The One Person Your Practice Cannot Replace
A surgeon I will not name built a very good practice on the back of maybe six referring physicians. Primary care doctors he had known for 20 years, who sent him their patients because they trusted him personally, not because of anything written down. The practice threw off strong numbers and looked, from the outside, like a durable business. It was really six relationships in one man's head, and every one of them was going to retire within a few years of him.
That is concentration risk, and in healthcare it hides better than in almost any other business.
Referrals are the revenue you do not control
Plenty of specialty practices live or die on where their patients come from, and the referral book usually sits with one or two people. A rainmaker physician. A founder. Sometimes a single office manager who has kept the relationships warm for a decade. The money flows in reliably, so nobody treats it as fragile. Then the rainmaker cuts back to 3 days a week, or a hospital system buys the primary care group that fed you half your volume, and the revenue that felt permanent turns out to have been on loan.
You can see this coming if you look. Pull your referrals by source for the last 2 years. If a handful of names account for most of your new patients, treat it as a risk to the business rather than a marketing success, and put it on the same list as anything else that could take the practice down. The fix is slow and worth starting now. Widen the base. Build referral relationships at the practice level instead of the personal level, so they survive one person's retirement. Put someone other than the founder in front of the referring offices. None of it is quick, which is exactly why waiting is expensive.
The other person who can switch your revenue off
Concentration has a paperwork cousin, and it catches owners who would never miss a clinical detail. Credentialing.
A provider who is not credentialed with your payers cannot bill them, full stop. And credentialing takes time. The process runs 90 to 120 days on a good day and stretches past 6 months when something goes wrong. Every one of those days you have a physician seeing patients, drawing a salary, and generating claims that cannot be submitted. Industry estimates put the cost of a credentialing delay somewhere between $9,000 and $15,000 per provider per month, and for a surgeon or high-volume specialist the figure runs far higher. One survey found a large share of health systems losing $1,000 to $5,000 per provider per day while enrollment sits in a queue.
I have sat across from owners who hired a strong physician, celebrated the win, and then watched 3 months of that salary go out with almost nothing coming back, because nobody started the credentialing paperwork until the provider's first day. The revenue was always going to arrive. It arrived a quarter late, and that quarter was real money they never got back.
Treat both as risks, because that is what they are
The thread connecting these is dependence on a single point of failure the financials do not show. Your P&L has no line for "half our referrals come from two doctors past sixty." It does not flag the new hire who cannot bill for another 11 weeks. These stay invisible right up until the day they are the only thing that matters.
So make them visible. Once a year, ask two plain questions. If our top two referral sources disappeared tomorrow, what happens to the business? And how many days pass between hiring a provider and collecting the first dollar of their work? Put real numbers on both. Then start the credentialing paperwork the day an offer is signed, not the day the provider starts, and begin widening the referral base years before you think you need to. Nobody handles this well by avoiding the risk. You handle it by naming it early enough to do something about it.