes how to manage legacy noise issues, in a responsible and proportionate way. This is where we come in.
Our role was not to ‘fix’ the site overnight. Instead, we were appointed as long-term acoustic consultants to help the operator demonstrate compliance with a Letter of Intent that required noise levels to be reduced over time, and (crucially) to evidence that reduction.
To achieve this, we decided to install long-term unattended noise monitors at carefully selected locations around the site; some were installed local to the plant equipment and some were installed at locations representative of nearby housing. These noise monitors were designed to run continuously, building a robust dataset that could track operational noise levels over months and years, rather than relying on snapshot surveys. Our objective was not to only report levels, but to understand how they changed and reduce over time when legacy equipment was replaced, upgraded, or decommissioned.
One of the underlying issues affecting this site was classic background noise creep. As the data centre expanded incrementally over the years, plant was added, load increased, and what had once been acceptable was slowly becoming problematic as the surrounding soundscape changed. On top of that, the introduction of new residential receptors brought the problem closer to our operator.
What is interesting, and perhaps counter-intuitive, is that the issues have largely solved themselves; given time and the right decisions.
As legacy equipment reached the end of its operational life, it was replaced with newer, quieter, and more efficient plant. This led to a notable reduction in the site’s overall noise output. Our role during this process was deliberately pragmatic: advising on plant selection, highlighting acoustic risks early, and gently steering specifications toward options that delivered meaningful noise reductions without operational compromise.
Our long-term monitoring told the story clearly. Over time, reported levels did reduce. The site moved (and continues to move) closer to the intent of the Letter of Intent, through informed data-led decision making.
This example, in many ways, shows how acoustics should work in complex urban environments. Not as a blunt constraint, but as a framework for continuous improvement. The real value here lies with our understanding of a site’s history, its constraints, and its future trajectory.
The fact that this data centre now blends into its surroundings, unnoticed by most people who walk past it every day, is the intended outcome. In dense cities, that invisibility is earned over time through careful monitoring, informed choices, and willingness from the operator to adapt to the realities of urban development.
To Conclude:
Urban data centres are only going to become more common, more constrained, and more closely scrutinised, and retrofitting them to meet modern planning expectations is rarely straightforward. This project demonstrates that long-term acoustic management, grounded in monitoring, trust, and collaboration, can deliver meaningful outcomes without disrupting operations or the surrounding community, turning acoustics from a reactive compliance exercise into a strategic tool for operators and a mechanism for enabling essential infrastructure to coexist with high-quality urban living. When a data centre becomes acoustically invisible in one of the busiest parts of a UK city, that is not luck, but the result of patient, evidence-led decision-making.