A conclusion is only as useful as the evidence behind it.

A postcode can contain several homes, and similar addresses can represent a house, a flat, a subdivided unit or a neighbouring plot. Propeteer begins by resolving the address and stable property identifiers before joining records from separate sources.

Where identity is uncertain, the result should remain uncertain. Historic transactions or certificates are not silently merged merely because their address text looks similar.

• Use exact address components and stable identifiers where available
• Keep numbered, suffixed and subdivided properties separate
• Surface material changes in property type or tenure for review

A planning decision, recorded sale or mapped boundary is evidence. A valuation range, likelihood assessment or development conclusion is an interpretation built from that evidence.

Propeteer keeps the underlying record visible and labels estimates as estimates. The purpose is to make a conclusion easier to understand, challenge and update—not to replace professional judgement with an unexplained score.

• Show the records used to support a conclusion
• Separate historic index movement from current market evidence
• Do not present modelled outputs as official determinations

Nearby does not automatically mean comparable. Proposal type, building form, scale, neighbour relationship, street character, policy context and decision date can all change the significance of a planning application.

The strongest precedents are those that resemble the proposed work and retain enough readable material to explain the decision. Conditions, refusal reasons and pending cases remain separate because they answer different questions.

• Compare the proposed work as well as the distance
• Treat approvals with conditions separately from unqualified approvals
• Exclude pending cases from decided approval rates

A previous sale indexed to today describes general market movement; it does not know what happened to the property. Recent comparable sales add current evidence, while floor area, property type, tenure, condition and micro-location affect how much weight each transaction can carry.

Propeteer combines compatible signals and explains disagreement. Weak or missing evidence should widen the range and reduce confidence rather than create false precision.

• Prioritise recent, similar completed transactions
• Use floor-area benchmarks only when the match is reliable
• Keep asking prices as context rather than completed-sale evidence

Schools, transport, crime, flood and environmental records can help frame a property decision, but each dataset has its own geography, publication date and limitations.

A mapped feature may be on the property, nearby or merely within the selected search radius. No result can also reflect unavailable coverage. Those states should remain distinguishable.

• Show the search radius and distance from the address
• Distinguish an on-site match from a nearby record
• Direct material findings to the relevant official search or adviser

Completeness varies between councils, datasets and individual properties. Confidence should reflect identity quality, evidence quantity, recency, comparability and whether the underlying documents can be read.

A useful result can be incomplete as long as the limitation is clear. Propeteer distinguishes a completed check, no matching result, unavailable coverage and a failed connection so missing information is not mistaken for reassurance.

• Retain dates and coverage information
• Explain important exclusions and failed matches
• Lower confidence when evidence is sparse or contradictory