Inside track: Smart buildings expert unpacks the energy challenge

It’s getting harder to manage buildings efficiently and economically. Most systems in use today struggle to provide anything resembling accurate, real-time reporting of occupancy levels. More recent trends – notably hybrid working and spiralling energy costs – have intensified the pressure to do better. Technologies like unobtrusive movement and ID detection, and Big Data, can make a huge impact on how well buildings are managed, argues CEO Paul Sheedy. Envirotec spoke to him about new technology, human behaviour, and how his platform can solve building management headaches and impact the bottom line.

This is the perfect storm for the sector. Rising energy costs, increasing challenges in getting FM staff, no clarity about what is happening within buildings, a new norm and a hybrid model which does not seem to have stability are all creating complete uncertainty for businesses. You need to know how to measure your building, only then can you manage it better, says Sheedy.

Building usage has become more unpredictable in recent years. The post-Covid world brought about hybrid working but a lot of people are going back into offices again this winter, as a result of ballooning energy costs making home working an expensive choice.

But building usage is in a period of flux. For facilities managers and finance directors, it all adds up to a significant headache with managing things like building energy usage and addressing net-zero emissions targets, not to mention ESG reporting and, to top it all off, EPC regulations that are going to deliver huge challenges for the sector and cannot be ignored.

Commercial buildings waste about 30% of their energy use at any given time1. Over 96% of buildings have no actionable data to assist with meeting net zero targets2. In 2030, the UK government will introduce new Energy Performance Certificate (EPC) standards where a rating of B or above will be mandatory. But a C rating will be needed by 2025, leaving only 25 months now for buildings to deliver. As of January this year, 80% of offices fall below these standards, and may not be able to be leased.

There are huge savings to be made from putting automated systems in place to ensure systems like heating, lighting and air conditioning are only used where they are needed. Reliable occupancy data is the key ingredient for change but is widely absent.

Detecting occupancy

Large-scale systems that sense when people enter or exit a building have tended to rely on smartcards and RF technologies like near-field communications (NFC) – the technology for access and bank cards for example. has developed a custom RF technology to incorporate into the same ID cards, allowing longer-range, less obtrusive detection, but keeping your access system as it is.

Recent clients have included the new UK headquarters of a global corporate consultancy, which uses the proprietary long-range RFID smart card. For, this involved some custom engineering to produce an antenna design that allows the ID card to work accurately over a long range but without interfering with the access control system and undermining the read rate of the cards.

The company has developed multiple different antenna and chip systems, “to work with almost every access control system on the market.” In a more recent collaboration with Manchester Metropolitan University, Sheedy says they are developing new innovations to harness energy transmitted from communications systems that will be deployed in buildings within the next decade.

Engagement is crucial

Cutting edge technology can be important, Sheedy suggests, but it’s essential to be guided by an understanding of how people will use it – or even whether they will use it. Some of his own experience has been with “this interaction between technology and human behaviour”, he says, in a previous role building a technology firm that was using Big Data and AI to analyse consumer transaction data in order to find ways to improve customer engagement.

Bluetooth Beacon, for example, failed to take into account how the demographics of that target audience (shoppers) actually processed information. In every case globally this specific technology failed – a factor he puts down to “purely the psychological behaviour not being taken into account.” Technology always has to be simply deployed, to deliver maximum impact with minimum interactions.

Detection methods

Accurate, unobtrusive monitoring of floor-level occupancy with 100% accuracy is cited in the literature for Unifi-id’s platform, which employs various movement-detection approaches. Vector recognition is another ingredient that looks likely to be increasingly important. “We developed a GDPR-compliant way of doing that,” says Sheedy. A vector-based detection system identifies employees, matching a face to a card number – ignoring anyone else, and not holding any photos.

Avoiding the “big brother” element appears to be important, and he clearly distinguishes’s platform from technologies such as BlueDot tracking, the AI product which played a role in tracking cases of COVID-19 across borders in the early days of the pandemic.

Making data actionable doesn’t require much granular detail at all, he says. It’s enough to know who’s on each floor, what time they arrived and left. Or to be able to gain a quick grasp of building occupancy: “On Monday it was 15%, Wednesday 65%, Thursday 70%, and so on.” It means HVAC systems can be managed for who’s actually in the building, rather than running them at 100% all the time. But, more importantly, you need to know how the occupancy graph looked on each of these days.

AI plays an important role in making data actionable, and much of the value of’s platform seems to come from overlaying environmental and other data on top of this real-time view of occupancy. For example, incorporating weather data makes for better HVAC management. If it’s a very bright day, there might be an issue with turning off the air conditioning on a side of the building that gets the sun. But if it’s an overcast day and the predictive analysis on occupancy is clear, it will be okay to turn it off at the appropriate time.

Factors like outdoor temperatures and tube strikes also correlate with occupancy, and the platform uses AI (including things like machine learning and neural networks) to predict occupancy rates. Heavy rain might prompt people to work from home, but Mondays and Fridays are the days where such an effect is most pronounced, he says. “By Tuesdays a guilt complex starts to hit and most people will pick up an umbrella and struggle into the office.”

The platform has been built using many years’ worth of datasets, and from establishing precise correlations between parameters inside the building (temperature, CO2, humidity, and so on) and outside (things like weather and tube strikes). The resulting actionable insights seem to open up a range of possibilities with a measurable impact on the bottom line.

Acting on the data

Cost reductions arise from energy savings, as a result of better management of things like HVAC and lighting. Sheedy emphasises an approach of striving to understand each particular situation, and the unique problems facing a particular building, large or small.

This might involve working with a finance director to identify the big costs, the red lines and so on. Recent examples have included a small building with “archaic lighting” – indeed “this was the biggest issue” – where’s input has reduced lighting costs by 77%. This involved making a change to the way the client was lighting his premises – arranging a 7-year lease on an LED lighting system.

“You cannot manage what you cannot measure,” as Sheedy explains, and the approach is to give people the means to collect the data, and the AI platform to interpret it, so that they know what they should be doing. Small tweaks can have a big impact.

Plugging energy leaks

One example of energy wastage is the scenario where someone – a security guard, for example – walks around for maybe one minute per floor, and this triggers the lights to go on, which stay on for 30 minutes or so. If you have the data to show the lights came on at 2am, and it was because of a security guard, then you can set up the system to turn them back off again without delay as they leave the floor. A simple adjustment like that – when aggregated over weeks and months – can amount to a lot of energy saved.

Other practical objectives which the platform is set up to address include establishing optimal evacuation drills. In the event of a bomb scare or other emergency, even the most seemingly up-to-date buildings in modern business districts will seldom have a system in place for knowing exactly who is inside at any given time, surely an unnecessary omission, he suggests. And a real-time grasp of occupancy ID also makes intrusion prevention a much more infallible affair.’s systems also offer a rapid RoI. When the full solution is installed, clients can see energy cost reductions of 30%, delivering an RoI in just a few months after implementation, he says.

While the firm’s client list includes high-profile names in sectors like banking and finance, and those with large office complexes, the platform encompasses a suite of technologies and expertise to implement smart buildings, fitting all kinds of sizes and requirements. And part of the appeal of the platform is that it sems to seamlessly integrate with common access control systems.

Sheedy summarises the system as one that is dedicated to making buildings smarter, greener, cheaper, and generally better overall. “It’s all about the simple solutions to make the biggest impact, for the bottom line and for the planet.”

Occupancy and weather data allows HVAC systems to be adjusted for optimal effifiency.