Executive Summary: There is no single correct interval, and any vendor offering one is selling a default rather than an answer. The right cadence is set by the shortest clock you are actually subject to, and for most secured lenders that clock is four months, not the twelve most portfolios default to. This article works out where the number comes from rather than asserting one.
Why does re-check frequency matter more than data quality?
Because a Secretary of State record is accurate on the day you pull it and says nothing about any day after that. The data is not the variable. The staleness is.
A borrower verified at origination is verified as of that date. Everything that follows, a status change, a dissolution, an officer change, a name change, happens in a record nobody is reading. The gap between when a change lands in the public record and when a lender notices it is not a data-quality problem, because the information was public and correct the whole time. It is a scheduling problem.
That reframing matters for how the decision gets made. Choosing a monitoring cadence is not a question of how good your data source is. It is a question of how long you are willing to be wrong before finding out, and that is a credit policy decision with a cost on both sides.
• Too infrequent and you inherit the consequences of changes you could have acted on. The loss is not caused by the change; it is caused by the delay.
• Too frequent and you spend credits and analyst attention on records that did not move. Most records do not move in most windows.
• The wrong cadence for the wrong borrower is the common failure. One interval applied uniformly overspends on stable exposure and underspends on the risky part of the book.
What is the shortest clock a lender is actually subject to?
This is the question that produces a number, and for secured lenders the answer is unexpectedly short.
Under UCC Article 9, if a debtor's name changes so a filed financing statement becomes seriously misleading, the filing perfects collateral acquired "before, or within four months after" the change, and does not perfect collateral acquired more than four months after unless an amendment is filed within that same four-month window.[1] The clock runs from the change itself, not from the date anyone tells you about it, and industry guidance is consistent that catching it is the secured party's job rather than the debtor's.[2]
Four months is the binding constraint, and it produces an uncomfortable arithmetic result. A monitoring cadence of exactly four months does not protect you, because a change occurring the day after a check is not seen until the next one, by which time the window has closed. To reliably act inside a four-month window, the interval has to be meaningfully shorter than four months. Monthly gives comfortable margin. Quarterly gives very little. Annual gives none at all.
Set against that, the other clocks in this area are long. Reinstatement windows after administrative dissolution run to three years in Texas for retroactive effect,[3] are open-ended in Florida,[4] and cap at five years in Georgia.[5] Those are generous by comparison, which is precisely why the UCC amendment window rather than the reinstatement window should drive the cadence for secured exposure.
Most portfolios monitor annually because annual review is the existing rhythm of the business, not because anything in the underlying law runs on a twelve-month cycle. The shortest clock is four months and it belongs to the lender, not the borrower.
What does the state record's own update rhythm imply?
There is a second constraint working in the opposite direction, and ignoring it wastes money.
Some fields change only when a periodic report is filed. Ownership and officer information is the clearest case: Texas states there is "no filing requirement with the secretary of state when there is an ownership change" for either corporations or LLCs, so that information refreshes when the next report is filed rather than when the change occurs.[6]
Checking a periodic-report field daily cannot produce daily information, because the underlying data changes at most once a reporting cycle. Other fields behave differently. Status changes and registered agent resignations post when they happen. A registered agent's resignation terminates the appointment "on the 31st day after the date the secretary of state receives notice," which is a real event on a real date, unrelated to any report cycle.[7]
So the record contains two kinds of fields:
• Event-driven fields. Status, registered agent, dissolution, name changes. These post when they happen and reward frequent checking.
• Cycle-driven fields. Officers, managers, and members. These refresh on the periodic report and reward well-timed checking rather than frequent checking.
• Fields the record never captures. Equity transfers in corporations and LLCs. No cadence detects what is never filed.
The practical implication: frequency serves event-driven fields, and timing serves cycle-driven ones. A check placed just after a borrower's periodic report is due extracts more from the officer fields than four evenly spaced checks that all happen to miss the filing.
How should cadence vary by borrower?
Uniform cadence is the most common design and the least defensible one. Tiering costs little and improves both sides of the trade.
The variables that should move the interval:
• Secured versus unsecured exposure. Secured lending carries the four-month UCC clock. Unsecured does not, which genuinely justifies a longer interval.
• Exposure size. The cost of a check is 1 credit. The cost of missing a change scales with the balance. Large exposures justify short intervals on arithmetic alone.
• Remaining term. A loan with two months left has limited room for a change to matter. A five-year facility has a great deal.
• Entity age and filing history. A business that has lapsed and reinstated before is more likely to lapse again. Its own filing history is the best available predictor.
• State of formation. Reinstatement windows and reporting cycles differ, and biennial-report states leave officer data stale for twice as long.
• Whether anything has already moved. A borrower that produced a change last quarter warrants a shorter interval than one that has been static for three years.
A workable default structure is a short interval for secured or large exposures, a moderate one for the general book, and a long one for small, unsecured, short-remaining-term loans, with any borrower that produces a change moving up a tier automatically until it goes quiet again.
The automatic escalation rule deserves emphasis, because it does most of the work for very little design effort. Entity-level problems cluster in time rather than arriving independently. A business that misses one periodic report is disproportionately likely to miss the next, because the cause is usually structural: an address the state cannot reach, a departed bookkeeper, or a decision to stop paying small recurring obligations. None of those resolve on their own between one filing deadline and the next.
That means the single best predictor of whether a borrower will produce a change in the next ninety days is whether it produced one in the last ninety. A tiering model that adjusts on observed behavior therefore outperforms one built purely on static attributes like exposure size or entity age, and it does so without anyone maintaining a scoring model. The rule is simply that movement earns attention and silence earns less of it, which concentrates spend on the part of the book where something is already happening.
Want to see how configurable Secretary of State re-checks fit into a portfolio workflow? Book a demo.
How does the right cadence differ by lending product?
The four-month constraint is general. How much margin you need against it depends on what you are lending and how quickly you could act on what you find, and that varies enough by product to be worth working through.
Merchant cash advance and short-term working capital. Terms are measured in months, so the total exposure period is often shorter than a single quarterly interval. That makes frequency at origination and early in the term far more valuable than frequency later. The practical design is a dense schedule during the first weeks, when the risk of a recently manufactured or recently acquired entity is highest, tapering as the balance amortizes. An annual cadence on a nine-month product checks the borrower approximately never.
Equipment finance and secured term lending. These carry the full four-month UCC exposure across multi-year terms, and the collateral is specific and identifiable. This is where monthly checking is easiest to justify on arithmetic and hardest to justify skipping. It is also where the name-change trap does the most damage, because after-acquired collateral provisions are common and a lapsed filing affects exactly the collateral acquired after the gap opened.[1]
Lines of credit and revolving facilities. The distinguishing feature is that new advances happen throughout the term, so the question is not only whether the borrower is sound but whether it is sound at the moment of each draw. Cadence matters less here than event triggering: a check immediately before a material advance is worth more than any fixed interval, because it places the verification at the moment the money moves.
Unsecured lending. No perfection to lose, so the four-month clock does not bind. Status and dissolution still matter for collectability and for standing to sue, but the interval can reasonably stretch. This is the one category where an annual or semi-annual cadence is defensible rather than merely traditional.
The common thread is that cadence should be anchored to when money is at risk rather than to when the calendar turns. Fixed intervals are simply the cheapest approximation of that, and they are a poor approximation for any product where exposure is uneven across the term.
What does this cost, and how should it be modeled?
Cadence decisions get made on intuition because the cost side feels unbounded. It is not, and the arithmetic is simple enough to do properly.
Business Monitoring re-checks a borrower's Secretary of State record on a cadence you configure, anywhere from daily to every 30 days, and each completed check costs 1 credit from the same shared pool as the rest of the API suite.[8]
That makes the cost linear and predictable. Annual monitoring of one borrower is 1 credit per year. Monthly is 12. Weekly is 52. Multiply by the number of borrowers in each tier and the yearly cost of a monitoring policy is a small, calculable figure known before committing to it.
Set that against what a missed change costs. The comparison is not credits against credits; it is credits against the exposure that a lapsed perfection or an undetected dissolution puts at risk. For any meaningful balance, the number of credits required to shorten the interval from annual to monthly is trivially small relative to a single avoided loss. The reason most portfolios monitor annually is rarely a considered cost decision. It is that nobody has run the multiplication.
The second cost is real and less often counted: analyst attention. Every alert consumes review time, and a design that fires on low-value field changes converts a cheap data cost into an expensive labor cost. That argues for severity classification and suppression rules rather than for longer intervals. Checking more often and alerting more selectively is usually better than checking rarely and alerting on everything.
What can a scheduled re-check actually tell you?
Being precise here matters, because cadence decisions built on an inflated view of what monitoring detects will be wrong regardless of the interval.
A scheduled re-check compares the current Secretary of State record against the previous one and reports what differs. Changes are classified by severity, which is what allows a status change to be routed differently from an address update rather than arriving as one undifferentiated stream.
What it does not do:
• Scope is the Secretary of State record only. No OFAC or other watchlist screening, no UCC filings, no court dockets, no professional licenses. Sanctions re-screening remains a customer-side workflow on a customer-side schedule, and no monitoring cadence here changes that.
• It cannot surface a change the state never recorded. Equity transfers in corporations and LLCs are the clearest example. No interval detects an unfiled event.
• It cannot tell you why something changed. The record shows the delta. Cause is your analysis.
• It does not resolve the four-month UCC question for you. It can surface a name change quickly enough for you to act. Whether an amendment is required, and filing it, is your workflow.
The underlying data is the same record a point-in-time lookup returns:
curl -X GET "https://apigateway.cobaltintelligence.com/v1/search?searchQuery=Acme%20Holdings%20LLC&state=TX" \
-H "x-api-key: YOUR_API_KEY"The value of a cadence is not better data. It is that something re-reads the record and remembers what it said last time.
How should a lender set the number?
Work from the constraint inward rather than from the calendar outward.
Start with the shortest clock that binds you. For secured lenders that is the four-month UCC amendment window. Choose an interval comfortably shorter than it, which in practice means monthly for anything secured. If you are unsecured, this constraint does not apply and a longer interval is genuinely defensible.
Time cycle-driven checks against reporting deadlines. Officer and ownership fields refresh on periodic reports. One check placed shortly after a borrower's report is due is worth more on those fields than several placed arbitrarily.
Tier by exposure and let behavior move borrowers between tiers. A static, uniform cadence spends the same on a $10,000 unsecured balance approaching maturity as on a $2 million secured facility with four years left. Automatic escalation after any change costs nothing and concentrates attention where something is already happening.
Then check the arithmetic. Credits per check multiplied by borrowers multiplied by frequency is a number you can compute in a spreadsheet before committing. Most lenders discover the interval they were defending on cost grounds was never actually expensive.
The honest summary: annual monitoring is the industry default and it is shorter than the clock most secured lenders are subject to by a factor of three. Fixing that is not a technology decision. It is a scheduling decision with a known price. For what happens when a change is missed, our guides to entity status transitions and the administrative dissolution reinstatement window cover the downstream consequences in detail.[9] The formation record itself is what governs a financing statement's continued effectiveness, which is why the two questions are the same question.[10]
.jpg){kind=small}
.png)