Eligibility is where borrowing bases are won and lost. A $20 million pool can support $13 million of availability or $11 million depending entirely on how well the tests are run. Olycor runs them as code, on every invoice, every time.
Receivables eligibility testing checks every invoice against the credit agreement's criteria: aging, cross-aging, concentration, obligor type, jurisdiction, and disputes. Each test determines whether that receivable counts as collateral. The output converts gross accounts receivable into eligible receivables, the number every borrowing base is built on.
Every credit agreement defines its own set, but the same eight families of tests appear in nearly all of them. Each one either passes an invoice, excludes it, or carves out part of its value.
Invoices past the agreement's limit, commonly 90 days from invoice date or 60 days past due, drop out of the pool entirely.
If more than a threshold share of an obligor's balance, often 25 to 50 percent, is past the cutoff, the obligor's entire balance is excluded, including the current invoices.
Any obligor above the cap, often 10 to 20 percent of eligible receivables, has the excess carved out. The invoices are fine; there are just too many of them.
Intercompany and affiliate balances are excluded. Government receivables are typically out unless properly assigned under the Assignment of Claims Act.
Obligors outside permitted countries are excluded, and invoices in currencies the agreement does not approve are excluded or haircut.
Invoices under dispute come out, at least to the extent of the disputed amount, until the dispute is resolved and the flag is cleared.
When an obligor also sells to you, the payable offsets the receivable. The netted exposure, not the gross invoice, is what counts.
Cash received but not yet applied to specific invoices reduces the pool. Until it is matched, the open amounts it will extinguish are overstated.
Most agreements sequence the tests: basic eligibility first (aging, obligor type, jurisdiction, disputes), then cross-aging, then concentration caps applied to the pool that survives. The sequence is not cosmetic. A concentration cap measured against a $20 million gross pool produces a different excess than the same cap measured against the $17.3 million that remains after aging and cross-aging.
Here is the same pool under two orderings. The largest obligor holds $4.5 million, of which $600,000 is already removed by the aging and cross-aging tests, and the concentration cap is 15 percent.
| Step | Tests first, cap last | Cap first, tests last |
|---|---|---|
| Gross receivables pool | $20,000,000 | $20,000,000 |
| Less: invoices over 90 days | ($1,800,000) | ($1,800,000) |
| Less: cross-aged obligors | ($900,000) | ($900,000) |
| Less: concentration excess | ($1,305,000) | ($1,500,000) |
| Eligible receivables | $15,995,000 | $15,800,000 |
In the first ordering, the cap is 15 percent of the surviving $17.3 million pool, and the obligor's remaining $3.9 million exceeds it by $1,305,000. In the second, the cap runs against the gross pool, the excess is $1,500,000, and $600,000 of the obligor's aged invoices are effectively deducted twice. The gap is $195,000 of eligible collateral, roughly $166,000 of availability at an 85 percent advance rate, lost to sequencing alone. Neither ordering is wrong in the abstract. The one in your credit agreement is the right one, and the software has to follow it exactly.
Every test is only as good as the fields feeding it. These are the nine that matter, and what goes wrong when each one is missing or unreliable.
| Field | Why it matters |
|---|---|
| Invoice number | The unit of testing. Without a stable identifier, exclusions cannot be traced to a specific invoice or reconciled month over month. |
| Obligor ID and group | Concentration and cross-aging run at the legal entity group level. Three spellings of the same customer must resolve to one counterparty before either test means anything. |
| Invoice date | The anchor when the aging cutoff is measured from invoice date. A wrong date silently moves invoices across the 90 day line in either direction. |
| Due date | The anchor when the cutoff is measured from due date, and the input to every past-due calculation cross-aging depends on. |
| Payment terms | Used to derive or validate due dates. Net 30 recorded where the contract says net 60 makes on-time invoices look 30 days late. |
| Open amount | The dollar amount actually tested. It must be net of applied payments, or the entire pool is overstated before a single rule runs. |
| Currency | Non-approved currencies are excluded or haircut. Mixed currency data without a currency code turns conversion into guesswork. |
| Dispute flag | Disputed amounts come out of the pool. A flag maintained in a side spreadsheet never reaches the test, so the exclusion never happens. |
| Credit memo linkage | Unlinked credit memos leave open amounts overstated and hide dilution. The memo has to reduce the specific invoice it offsets. |
The rules rarely fail. The data and the process around them do. These five patterns account for most of the eligibility errors that surface in field exams.
| Where it breaks | What it looks like | The consequence |
|---|---|---|
| Due dates derived from wrong terms | Every invoice for one customer carries net 30 when the signed contract says net 60 | Invoices show past due a month early. Eligibility is understated, and cross-aging can wipe out the obligor's entire balance for no real reason. |
| One customer spelled three ways | Acme Corp, ACME Corporation, and Acme (US) each sit comfortably under a 15 percent cap | The true 22 percent exposure never appears on the certificate. It appears in the field exam, which is the worst place to discover it. |
| Stale dispute flags | A $180,000 dispute resolved in March is still flagged in June | Eligible collateral understated for three months. Borrowing capacity the company paid for sits unused. |
| Credit memos not linked to invoices | $240,000 of memos float as unmatched negative lines at the bottom of the aging | Open amounts overstated now, and the gap surfaces later as unexplained dilution that the lender prices into reserves. |
| Rules live in one person's spreadsheet | A hidden tab still applies the 25 percent cross-aging threshold an amendment moved to 35 percent last year | The certificate is wrong in a way nobody can see, and the one person who could explain the tab left in Q1. |
The test families are standard. Almost everything inside them is negotiated. Concentration caps run anywhere from 10 to 20 percent, cross-aging triggers from 25 to 50 percent. The aging cutoff can be measured from invoice date or from due date, and the difference is not academic: 90 days from invoice roughly equals 60 days past due on net 30 terms, but on net 60 terms the same clause behaves completely differently.
Larger facilities add tiers and carve-ins. Investment grade obligors might get a 20 percent cap while unrated names sit at 10. Government receivables, excluded by default, come back in when properly assigned. Foreign obligors come back in when covered by credit insurance or backed by a letter of credit. Two borrowers with identical receivables can report meaningfully different eligible pools, and both can be right, because different agreements govern them.
Olycor runs eligibility as a deterministic rule engine configured to each facility's credit agreement: your thresholds, your cutoff convention, your test sequence. Every excluded invoice carries the rule version that excluded it and a link to the source file and row it came from. Anything ambiguous in the data, a new obligor name, a suspicious due date, becomes an exception for a person to resolve rather than a silent assumption. And because the rules run as code, you can test a what-if before submitting: what the pool looks like if the disputed $180,000 resolves, or if the largest obligor pays down $1 million. AI helps interpret messy inputs on the way in; it never touches a calculated number.
Olycor is borrower software, but its output is built to be checked, not taken on faith. A credit officer or field examiner can open any invoice and see its pass or fail result for every test, with the specific rule and threshold that fired: this invoice failed cross-aging because 41 percent of the obligor's balance exceeded the 35 percent trigger. Runs are reproducible, so the same inputs under the same rule version return the same eligible total on demand. And every eligible receivables figure ties back to the AR aging it came from, line by line, which turns reconciliation from a sampling exercise into a lookup. Olycor does not replace lender review; it makes the review faster and harder to argue with.
Or browse the full receivables finance glossary and the eligibility rule library.
Last updated July 9, 2026. This page explains general market practice; eligibility criteria, thresholds, and test sequencing vary by credit agreement, and your agreement governs. Olycor does not provide legal, tax, accounting, or credit advice.
Olycor is onboarding borrowers who want eligibility to run as code instead of living in one analyst's spreadsheet.