We’re very pleased to share that Thesqua.re has published a peer-reviewed scientific research paper on EcoGrade, our sustainability metric.

What’s It All About?

For the last couple of years, we have been hard at work creating an industry-leading sustainability score. We use it to show how each apartment measures up in terms of four green credentials: its efficiency (has it got double glazing, efficient lighting, etc), its energy consumption (less energy usage is better), its access to green transport, and whether its power is supplied by renewable energy source. From these data, we also calculate the CO2 score.

In many cases, consumers understandably get confused when every apartment of a similar type in the country shows the same CO2 score. How can one choose a sustainable option if all two-bedroom apartments show the same CO2 value?

The EcoGrade is different. We use official government-approved data. We carefully match the data to the specific address in question. We spend a lot of time worrying about every detail. 

For example, when we calculate the availability of green transport, we look at all forms nearby: metro train stations, electric vehicles, and electric scooters. We have data on exactly where these may be at any given time. 

But what if all the electric scooters happen to be far away on the day we calculate the score? We check their locations on multiple days at different times and calculate the average distances from the apartment in question. 

We even take into account the curvature of the Earth when we calculate how long it takes to walk from the apartment to the transport!

Sometimes there is missing data. Not every apartment has exact data describing its sustainability. Our solution is to use mathematical interpolation to figure out what the data should be, looking at similar neighbouring properties. 

But just figuring out similarity is complicated – the apartment listings usually tell us the number of rooms but do not have accurate floor areas, while the sustainability data usually has floor areas but not the number of rooms. 

So we have to calculate mappings between them, specific to each region, using known data. How do we know all our maths is correct? That’s where this research paper comes in.

We created ten thousand random residential addresses, in ten UK cities, to test EcoGrade scores on. We ensured the addresses comprised different sized dwellings in different areas, and that some had good data, while others didn’t. 

In this way, we simulated real rental listings but removed any biases that rental listing data alone might have. We showed using some fancy statistics that any time we derive data to fill in missing values, our derivations are so close to reality that you can’t tell the difference.

But the cool thing about this research is not only did we prove EcoGrade is as accurate as we can make it… we also now have the first-ever comparisons between properties in the UK. For the first time, we can tell you what kind of apartment is more sustainable, and where more sustainable options may be located.

EcoGrade belongs to Thesqua.re. It’s our proprietary metric and we’re proud to say it is the best in the industry. Only on Thesqua.re you can see sustainability information that’s based on real government-approved data specific to the exact property with this level of attention paid to maintaining accuracy.

Who Developed It?

Our talented team of developers and advisors: Rajat Mathur, Shrey Jindal, Prof Peter Bentley , Dr Soo Ling Lim, and Beth Foster.

Where Was the Work Published?

The paper is called: Address-Specific Sustainable Accommodation Choice Through Real-World Data Integration .

It was published in the IEEE International Conference on Intelligent Data Science Technologies and Applications (IDSTA2024) and was presented by Peter Bentley at the conference in October 2024. In addition, Peter was also invited by the organisers to give the first keynote lecture for the whole conference.

Why Publish It As A Scientific Paper?

Instead of writing a “white paper”, we published our work in a peer-reviewed conference where independent experts ensure our claims are valid before it is accepted. Hence, proving the algorithm of our model is valid. This is why you can trust what we say.

What’s Next?

While this work focused on the UK, we have already rolled out EcoGrade for other countries, including the USA, France, Spain and the Netherlands. It’s an ongoing project because we are trying to do this right. 

We always try to find the best data sources for each country, and we spend a lot of effort cleaning and understanding each dataset (because each country measures different things in different ways).

And we also have one other trick up our sleeves. In countries where there is no data at all, what can anyone do? The answer is AI. Check out the next research briefing, Decoding Serviced Apartment Buildings with AI to find out more!