Finding relevant planning information can be such a chore. So we made it easier. Introducing PAI (Planning Appeal Intelligence)
Current planning application and appeal finders rely heavily on keyword searches. This makes locating relevant cases laborious and time-consuming, as users often have to sift through mountains of irrelevant data before finding a useful precedent. For many, this makes crucial planning insight inaccessible or prohibitively expensive for smaller projects
In collaboration with David Sinclair, we have developed PAI (Planning Appeals Intelligence). Unlike standard AI models, PAI’s knowledge base is exclusively sourced from live planning appeal decision history. This creates a high-level, reliable, and up-to-date database designed to provide accurate and relevant results quickly.
Why We Developed PAI
Navigating the UK planning system has often felt like a "dark art," but recent shifts have made the stakes even higher. For small-scale projects and ambitious homeowners, the barrier to entry is steep. Professional consultancy fees can frequently exceed 15% of total project costs, leaving many to navigate complex policies without the proper data tools.
The problem is two-fold: accessibility and accuracy. Current databases rely on outdated "lexical" keyword searches that ignore the nuanced logic of planning policy. You might spend hours digging through hundreds of irrelevant results while the critical insight remains buried. While general AI models are popular, their training is often too broad for the surgical precision required for planning; they can "hallucinate" information, which is a risk no project can afford.
With the April 2026 "Submit Once" reforms, a poorly prepared initial application is now a "dead end." Because new evidence is largely barred at the appeal stage, an initial refusal is significantly harder to overturn. With appeal success rates hovering at just 33%, there is an urgent need for a tool that de-risks projects by injecting technical intent from day one.
How We Did It
At Studio Bark, we believe any planning or design challenge is an opportunity to develop inspiring, future-proof architecture. To tackle the "Residential Advice Gap," we collaborated with David Sinclair to develop PAI (Planning Appeals Intelligence). Our goal was to create a tool that moves beyond keywords and instead focuses on the technical logic that actually wins cases.
Unlike standard AI models, PAI is built on a high-level, reliable database sourced exclusively from live planning appeal decision history. What makes PAI truly unique is its ability to identify the exact logic used by Inspectors to rule on cases. It doesn't just return results; it provides answers to complex planning queries based on real-world outcomes.
What The Tool Can Do
Finding appeals faster with PAI
Current government databases rely on Lexical Search (keywords). If you search for "overlooking," you miss cases that use the term "loss of privacy" or "inter-visibility." PAI understands the concept of a planning conflict. It finds cases based on the Technical Intent of the inspector’s logic, not just the words they used. This improves the efficiency of the search and reduces time spent reviewing search results.
Gaining planning insight with PAI
Due to its intelligent search methods, PAI is able to give answers to complex planning queries. It does this by gathering information from appeal reports from both approved and dismissed appeal cases, and uses this to build its evidence based response. These references are shown within its response in order to maintain an audit trail for the user to monitor.
Writing planning statements with PAI
PAI is able to use its database of planning appeal documentation to assist in the drafting of planning statements. PAI is able to reference identified planning risks against policy and write an argument for why it should be included within the proposal.
How It Helps Your Practice
For architectural SMEs and planning consultants, PAI transforms the research process from hours of manual PDF screening into seconds of targeted retrieval.
- Professional Confidence: By providing direct citations to real PINS cases, PAI allows architects to satisfy professional liability requirements and submit applications with confidence.
- Increased Productivity: Small practices can now access high-level planning intelligence that was previously financially non-viable, allowing them to act as their own expert guides through policy nuances.
- De-Risking Applications: By identifying the exact logic used to overcome constraints in the past, PAI helps you navigate the nuances of planning policy and avoid refused applications.
Get Involved
We spent the first half of 2026 refining PAI to be usable in real world application and we will shortly be launching our first beta trail. The outcomes from this trail will help us to develop the tool further and increase its usability.
If you would like to be involved in the beta, please click here to fill out a quick google form so that we give you access.