Commercial real estate operator
Acquisition intelligence
Deployed and run at scale during a completed engagement. Bedrock does not operate it now and makes no claim about its use since.
Jurisdiction research stopped living in a browser tab and became a record the next parcel reads from.
The hunt that started over every time
Every candidate parcel began the same hunt. Find the authority that governs it. Find out whether that authority publishes anything a person can rely on. Read the ordinance. Decide whether the use is permitted, or argue about it.
None of those answers were written down anywhere the next parcel could read them, so the second site in a town cost what the first one had cost, and a reading nobody recorded was argued again a month later.
A county is not one rule
A county is a set of towns and the towns do not agree. Some publish an ordinance and map layers; some publish nothing a system can read. The register keeps each authority separately, so a town that has been researched stays researched, and a town where nothing has been verified is visible as exactly that rather than quietly treated as empty.
- Source verified and in useA · B · D · F
- Nothing verified yetC
- Withheld by the readiness auditE
- Candidate parcelOne town holds it
A source that failed is kept with the reason it failed, so it is penalised on every later search rather than tried again as new, in this county or any other. A person verified every source before it was used. Local words for the same use are reconciled in an alias list rather than in code, and a reading a person deletes on purpose stays deleted: the automated reseed cannot bring it back. Before a town could be called ready, an audit scored its data against published thresholds and withheld it if any failed.
Three sources tried against one authority
- VerifiedThe county parcel serviceIn use for every parcel in the county
- RejectedRegional zoning mirror, 2019Attributes belong to the neighbouring county. Kept on the deny-list, and penalised on every later search
- Not yet verifiedTown C planning pageNobody has passed it, so Town C is shown as unverified rather than treated as empty
An acquisition analyst opens the county with a list of candidate parcels to get through and a set of towns whose ordinances nobody can hold in their head.
A parcel the rules cannot settle
The buying criteria are stated once and run against every parcel rather than the three sites somebody had time for. A hard failure excludes a site outright. A weaker concern demotes it and carries the reason with it. And where the ordinance does not answer the question, nothing is scored at all.
The analyst’s buy box surfaced a highway-commercial parcel on Frontage Road for self-storage. The ordinance permits self-storage in that district only where every bay is internal to the structure and no outdoor storage sits within fifty feet of a residential boundary — and the record does not say which this parcel is.
Where it is unclear, a person reads it
A system of this kind can always produce an answer. The question is what it does when the evidence does not support one. In the sequence above the ordinance did not settle the question on its own, so the row waited in the queue for a person, with the section quoted beside it and its confidence stated rather than the verdict being treated as settled.
The harder case is a contradiction. The parcel record carries one zone code and the district that contains the parcel carries another. The verification panel does not pick one. It resolves to Conflict, and the parcel goes nowhere until an analyst has read both layers and recorded a verdict — stored as human-recorded, written back to the town, and safe from the next reseed.
Which district a parcel is bound to is part of the record for the same reason. A parcel is tied to the district that contains it rather than the district nearest to it, and the record says which. That distinction decides whether everything downstream is about the right piece of ground.
- Authority
- Mill Bend
- District
- C-3
- Bound
- Contained, not nearest
- Size
- 3.41 acres
What the next parcel inherits
What comes off the work is repetition. The first parcel in a town pays for the reading. Every parcel after it reads what was recorded — including the source that was rejected, and why.
| The first parcel in a town | Every parcel after it |
|---|---|
| Find who governs it | Read it from the register |
| Hunt for sources and test each one | Read the verified source, skip the rejected one |
| Read the ordinance yourself | Read the stored reading and the section it came from |
| Guess, or go and ask somebody | Ask only where nobody has settled it yet |
Four parts ran; one was only built
Four of these parts ran in production during the engagement. Research retention kept sources, readings and citations per jurisdiction rather than per search. Buy-box screening scored the whole candidate set against a written buy box. The adjudication queue took batch decisions over a backlog that row-by-row review could not clear. The opportunity digest matched listings to parcels and sent them on a schedule, to one recipient.
The fifth part is market context — competition, population and hazard. It was built and wired, and routine use of it is not evidenced.
Research time per parcel was never measured, so what this removed is repetition rather than a saving anyone timed. Figures for how much of a county had been researched are not published here either: the system’s own definitions of researched and ready are stricter than the readings those figures invited.