You need to enable JavaScript to run this app.
By Travis Yule — CEO & Founder, Full Send Funding
Published 2026-02-12 · Updated 2026-09-30
There is no single business credit score — there are four, on incompatible scales. What each measures, the order that builds a file, and what changed at the SBA in 2026.
In one sentence: Business credit is a separate file built from reported experiences rather than from paying bills, in a fixed order — real entity, bureau file, accounts verified to report, paid early only where somebody is reporting — and the points above PAYDEX 80 are bought with working capital at a price you can compute.
Business credit is a separate file from your personal credit, built on a separate identifier, scored on scales that look nothing like a FICO score. It eventually unlocks supplier terms and bank-priced capital. It is built in a specific order, skipping steps wastes months, and a great deal of what is sold as "business credit building" is selling you the parts that are free.
Two things about it are almost universally misunderstood, and both cost money. A commercial file is built from reported experiences rather than from paying your bills, so a supplier who furnishes no data to a bureau contributes nothing to your file however promptly you pay it. And the behaviour that lifts the best-known business score above its on-time ceiling — paying before terms — is not free. It is bought with working capital, at a price you can compute, and there is a version of it that costs a sixth as much.
Here is the sequence that works, what each bureau measures, what a lender reads in a commercial file that you never see, and what business credit does not do.
There is no single business credit score. There are several, they use incompatible ranges, and a lender pulls the one that fits its process.
The scales are not comparable to each other or to a consumer FICO score. Ask a lender which one it is pulling.
The practical consequence: ask which score a lender is pulling. "Your business credit is a 72" means something completely different at D&B than at Equifax, and a number quoted without its scale is not information — a 72 on Equifax's 101–992 scale is not even a valid value.
A second consequence catches people later: the score you subscribe to see is not necessarily the score a counterparty buys. Monitoring products are sold to business owners and commercial reports to lenders and suppliers; they are packaged differently and do not always carry the same number. Treat your own monitoring figure as a direction of travel, not as the number on the other side of the desk.
A commercial file holds three kinds of data, and they behave completely differently.
Trade experiences. Each one is a supplier or a lender reporting what it extended, how much, and how the business paid relative to the terms it was given. This is the part you build, and it does not exist until somebody furnishes it. The file records reports, not payments.
Public records. Judgments, tax liens, bankruptcies, UCC-1 financing statements. This is the half you do not write, and the half that can change without you doing anything wrong — a funder you repaid two years ago can still have a live filing against your business because nobody filed the termination.
Firmographics. Legal name, address, phone, entity type, industry code, years in operation, size. None of it is a credit judgment, and everything downstream depends on it, because firmographics are how inbound trade data gets matched to the right file. That is why step 1 below is not paperwork.
Each score is computed from some or all of the three (PAYDEX from payment experience alone, Intelliscore Plus from all three), and a counterparty reading a real file spends most of its attention on the trade lines and the public records rather than on the figure at the top — which is why "what is my score" is a weaker question than "what is on my file".
Which score gets pulled follows from what is being decided. A supplier weighing $9,000 of net-30 terms wants one narrow thing — whether this business pays its suppliers and how late — which is a payment score. A lender deciding a five-figure facility wants the risk of serious delinquency, built from trade lines, public records and firmographics together. FICO's SBSS blends business and owner data into one number; what changed about its role at the SBA has its own section below.
Bureaus need a distinct, consistently identified entity to score. That means:
The first three are the work of the first week. The fourth quietly decides whether any of the rest works.
A bureau does not receive a payment. It receives a record from a supplier's accounts-receivable system carrying whatever name, address and phone that supplier keyed in when it opened you as a customer — which is whatever was on the document you handed them. The bureau matches that record to a file; if it matches nothing closely enough, it opens a new one.
So "Northside Mechanical LLC" at Suite 4 and "Northside Mechanical" at #4 can end up as two partial files, each holding half your history, neither strong enough to score. Every month spent paying promptly builds two half-files instead of one, and nothing in the experience tells you it is happening.
Pick the exact string — legal name, one address format, one phone number — and put it on the invoice template, the W-9, the credit applications suppliers ask you to fill in, the certificate of insurance, the bank account title and the website footer. The legal name also has to resolve to the entity behind your EIN, the most common identity defect in a funding file; the document package a funding request needs covers name control and the ways it breaks.
One further consequence of building on an identifier: a new entity is a new file. An owner who moves the business into a newly formed LLC or corporation, or re-forms it in another state, under a new EIN is starting the clock again as far as a bureau is concerned, however long the operation has been trading. If a restructuring is coming, do it before you spend a year building history that will not travel with you.
Nothing can report to a file that does not exist.
Request a D-U-N-S number from Dun & Bradstreet directly. It is free, and anyone charging you for one is charging for a form. Experian and Equifax build files from data they acquire rather than from an application, so the way onto those is step 3.
Before you start, find out what already exists. A file may be open with firmographics that are simply wrong — an old address, an industry code describing a business you no longer run — and wrong firmographics are how good trade data ends up on the wrong file. Discovering the problem in month eleven, when a supplier declines terms, is the expensive way to find out.
This is where most credit-building programs go wrong. The assumption is that paying suppliers builds credit; it builds credit only if the supplier reports, and most small suppliers do not.
Three to five genuinely reporting accounts, paid on time or early for six to twelve months, is the realistic runway to a usable profile. Fewer than three and the file is too thin to score reliably.
If building a commercial file is part of why you are borrowing, that belongs in the choice of structure rather than beside it — the comparison of an advance against a term loan treats it as one, because a reported instalment loan does something for the file that a purchase of future receivables does not.
"Do you report?" is answered by a salesperson. The file answers it, and sixty to ninety days after the first paid invoice is the practical time to look.
So: open the account, buy on it, pay it, then pull your own report and look for that supplier's trade line by name. If it is not there after two full billing cycles, that account is building nothing. You have not lost the goods — you bought things you needed — but you have lost the months you thought you were banking.
The corollary is the rule that makes this exercise sane: buy what you were going to buy anyway, on terms, from suppliers that report. A trade line created by ordering inventory you did not need costs you the whole purchase, not a credit-building fee. The cheapest trade lines are the ones already sitting in your accounts payable, so the first job is not opening accounts — it is finding out which current suppliers report, and asking the ones that do not.
A business credit card is worth having, and it is worth knowing which file it lands on. Many small-business cards report to the consumer bureaus, or report there only when the account goes negative — ask. Where a card reports to your personal file, a balance carried on it is a balance on your personal file, which is usually the thing a business card is opened to avoid; and where it reports only there, it is not building the business file you opened it for. Ask which bureaus the issuer furnishes to, then check the answer against your own file.
This is the single largest difference between business and consumer credit scoring, and almost nobody exploits it — or prices it.
D&B's PAYDEX is built on days relative to terms. A score of 80 means paying exactly on terms — the thing most owners think of as perfect. Scores above 80 are earned by paying before terms, with 100 representing consistent payment around thirty days ahead. Below 80, the score falls as the average days-beyond-terms rises. So a business that pays every invoice precisely on the due date, forever, sits at 80 and stops.
Paying early is not a habit. It is a purchase, and the currency is working capital.
Take a business carrying $54,000 a month of trade payables — $1,800 a day of supplier credit across a thirty-day month. Moving its average payment from day 30 to day 20 hands back ten days of that credit: 10 × $1,800 = $18,000 that leaves the bank account earlier and stays out of it for as long as the habit holds. Not a one-time $18,000 — a permanent $18,000 reduction in the cash the business operates with, because next month's payables come ten days earlier too, and the month after that.
That is the same lever the cash conversion cycle runs on, pulled backwards: the operator's guide to the cycle works the arithmetic of stretching supplier payments to release cash, and what is being described here is the decision to spend that same cash to buy a score. Worth it sometimes. Worth knowing the price always.
Now put the two mechanisms together. Only reported experiences can be scored, and only some of your suppliers report.
If three of your fourteen suppliers furnish data, and those three carry $9,000 a month of the $54,000 payables book — $300 a day — then paying those three ten days early costs 10 × $300 = $3,000 of cash held out, not $18,000. Same ten days, same visible behaviour in the file, one-sixth of the working capital, because five-sixths of the book was buying nothing. The eleven suppliers who do not report can keep being paid on the due date. Paying them early is invisible to every score you are building.
There is one case where the points are free, and it is the place to start: a reporting supplier that also offers a discount for paying early. The discount pays for the cash handed over sooner and the score movement comes along with it. On $9,000 a month of purchases, terms of 2/10 net 30 are worth 2% of $9,000 — $180 a month, $2,160 a year — against roughly $6,000 of cash surrendered twenty days earlier, and the cash conversion cycle article above gives the formula for what that is worth annualised. Sort the reporting suppliers by whether they discount, and work down from the ones that do.
The honest limit: we cannot tell you how many points any of this buys. No bureau publishes how many points a change in one trade line moves its score, and a file with three reporting trades behaves differently from one with thirty. What can be stated is the direction, the cost, and that paying early where nobody is reporting is a pure loss.
The reason to build a file is usually given as "so a bank will lend to us" — which is the last thing it does, not the first.
Supplier terms come first, and they are the best deal in small-business finance. A supplier extending net-30 on $9,000 a month of purchases is financing the business 30 × $300 = $9,000, permanently and revolving. Move the same supplier to net-45 and it is 45 × $300 = $13,500. There is no stated cost — a supplier who prices early payment separately is charging for the terms — and the capacity arrives months before a lender will look at the file. This is what a payment score is for.
Then bank-priced capital, last. No commercial lender approves a facility because a business credit file is strong. The file gets you read; coverage, history and a guarantee decide it. How a bank line compares with the other working-capital instruments is the question that follows, and it is not a question about scores.
Mostly two things.
The public-record half. UCC-1 financing statements filed by previous funders sit in the record and are read as evidence of a position, whether or not the position is still live. A filing survives payoff: the security interest ends under the agreement, but the public filing stays until somebody files a UCC-3 termination, and plenty of secured parties do not file one promptly. So a business that repaid a facility two years ago can still show what looks like an open blanket lien — and the filing appears on business credit reports as well as being found by every counterparty who searches the Secretary of State. What a filing actually covers, and how to get a stale one removed is a subject of its own. The point here is narrow: this is the one part of your file you can be damaged by without having missed a payment, and the search that finds it takes ten minutes.
What we file on working capital up to $2 million: nothing. On deals up to $2 million, our working-capital agreements carry no UCC-1 lien filed against the business and no confession of judgment, so such an advance from us leaves no public-record trace for the next lender to find and, unless you ask, no trade line either. Decide which way you want the trade line before signing: nothing to explain later, and nothing built either.
This is information, not legal advice; your filings and your agreements control, and a commercial attorney should read anything that matters.
Worth knowing if an SBA loan is the destination, because the rule most articles describe is no longer in force.
The SBA historically pre-screened 7(a) Small Loan applications with a FICO SBSS score and a published minimum — most recently 165. Procedural Notice 5000-875701, effective January 16, 2026,[1] and the supplemental guidance that followed it, effective March 1, 2026,[2] sunset that screen: 7(a) Small Loan applications no longer receive an SBSS score and the SBA no longer screens on one. SOP 50 10 8 replaced it with conventional commercial credit analysis, and the floor it sets is a minimum debt service coverage ratio of 1.0× (1:1) for 7(a) Small Loans — loans of $350,000 or less — and 1.15× for loans over $350,000.[3] Lenders may apply the same policies and processes they use on their non-SBA loans of the same type and size, use credit scoring models of their own, or follow the underwriting criteria used for standard 7(a) loans.[1]
The practical translation: chasing a specific SBSS number is no longer the play. Demonstrable debt service coverage is, and that is a cash flow target rather than a credit-file target.
Worth seeing what the floor means in dollars. A business with $198,000 of cash available to service debt against $176,000 of scheduled principal and interest covers its debt 1.125 times — clear of the 1.0× floor on a Small Loan of $350,000 or less, short of the 1.15× required above that. Two things follow. The denominator includes the debt service you already carry, which is why a new recurring debit appearing in your statements mid-underwriting moves the ratio the wrong way. And the floor is the SBA's, not the lender's: a bank's own credit policy commonly asks for more, and that figure lives in the policy rather than in any public document. The full comparison of a 7(a) loan against fast working capital works the timeline and the rate difference that hang off it.
Three expectations that cause real disappointment.
It does not remove the personal guarantee. Nearly all small-business credit is personally guaranteed regardless of the business file, and a strong file improves pricing, size and terms rather than eliminating the guarantee. What the guarantee actually says, and which of its forms is in front of you is worth reading closely before it is signed.
It does not build quickly. Six to twelve months of reported history is the floor, and a file with two trade lines is not a profile. Anyone promising a usable business score in thirty days is selling something.
It does not substitute for cash flow. Every commercial lender ultimately underwrites the ability to pay. A pristine file on a business with no coverage is a decline with a good-looking score attached — and the reasons applications are actually declined are read from the bank account and the public record far more often than from any score.
Some of it is useful. Most of what is charged for is either free or is history you did not earn. A D-U-N-S number is free on request, so a package whose headline deliverable is "we'll get your D-U-N-S" is charging for a form.
"Seasoned trade lines" and shelf companies sell history rather than a way to create it. A file whose trade lines all appear at once, unconnected to the payables and deposits the business actually has, reads differently from one that grew, and an underwriter holding four months of bank statements beside a commercial file is in a position to notice whether the two describe the same business.
Monitoring subscriptions are the category most likely to be worth paying for, for the least glamorous reason: you cannot run the verification loop in step 3 without seeing your own file. The stale filing described above needs no subscription: the Secretary of State search finds it. What you are buying is visibility, not a score.
Week one. Entity, EIN, business bank account, one exact name-address-phone string everywhere, a D-U-N-S number requested, and a Secretary of State search of your own business so you know what is already in the public record.
Months one to two. Open three to five accounts you have asked about reporting — and would have opened anyway.
Months three to five. Sixty to ninety days after each account's first paid invoice, pull your own file and confirm it appears as a tradeline. Replace any that have not appeared by then.
Months four to twelve. Pay the reporting accounts early and the rest on terms, and keep the string consistent on every new supplier application. This part cannot be compressed: it is history, and it accrues at one month per month.
Month twelve. Ask a supplier for a terms increase before asking a bank for anything — the supplier is the counterparty most likely to say yes, and the one most likely to tell you what it saw.
None of this produces capital this quarter. The file is a twelve-month asset, and if money is needed now, it is not an answer to that question. The mistake is not building the file — it is sequencing the file ahead of the decision actually in front of you. Waiting postpones the growth that produces the deposits, and the deposits are what a revenue-based underwriter reads anyway. Run both: the file in the background, the capital decision on its own merits, now.
While the file builds — and it does take the better part of a year — revenue-based funding runs on a different axis entirely: your deposits, not your score. Average daily balance, deposit consistency and negative days do the work a credit file would otherwise do.
Qualification is four months in business and $10,000 a month in revenue, all credit profiles considered, on a soft-pull application that leaves no mark. Decisions come within 24 business hours, and on deals up to $2 million, 90% of complete applications that meet our requirements are approved and funding follows within 24 hours of approval — one to three business days end to end.
Used deliberately, that capital funds the growth that makes the eventual bank relationship worth having, and a clean, completed funding record is itself a reference the next funder will ask about. Be clear about which tool does which job: an advance — underwritten on revenue, not the credit file — funds the gap, and the file gets built by the deliberate steps above, in parallel.
Four scores on incompatible scales, so a number quoted without its scale is not information. A commercial file holds reported trade experiences, public records and firmographics, and only the first is built by how you pay. Nothing reports to a file that does not exist, so the order is entity, EIN, bank account, one exact identifying string, a D-U-N-S number, then three to five accounts verified against your own file rather than a salesperson's answer. PAYDEX stops at 80 for paying on terms and rises only for paying early, which is bought with working capital: ten days early across a $54,000 monthly payables book holds $18,000 out permanently, while ten days early on only the $9,000 of it that reports costs $3,000. Supplier terms are the first return and a bank facility the last — and the file does not remove the personal guarantee, does not build in under six months, and does not substitute for cash flow.
The score table lists the four scales as each bureau publishes them, and the sequence is the firm's own practice advising applicants; the runway estimates — three to five reporting accounts, six to twelve months, and the sixty-to-ninety-day window for checking that an account reports — are practice, not measurement. The worked payables arithmetic is illustrative rather than a case: a $54,000 monthly payables book at $1,800 a day, of which $9,000 a month at $300 a day sits with suppliers that report, with every figure in that passage derived from those stated inputs. The coverage example — $198,000 of cash available against $176,000 of scheduled principal and interest, or 1.125× — is arithmetic on the SBA's published floors rather than a file we underwrote. The SBA change is stated from the two procedural notices cited, and the debt service coverage floors from the SBA's information notice issuing SOP 50 10 8,[3] as each read at the time of writing; SBA guidance changes, and the current standard operating procedure controls.
In order: register a real entity and get an EIN; open a dedicated business bank account; use one exact business name, address and phone everywhere; request a free D-U-N-S number from Dun & Bradstreet; then open three to five accounts that actually report to a commercial bureau and pay them early. Six to twelve months of reported history is the realistic runway to a usable profile.
Dun & Bradstreet, Experian Business and Equifax Business are the three that hold commercial files, and FICO’s SBSS is a blended score built on top of business and owner data. They use incompatible scales — PAYDEX and Intelliscore Plus run 1–100, Equifax’s Business Credit Risk Score runs 101–992, SBSS runs 0–300 — so a score means nothing without knowing which one it is.
80 means the business pays exactly on terms, which is the benchmark most suppliers and lenders look for. What surprises people is that 80 is the ceiling for on-time payment: everything above it is earned by paying before terms, with 100 representing consistent payment roughly thirty days ahead. A business that pays every invoice precisely on the due date forever sits at 80 and stops there.
Only if the vendor reports, and most small suppliers do not. This is where credit-building programs quietly fail. Ask directly before opening an account: do you report payment history, and to which bureaus. The same question applies to business credit cards, many of which report to consumer bureaus, or only report negative events.
Not for revenue-based funding, which is underwritten from bank statements rather than a credit file — deposit consistency, average daily balance and negative days do the work a score would otherwise do. Qualification starts at four months in business and $10,000 a month in revenue, all credit profiles considered, on a soft-pull application. A strong business file matters when the destination is a bank line or an SBA loan.
No. Procedural Notice 5000-875701, effective January 16, 2026, sunset the SBSS pre-screen for 7(a) Small Loans, with supplemental guidance effective March 1, 2026; those applications no longer receive an SBSS score. SOP 50 10 8 replaced the screen with conventional commercial credit analysis and sets the minimum debt service coverage ratio at 1.0× for Small Loans of $350,000 or less and 1.15× above that. The target is now demonstrable coverage rather than a specific score.
Almost never. Nearly all small-business credit is personally guaranteed regardless of the business file; a strong file improves pricing, size and terms rather than eliminating the guarantee. It is worth reading what the guarantee actually covers, since the difference between guaranteeing payment and guaranteeing conduct is substantial.
Six to twelve months of reported history at a minimum, and a file with two tradelines is not a profile. Anyone promising a usable business score in thirty days is selling something. The steps themselves — entity, EIN, D-U-N-S, consistent listings — take a week; the history has to accumulate.
Open the account, buy on it, pay it, then pull your own business credit report sixty to ninety days later and look for that supplier's trade line by name. A salesperson's answer is not evidence; the file is. If nothing has appeared after two full billing cycles, that account is building nothing, and the fix is to stop counting on it and open one that does report. This is also the reason a monitoring subscription is usually the only part of credit-building worth paying for: you cannot run the check without seeing your own file.
Yes, and the price is working capital rather than a fee. A business carrying $54,000 a month of trade payables holds $1,800 a day of supplier credit; moving its average payment from day 30 to day 20 hands back ten days of it, which is $18,000 of cash out of the account and kept out for as long as the habit holds. The cheaper version is to pay early only on the accounts that report — ten days early on $9,000 a month of reporting purchases costs $3,000, not $18,000. Paying early where nobody reports is invisible to every score you are building.
The file is the record: reported trade experiences, public records such as judgments, tax liens and UCC filings, and firmographics like legal name, address and industry code. The score is a number computed on top of it, and each bureau computes its own number from its own file, and a blended score such as FICO SBSS is computed on top of business and owner data. A counterparty reading a real file spends most of its attention on the trade lines and the public records rather than on the figure at the top. That is why a score quoted on its own, without its scale or the file underneath it, tells you very little.
You can buy history, and it does not read like history you earned. A file whose trade lines all appear at once, unconnected to the payables and deposits the business actually has, looks different from one that grew — and an underwriter holding four months of bank statements beside the commercial file is in a position to notice whether the two describe the same business. The parts of credit-building genuinely worth money are visibility into your own file and suppliers you would have bought from anyway. A D-U-N-S number is free.
In practice, yes. A commercial file is built on an identifier, so a new entity with a new EIN is a new business as far as a bureau is concerned, however long the operation has been trading. An owner who moves the operation into a newly formed LLC or corporation under a new EIN, including re-forming it in another state rather than registering the existing entity there, starts the history again. If a restructuring is coming, do it before spending a year building a file that will not travel with you.
Supplier terms, and they are the best deal available to a small business. A supplier extending net-30 on $9,000 a month of purchases is financing $9,000 permanently and revolving, at no stated interest cost; at net-45 the same purchases finance $13,500. That capacity arrives months before a lender will look at the file, and the supplier is the counterparty most likely to tell you what it saw. Ask a supplier for a terms increase before asking a bank for anything.
Travis Yule worked in business funding before starting Full Send Funding in 2021, and has more than five years in business funding in all. He leads the company from Middle Grove, New York, and writes about working capital from the underwriting side of the table — what the numbers actually have to say before a business gets funded.