The Key Man Risk in Factoring

Written by: Matt Rybak, COO, Tank Payments

As a vendor, I occasionally remember to shut up and listen. And when I do, what factors tell me is interesting. If you spend enough time doing this, it's tempting to draw through-lines between conversations, linking thoughts to other thoughts, even when they didn't seem the same at first.

Here's what I mean. In one conversation, a founder tells me they're worried about stepping away. Not from the business—from the details. They built a portfolio worth hundreds of millions on personal relationships and an intuition for risk that took twenty years to develop. How do you hand that off? In another conversation, a growing factor tells me they can't maintain service quality past a certain headcount. The new hires are competent, but they don't have the instinct yet. The clients can tell.

These sound like different problems. They aren't. Both are the same thing: your business depends on the brilliance of specific people, and brilliance doesn't scale by default.

I try to learn as much as I can about the history of this industry. I was fortunate to have lunch with my friend Glen Shu, who assures me he was around when banks de-bundled factoring from their portfolio of services. From that de-bundling, the most successful factors were the ones who led, what might be jokingly referred to, as cults of personality. They were effective, trustworthy, and so attuned to their clients that the relationships felt more like partnerships than transactions. In many cases these businesses were small enough and profitable enough that one talented operator could preside over a portfolio and still maintain a personal connection to every client as growth skyrocketed.

Factoring is a uniquely personal form of finance. That's always been a strength. But it becomes a vulnerability when the person is the process.

My perspective here is that of a software vendor, someone who is excited about the prospect of helping factors achieve scale through efficiency. But I haven't operated a factoring company, and odds are I'll never be on the front lines. I hope that my colleagues in factoring will reach out after reading this to help me understand risks and solutions one click deeper.

Where the Risk Actually Lives

According to SCORE, a research entity associated with the Small Business Administration, 47% of family business owners expecting to retire within five years haven't named a successor yet. Furthermore, SCORE estimates that only 30% of family businesses survive into a second generation. These numbers aren't specific to factoring, but factoring is especially exposed.

You can see it in the recent consolidation wave. When firms like Outgo or Bay View Funding change hands, the acquiring company isn't just buying a portfolio. They're buying relationships, judgment, and institutional memory that may or may not survive the transition. "The businesses most active in M&A are the ones that have identified ways of scaling good business practices to acquired portfolios and newly onboarded employees," says Johnny Bond of Performance Trust Capital Partners, an investment bank active in the factoring space. “The best buyers nail the integration.”

Most factors already understand this at the leadership level. What's less obvious is how deep the key-man risk goes operationally.

Cash application and collections.

I speak with factors whose leadership team is shocked to discover that applying cash is basically the domain of one subject matter expert. That person looks at a remittance with no invoice numbers, a misspelled client name, and a payment that doesn't match anything in the system, and they just know. "We have an Excel spreadsheet hundreds of lines long that we refer to when a mysterious payment hits our account and we can't tell which debtor remitted it," says Adrienne Croteau of OperFi, a very "key" person on the receivables side of the business. This is one of the more straightforward parts of the business to document. Even then, when that person is out, cash posting can take three times as long. When they leave, the knowledge leaves with them.

Underwriting judgment.

The person who decides advance rates, debtor approvals, and concentration exceptions has seen a thousand deals. They know which brokers always pay on time and which ones go quiet on day 35. That intuition is worth millions in avoided losses, and in most cases none of it is written down.

Client relationships.

Your account executive knows that a particular client's cash flow dips every March because their biggest debtor does annual budget resets. They know not to call on Fridays because the owner is on the road. They know that the client considered leaving last year and what brought them back. When that AE moves on, the client feels it before you do.

This is not representative of every factoring company, but I’d be hard-pressed to think that these categories of risk don’t resonate for most.

Digitize First, Then Delegate

Boston Consulting Group recently coined the term "Enterprise as Code" to describe what happens when businesses explicitly capture their implicit operating model. Their argument: once you've written down how the business actually runs, both people and systems can understand, test, and improve it.

That framing matters because it names the two-step sequence that actually mitigates key-man risk. Step one is digitization: getting the knowledge out of people's heads and into structured, explicit form. Step two is delegation: handing that codified knowledge to something that can act on it. A new hire, a new department, or increasingly, an AI agent that can execute the task the way your best person would have.

Neither step works alone. Documentation can collect dust in a binder on a shelf. AI, without institutional context, is a powerful tool with no idea how your business actually works. The combination is what gets businesses closer to protecting against key man risk. Your private, factor-specific, best practices can be executed by a system that actually does the work.

I don't know exactly how far AI will go in reshaping this industry. What I can tell you from Tank's efforts introducing AI into factoring businesses is that the factors who prepare for this will likely have options that the unprepared ones won't. And the preparation is the same whether you're training a new hire, onboarding an acquired team, or giving an AI agent the context it needs to apply cash, flag a collections risk, or route a payment.

Here's what that preparation looks like, based on what I've heard from factors.

Map your critical workflows end to end.

Not the ones in your employee handbook. The real ones. How does cash actually get applied? What does your best underwriter check before approving a new debtor, and in what order? What triggers a collections escalation from an email to a phone call? Most of this has never been written down, because the people doing the work have been doing it long enough that the steps feel automatic. The exercise alone is valuable. You'll find inconsistencies between what people think the process is and what actually happens.

Separate the mechanical from the judgment.

Some steps are rule-following: check the aging report, verify the NOA is on file, confirm the debtor's credit limit. These are transferable today, to a new hire or to software. An AI agent can check an aging report. It can verify an NOA. It can flag a credit limit breach. Other steps are judgment: this debtor's payment pattern changed, should I be concerned? This client's volume dropped 40% in a month, is it seasonal or a red flag? The judgment steps are where your key-man risk concentrates. But here's what's changed: if you've captured the why behind those judgment calls, not just the what, an AI agent can start making the same distinctions your best person makes. Not perfectly, not on day one. But well enough to handle the routine 70% so your experts can focus on the 30% that genuinely requires human intuition.

Treat this as infrastructure, not a one-time project.

The reason most process documentation fails is that it's treated as a compliance exercise. Write the manual, shelve it, forget it exists. Instead, make it something that evolves every time the process changes. This is the foundation that makes delegation possible. And in a world where AI agents are getting materially better every quarter, the factors who have their knowledge structured and current will be the ones who can take advantage of each new capability as it arrives.

The Point

AI is like a new employee, not a new software. That's something you've heard me say if I've ever been lucky enough to catch you for a demo. But a new employee is useless on day one without training materials, institutional context, and someone showing them how things are done here. The same is true for AI. The companies seeing real results are the ones that did the digitization work first, and are now watching AI agents take on tasks that used to require their most experienced people. We reach for this with two hands at Tank, but we're like everyone else. We've found true benefits from what we've tried so far, and we aren't sure where the benefits stop. So we're acting on a little bit of faith.

What I'm certain about is this: the factors who started as cults of personality don't have to end that way. The personality can outlast the person, if you do the work to make it transferable. Write it down. Be specific. Update it. And then let something act on it. Whether your next key employee is a person or a machine, they'll need the same thing: your hard-won knowledge, still private, but made legible.

Your best people's brilliance deserves to outlast their tenure.

About the Author

Matt Rybak is COO of Tank Payments, a money-movement infrastructure company serving the freight finance industry.

The views expressed in the Commercial Factor website are those of the authors and do not necessarily represent the views of, and should not be attributed to, the International Factoring Association