Case Studies | SaaS B2B

Maximising site value with AI

ML House Placement Prediction

About

A SaaS B2B business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The engagement had three focus areas, each essential to turning a manual placement tool into a trustworthy recommendation engine.

Reasoning over real site geometry

A plot is not a tidy rectangle. Each site arrives as a polygon, sometimes several, defined by precise coordinates, and any recommendation has to respect that exact shape. The solution had to place dwellings inside the real boundary of the land rather than an idealised footprint.

Encoding planning compliance

Every site carries rules that limit what can be built, from a permitted density of dwellings per hectare to the requirement that schemes over five homes allocate a proportion to affordable housing, which generates significantly less revenue. Those rules had to be applied automatically, not checked after the fact.

Optimising for developer value

Within those constraints the goal is commercial: choose the dwelling types and the allocation of each that maximise the revenue the developer can realise from the plot, balancing optimum floor area, build cost and land value across the mix.

Solution

The engine takes three inputs for a given site: the site shape expressed as a GeoJSON polygon, the planning constraints for that location, and a catalogue of the dwelling types that can be placed there. These are assembled into a structured prompt and sent to a large language model running in Amazon Bedrock, which reasons about how to fit and combine dwellings within the boundary while honouring the planning rules.

14.49

Dwellings per hectare density modelled per site

40%

Affordable-housing allocation enforced on schemes over five homes

Ranked

Optimum property mix returned for each plot

By the numbers:

  • 14.49 - Dwellings per hectare density modelled per site
  • 40% - Affordable-housing allocation enforced on schemes over five homes
  • Ranked - Optimum property mix returned for each plot
Changes

Cloud Combinator delivered a working placement engine that ingests a site polygon and its planning constraints and returns a ranked, revenue-optimised property mix, moving the client from manual, developer-led placement to an AI-generated recommendation grounded in the real shape and rules of each plot.

  • Geometry-aware placement — The engine reasons over the exact site polygon, so recommendations respect the real boundary of the land rather than a simplified footprint.
  • Compliance built in — Density limits and the affordable-housing requirement for larger schemes are applied automatically as the mix is chosen, not bolted on afterwards.
  • Value-led recommendations — Each recommendation carries modelled revenue, build cost, land value and allocation, so the developer can see the commercial case behind the suggested mix.
  • Handover — The client received a Bedrock-based placement engine and the prompt and data patterns behind it, ready to fold into the wider appraisal platform.

With an AI placement engine proven on real sites, the client is positioned to make optimum, compliant scheme design a native part of every appraisal, so developers can see the highest-value use of a plot from the first click.

AWS Stack

Amazon Bedrock

For large language model reasoning that fits and combines dwellings within a site boundary and its planning rules.

Amazon S3

For storing site geometry, planning constraints and property datasets that feed the engine.

AWS Lambda

For orchestrating the placement pipeline from input site to ranked recommendation.

YOU MIGHT LIKE

Related success stories

View all case studies

Case Studies | Insights

Utilising Language Recognition, Speed, and Enhanced Security to Make Social Media a Force for Good

  • Here, we take a detailed look at how the Cloud Combinator team collaborated with another cutting-edge AI service provider that provides intelligent systems to “make social media more social” for brands and users alike.
  • Arwen AI is a UK-based startup specialising in AI solutions to manage and enhance brands’ social media interactions. Founded in 2020 by Matt McGrory, Dr. David Cole, and Joel Bailey, Arwen. AI focuses on using AI to automatically detect and remove spam, toxic comments, and other unwanted content from social media platforms.
  • The team at Arwen have three core products. ‘Moderate’ is focused on identifying and removing toxic content from social media channels. ‘Engage’ helps brands identify and engage with meaningful conversations on social media, and ‘Customize’ allows brands to apply bespoke algorithms to their channels - creating an even more effective moderation and engagement.
Read more
CONTACT US

Ready to turn AI into impact?

We'll help you spot the highest-value opportunities, reduce risk around your first AI initiative, and define a clear path to results from day one.

Why talk to us:

Outcome-driven recommendations

AWS-recognised delivery expertise

Risk-aware AI adoption

Clear next step, not a sales pitch

Start with a focused 20-minute conversation about your goals — no pressure, no commitment.

We use cookies

This website uses cookies to enhance user experience and to analyze performance and traffic on our website.

See our Privacy Policy for details.