Why it holds up in front of a judge
Every figure on a report is the output of a rule you can read and a document you can pull. This page is the rules. If any of them is wrong, the report is wrong in a direction you can predict, which is the whole design.
How a claim gets retired
A mortgage is a claim until a satisfaction retires it. Both are recorded documents with their own file numbers, and the register carries a reference from the second to the first. That reference is what does the work.
In the sample report, a $780,000 Chase mortgage was recorded in 2004, and a satisfaction recorded in 2011 names it. (The sample's file numbers are masked, because the sample is not a real parcel.) So the 2004 mortgage is not debt in 2026, and it is shown struck through with both file numbers printed, rather than counted against you or silently deleted.
Across the four boroughs, 43.8% of all recorded claims carry a satisfaction like that one. A search that does not resolve them reports roughly twice the debt that actually stands. That is not a rounding problem, it is the difference between a case worth filing and a case that looks underwater.
Where the linking thins out When the register cannot prove a satisfaction, and what the report does about it.
Older satisfactions do not always link, so some historic mortgages still read as live even where a paper satisfaction exists. Any parcel whose verdict turns on one of those is marked on the report, so you know when the answer depends on a link we could not make.
Why the number is a floor and not an estimate
We could report an estimate of the current balance on each live lien. Nobody publishes balances, so that estimate would be a model output. We report the recorded face amount instead.
A mortgage recorded at $920,000 in 2011 has almost certainly been paid down. We count the full $920,000 anyway. The reachable figure that comes out the other side is therefore at or below the truth, never above it. The same rule runs everywhere it can:
- Debt counts at its recorded amount, because no recorded document shows a payment.
- A lien we cannot prove was released stays on the report as live.
- A parcel with any unsearched register returns no recorder data, never unencumbered.
Each of those costs you a case you might have won, and none of them sends you to levy on nothing. That is the asymmetry the whole system is built around, because the two errors do not cost the same: a missed case costs you an opportunity, and a levy on a fully claimed property costs you filing fees, marshal fees and your client's confidence.
What is not yet measured The direction of the error is argued. The size of it is not.
The direction of error is argued and stated on every report. The magnitude is not yet quantified. Validation against title searches is in progress, and the result goes on the limits page whichever way it falls.
How the legal screens fire
Florida homestead is not a judgment call in this system. It is a rule engine running the tests in Article X, Section 4 of the Florida Constitution: acreage, residency and the form of ownership.
Each rule that fires prints its authority beside the verdict, so the report tells you not just that a parcel screened as homestead but which test carried it and on what evidence. A determination a model produced is not defensible in a filing. A rule with a citation is.
How fraudulent transfer reachback is handled A deed that follows a judgment inside the window, cited to the statute rather than flagged as suspicious.
The same applies to fraudulent transfer reachback. When a deed follows a judgment inside the window, the report cites FS 726.110 and states how much of the four years remains, rather than telling you a transfer looks suspicious.
Lead, finding, and the difference
A finding is resolved by identifier. A lead is matched on something weaker, usually a name, and the report says so on its face.
The Broward transfer example is a lead, not a finding. A judgment against one spouse, then nineteen days later a deed for one dollar into tenancy by the entireties. The pattern is exactly what judgment proofing looks like. It is also exactly what adding a spouse to a title looks like, and that is usually lawful. So the report gives you the documents, the reachback window, the name frequency across 48 years of the county index, and the label. It does not give you a conclusion.
Ownership links carry the same distinction. Where a debtor is tied to an entity by a registry identifier, the report says resolved. Where the tie is a name match, it says matched by name alone and tells you how many people in that jurisdiction share the name.
The join that makes Broward different
In most counties, pairing a judgment with a later conveyance means matching two records on a name and hoping. Broward's index lets the two be tied on the record itself.
So the pairing is checkable rather than a guess, and every lead it produces can be verified against the docket rather than taken on trust. That is why the transfer product exists in Broward first and not somewhere larger.
Where the data came from
Every source is the office that keeps the record.
No paywall circumvented, no private system entered, no aggregator feed whose terms would restrict what you do with the result. Where a source's own terms restrict commercial use, we hold it back rather than serve it. That last one costs us coverage and we have kept to it anyway, because a report you cannot use freely is not worth holding.
What a report will not tell you
An empty list is not an all clear. Where a register was never searched, the report returns no recorder data rather than unencumbered. Telling you a property is clean when nobody looked is the one failure this system refuses.
Send a name from a case you have closed. We run it and you tell us whether we were right. If we were not, you have lost an email.