AI Helps Lewisham Identify Over 3,000 Potential Housing Sites, Lewisham 2026

Key Points

• Identification of Sites: Lewisham Council identified more than 3,000 small locations with housing development potential by utilizing an artificial intelligence mapping tool.
• Government Support: The pilot project was funded through the 2024 PropTech Innovation Fund and published as a official case study by the Ministry of Housing, Communities and Local Government (MHCLG).
• Technical Collaboration: The software, named the Small Sites AI Finder, was developed in partnership with architecture and urban design practice RCKa.
• Methodological Shift: The tool addresses traditional reliance on manual officer reviews and localized memory, providing a standardized borough-wide land assessment.
• Retention of Human Oversight: AI-generated results do not automatically grant planning status; planning officers must conduct manual reviews and detailed assessments before any site moves forward.
• Policy Integration: Alongside spatial mapping, Lewisham reviewed planning application data to measure the real-world impact of its Small Sites Design Guide Supplementary Planning Document.

Lewisham (South London News) August 5, 2026 – Lewisham Council has uncovered more than 3,000 small sites with potential for residential development across the borough after piloting an artificial intelligence mapping tool designed to streamline planning operations and accelerate housing delivery. The trial, formally published as a case study by the Ministry of Housing, Communities and Local Government (MHCLG), was executed under the government’s 2024 PropTech Innovation Fund. By leveraging spatial data algorithms, the initiative demonstrated how digital automation can assist local planning authorities in identifying development capacity at scale while establishing a robust, evidence-based foundation for future municipal policy.

The core technology behind the project, known as the Small Sites AI Finder, was created in partnership with architecture and urban design firm RCKa. The digital tool was developed to analyse land patterns, site dimensions, spatial boundaries, and contextual constraints across the entire London borough of Lewisham. The resulting dataset provided local authority planners with a significantly broader picture of land availability than previously achievable through traditional analytical methods.

Alongside the spatial AI analysis, Lewisham Council evaluated planning application datasets to judge the effectiveness of its existing Small Sites Design Guide Supplementary Planning Document. This dual approach allowed the local authority to measure how historical policy guidance influenced actual housing delivery while simultaneously mapping future infill opportunities.

Why Was the Traditional Manual Planning Assessment Method Proving Inefficient?

Historically, local authorities across the United Kingdom have faced operational hurdles when identifying small-scale housing opportunities. Planning teams traditionally relied heavily on the individual geographic knowledge of planning officers, site visits, and piecemeal manual reviews.

According to the government case study, this manual approach introduced several systemic limitations:

• Inconsistent Scope: Officers could not systematically evaluate every parcel of land across an entire borough simultaneously, leading to overlooked gaps in land availability.
• Resource Intensity: Desktop reviews and physical inspections required hundreds of officer hours, stretching local authority capacity.
• Fragmented Evidence: Gathering comprehensive data for local plan policies or housing land availability assessments was difficult to standardize across different wards.

The introduction of the Small Sites AI Finder enabled spatial data across Lewisham to be scanned continuously under uniform criteria. Planners could filter locations based on physical characteristics and preliminary constraints prior to committing resources to detailed on-site evaluations.

What Role Does Human Expertise Play in Evaluating AI-Generated Housing Outputs?

Despite the scale of the automated output, the Ministry of Housing, Communities and Local Government emphasized that the technology serves as a supportive mechanism rather than a replacement for professional human judgment. The generation of over 3,000 site leads represents an initial evidence-gathering phase rather than formal development allocation.

Each AI-identified site remains subject to rigorous manual evaluation by qualified planning officers. Professional reviews are required to verify ground conditions, heritage protections, local transport access, environmental designations, and overall site suitability before any location can be formally considered within local planning frameworks or housing land supply targets.

The technology acts primarily as a digital filter, eliminating the initial administrative burden of manual map searches while preserving the statutory decision-making duties of the local planning authority.

How Does This Pilot Fit Into the Government’s Broader Digital Planning Strategy?

The Lewisham case study represents one element of a wider effort by central government to digitize the planning system through the PropTech Innovation Fund. The scheme provides financial and technical backing to local councils testing digital tools capable of modernizing planning procedures, improving public engagement, and speeding up development decisions.

The initiative operates alongside complementary digital projects across the public sector, such as the MHCLG ‘Extract’ tool, which is designed to digitize decades of legacy local authority planning records. By converting historical unstructured documents and physical maps into machine-readable spatial data, government initiatives aim to create a national planning dataset that reduces friction for both public sector planners and housing providers.

What Is the Background of the PropTech Innovation Fund Initiative?

The PropTech Innovation Fund was launched by central government to accelerate the adoption of technology across local planning authorities in England. Established under the Ministry of Housing, Communities and Local Government, the fund has distributed multiple rounds of grant funding to municipal councils to pilot emerging technologies, including 3D spatial visualization, automated constraint mapping, and digital public consultation platforms.

Historically, local council planning departments have operated using legacy software and manual, paper-based administrative records. This structural inefficiency contributed to delays in preparing Local Plans and assessing housing land availability. The 2024 tranche of the fund specifically prioritized tools capable of unblocking housing supply, enhancing spatial data transparency, and supporting small and medium-sized (SME) housebuilders through clearer site identification frameworks. Lewisham’s deployment of the Small Sites AI Finder alongside architectural practice RCKa emerged directly from this funding stream, aiming to create scalable technical models that could be replicated by other local authorities across the country.

Explore More Lewisham Council News

Lewisham Council Rejects Home Office Immigration Raids

Lewisham Council votes to refuse assisting Home Office immigration raids

What Are the Future Implications for Local Planning Authorities and the Housing Sector?

The widespread adoption of AI-driven spatial mapping tools is poised to alter how local planning authorities, land buyers, and suburban communities approach urban housing supply.

For local authority planning departments, the integration of algorithmic land scanning will likely drastically shorten the timeframe needed to compile Strategic Housing Land Availability Assessments (SHLAAs). By automating the discovery of infill land, brownfield plots, and underutilized urban spaces, planning teams can maintain up-to-date land supply pipelines with lower operational overhead. This enables councils to make evidence-based policy decisions and defend housing targets during public examinations of their Local Plans.

For small and medium-sized developers, standardized AI site identification could lower barriers to entry. Small sites often represent the primary target for regional SME builders, yet identifying viable parcels has historically required substantial speculative investment. If local authorities publish verified datasets of suitable small sites, development risk is reduced, potentially incentivizing smaller construction firms to re-enter urban housing markets.

For local communities and residents, the transition toward digital site identification increases transparency regarding potential urban density changes. However, it also places greater emphasis on early consultation mechanisms, ensuring that automated land identification is balanced with local infrastructure constraints, green space protections, and neighborhood amenities.