Key person risk in a factor is not a headcount problem. It is the set of decisions that only one person can currently make: the credit rules nobody wrote down, the debtors who pay because of a relationship rather than a process, the exceptions granted for reasons that live in one memory. When that seat empties, the work continues and the judgment behind it does not.
Three of our own pages name this problem in one line each and then move on. The operations audit page says the staffing domain checks “where knowledge lives in one person’s head instead of a documented process.” The fractional COO page describes a book that “runs on how we’ve always done it rather than a documented process.” The fractional COO for factoring companies page says collections and onboarding run on “the way we’ve always done it.” None of them says what the documented version looks like, or what specifically breaks when a particular chair goes empty. So here it is.
Search the phrase and you get human-resources content: succession-planning templates, retention frameworks, and a great deal about documenting processes in the abstract. It is written for a company, any company. The industry’s own publication has run a piece on this, “The Key Man Risk in Factoring” in the IFA’s Commercial Factor, and it puts the risk in the right three places before turning to a technology answer. This page stays on the operational half: which seats, what is inside each one, and what writing it down actually means.
The risk is in the judgment, not the task
The useful way to sort a factoring operation is by how much of each job is procedure and how much is a call.
Pulling a UCC search, cutting a funding file, sending an invoice packet, posting a payment against a clean remittance: those are procedures. Somebody can learn them in a week from a written sequence, and if the person doing them leaves, you have a staffing gap and a training cost. That is inconvenient. It is not key man risk.
The call is different. Deciding a carrier’s authority history reads wrong even though every document checks out. Knowing which broker pays in fifty days despite thirty-day terms and is fine, and which broker just started doing the same thing and is not. Granting an advance-rate exception to a client because of something that happened in year two. None of that is on a checklist, and the person carrying it usually cannot list it if you ask them to. That is the exposure, and it concentrates in a small number of chairs.
The underwriting seat
This is the one most owners hold longest, and for good reason. It is also where an undocumented departure costs the most, because the loss is invisible until a bad deal funds.
What sits in that seat is not the criteria. Criteria are writable and we have written them: the underwriting criteria guide covers the client and debtor tests, and the broker credit scoring guide covers how a debtor limit gets set and reviewed. What sits in the seat is everything around them. The deals this shop does not do and has never written down as a rule. The limit on a particular debtor that was cut two years ago for a reason nobody recorded. The pattern that makes an underwriter slow down: three carriers from the same address, an authority reinstated last month, a broker whose payment behavior changed the week its largest customer went quiet.
Replace that person with a competent credit analyst from banking and you get someone who can read a financial statement and cannot yet read a load. We say in the staffing guide that handing off credit authority means first writing down the credit policy the owner has been running in their head. This is the other half of that sentence: the credit policy is the floor, and the notes behind each decision are what makes the floor usable by the next person.
Two things make this seat survivable. The first is a written credit policy with approval authority stated by dollar band and by exception type, so the next person knows what they may decide alone. The second is a reason recorded on every limit and every exception, at the moment it is granted, in the file rather than in an email thread. A limit with no reason attached cannot be reviewed by anyone except the person who set it, which means it never gets reviewed at all.
The collections desk
Collections looks like the most procedural function in the building. Some of it is. The part that is not is the part that walks out the door.
Our collections process guide owns the aging ladder and the escalation sequence, and we are not restating it. What that page cannot capture is the layer your collector has built on top of it: which accounts-payable contact at which broker actually answers, which portal needs a particular reference format or the payment stalls, which debtor’s disputes are real and which are a stalling pattern, and which invoice got short-paid last quarter for a reason that will repeat this quarter.
That knowledge is worth money and it is almost never written anywhere. The documented version is unglamorous. A contact record per debtor that holds the name, the direct line, the remittance route, and the last thing that worked, updated by the person who learned it. A dispute history that a second collector can read without calling the first one. An escalation note that says what was tried and what the debtor said, not just that a call was made.
There is a related failure worth separating out. When the collector leaves and the notes are thin, the usual response is to start the aging over from the top, which is how a shop ends up calling a debtor about an invoice that was resolved in a conversation nobody recorded. That is not only an efficiency loss. It costs you credibility with a debtor you will be calling again next month.
The account executive and the client relationship
This is the seat that shows up as revenue risk rather than credit risk, and it is the one most likely to be treated as unmanageable.
An account executive who has carried a portfolio for four years holds the history behind every exception on it: why this client’s reserve releases on a different rhythm, what was promised verbally during a rough quarter, which carrier calls before they run into trouble and which one goes quiet instead. A client who has only ever dealt with one person at your company does not have a relationship with your company. They have one with that person.
The written version here is thinner than people expect, and it is still worth doing. Every material commitment made to a client belongs in the client file in writing, including the ones made to keep somebody through a bad month, because a promise nobody can find is a dispute waiting for the day the person who made it is not available to confirm it. Every client should have met a second person at your shop before you need them to. And the annual or semiannual account review that reads a live client’s file against the terms it was written on is where an undocumented arrangement surfaces while it is still small.
Employment questions around this seat are real and they are not ours to answer. Whether and how you can restrict where a departing account executive goes, and what they may take with them, is a matter for employment counsel in your state, not an operations question.
The posting desk
Cash application is a procedure with one judgment call buried in it, and the judgment call is the risk.
The cash application guide covers the desk in full, so the only point here is the key man one. The payer-alias knowledge, which bank-statement name maps to which debtor in your system, is usually in one person’s head and it is the single cheapest thing in this article to write down. So is the suspense policy and the short-pay reason-code list. A posting desk where the aliases and the codes live in a table rather than in a clerk’s memory is a desk a temporary replacement can actually run.
The owner who is still approving everything
The seat most factoring owners underestimate is their own.
An owner approving every onboarding at 11 PM is a staffing finding, as the staffing guide puts it, and it is a key man finding too. If no funding above a certain size can happen without one signature, the operation has a single point of failure with a phone number. The test is simple and slightly uncomfortable: what stops if you are unreachable for two weeks? Not what gets harder. What stops.
Answering that honestly usually produces a short list, and the list is the work. It tends to include credit authority above some threshold, bank release authority, anything to do with the lender relationship, and the client conversations only the owner has ever had. Each one needs either a written delegation with a stated limit, or a named second person who is genuinely current on it, or an explicit decision that it waits. All three are acceptable. What is not acceptable is not knowing which one applies.
Who reads this from outside
Two audiences grade your documentation, and neither of them uses the phrase key man risk.
Your lender is the first. In an IFA Commercial Factor Q&A about completing a facility for a factoring company, the lender described what it looks at inside the factor’s own operation as “what type of KYC and credit policies they have and how they monitor their portfolio,” and said it has helped clients “write and set up their internal policies and provide best practices on how to adhere to their guidelines.” A written policy is not a compliance nicety to a rediscount lender. It is part of what they are underwriting, alongside the collateral. Our funding-stack guide covers the field exam that follows from it.
A buyer is the second. Our guide to what drives a factoring company’s value makes the case that value is built in the operation rather than at the closing table, and this is a concrete instance of it. A book whose credit decisions, limits, and client arrangements are legible to someone who was not in the room is a book a buyer can price. A book that requires the seller to stay for a year to explain it is a different transaction.
What documentation does not do
We are not going to tell you that writing things down prevents a departure, protects a valuation, or stops a loss. It does none of those.
What it does is change what a departure costs. The difference between a seat emptying and an operation stalling is whether the next person can reconstruct the reasoning, and that is decided months earlier by whether anyone bothered to record it. The honest version of this work is boring and continuous rather than a project with an end date, and most shops that do it well got there by writing down one decision at a time as it was made.
One caution on the fashionable answer. The industry piece cited above moves from documentation to delegating the codified knowledge to automated systems, which is a reasonable argument from where it is written and is not our position. Our published line is that AI can flag anomalies but an experienced operator still has to read which deals feel wrong. Writing the judgment down makes it teachable. It does not make it mechanical, and a factor that treats a codified credit policy as a replacement for an underwriter has swapped one concentration for another.
Where this stops being operational
FRC works on process, and there are two lines here worth naming plainly.
Employment agreements, non-compete and non-solicitation terms, and what a departing employee may take with them are legal questions for qualified employment counsel in your state. Key-person insurance is the first thing most general advice reaches for, and it is an insurance question. FRC does not give legal, tax, or insurance advice, and nothing on this page is any of the three.
Where an outside read fits
Key person risk is rarely one dramatic exposure. It is a credit policy that was never written, a debtor contact list in one inbox, a payer-alias table in one memory, and an owner who is the approval step for four different things. Each is small. Together they are why a two-week absence turns into a quarter of catching up.
Staffing is one of the six domains in FRC’s 90-Day Operational Audit, alongside collections, onboarding, risk, technology, and financial reporting, and this exposure shows up across most of them rather than in one. The output is a written assessment and a prioritized 12-month roadmap, with no long-term commitment.
FRC’s founder, Renee Williams, built a freight factoring company from the ground up over 25+ years, leading credit and underwriting operations and building the processes that let the desk run when the person who designed it is not in the room. The operators we have worked with, MinWest Financial and MyFactor among them, came in understanding finance and not the operational realities of transportation factoring. Call 800-648-5450 and Renee picks up directly.