‍Arebray is a development-focused firm of Chartered Surveyors that advises developers and landowners. Its core work is development viability, helping developers make schemes stack up by adjusting affordable housing, housing mix or tenure. Its schemes vary widely in size, from small sites of around a dozen plots to large sites of over 1,000 units.
We spoke with Richard Bray, founder of Arebray, about how Searchland's MCP has changed what goes into every report.
Challenges before Searchland's MCP:
- Only 20 to 30 comparables could be analysed manually for each 30-hour report timeline
- No ready-made data separating new-build sales from existing stock
- No time to evidence trends in a commercial report, such as how price per square foot changes with property size
‍
Same hours, far deeper analysis
Most conversations about AI focus on saving time, but Richard has gone down a different route. Richard takes around 30 hours to complete each viability report for his clients. Rather than finishing reports in less time with the same amount of depth, he’s put the extra capacity into the depth of the work.
"When you're doing it manually, you could probably analyse 20 or 30 comparables in that time frame. Now, using the MCP tool, I can do at least 400-500 in a similar window."
Richard still picks out the strongest comparables by hand and bases his values on them. The difference is that he now starts from a data set more than ten times larger.
‍
Analysis that wasn't possible before
With Searchland's MCP connected to his Claude AI Assistant, Richard has built his own skills and agents that pull Searchland data and present it in the same format every time. That has opened up analysis he simply couldn't fit into a report before.
One example is new-build comparables. Data that splits new builds from existing stock doesn't really exist on its own, so Richard creates it. He cross-references when planning consents were granted against when new postcodes were created, then matches that against sales data to isolate everything built in the last 10 years.
"Having core, reliable data is the important thing. When you add in AI capabilities, it’s very powerful."
The bigger data set also lets Richard show his workings. In valuation, the "quantum effect" means larger homes sell for less per square foot than smaller ones. With hundreds of comparables to hand, Richard can now plot them on a scatter graph with a trend line, giving clear justification for the rate he applies to a 1,000 or 2,000 sq ft home.
"Although the theory has always been there, I've never been able to physically show it in a report. If I've already got the data there, running an extra graph makes the report look that much better at no extra cost."
‍
Accuracy Arebray can stand behind
Training AI to do exactly what he needs, and nothing more, has been a learning process. Richard wants the same table of results every time and confidence that it's correct, so he ran the new workflow in parallel with his manual process for two months before rolling it out fully.
He has also built a separate proofreading agent that checks every piece of work, running on a different AI model from the one that produced it.
‍
What's next for Arebray?
Rather than using AI to cut hours, Richard is betting on quality, backed by a loyal client base, many of whom have worked with him for more than 10 years.
He is also still finding new uses for the MCP. Once a week, an automated search runs through the previous week's planning decisions and appeals and flags anything related to viability, steadily building a database of relevant cases. New requests keep coming in too, such as a client looking for sites within 800 metres of a railway station.
‍
"I'm still finding new things to do with it. As I sharpen that side of things, I'll find more and more uses for the MCP and the data that comes from it, because there's a whole load of data that I'm probably not even using at the moment."
Next, Richard plans to try Searchland's in-platform AI to speed up his SHLAA cross-referencing.
We look forward to supporting Arebray as Richard keeps pushing what's possible with AI and Searchland data.
Quotes from Richard Bray, Founder, Arebray
‍
Want to see how Searchland's MCP can power your own reports? Book a demo with our team today.
%20(1).webp)
