Vol 1 · Issue 103 · Friday, September 4, 2026

The worst phone call in a service business isn't the client who fires you.

It's the client who says yes to something enormous, and you're standing there doing math in your head, realizing you cannot actually deliver it, and you have about four seconds to decide whether to say so out loud.

Most owners don't say so. They say yes, hang up, and spend the next ninety days in a slow motion car accident. The work goes out late. The quality slips. The team gets crispy. And the client who was going to be your best case study becomes the one you quietly stop mentioning.

That call is coming for a lot of you in the next sixty days, because Q4 is when the big ones land. And the fix has to be in place before the phone rings, not after.

The obvious answer is the wrong one

When an owner realizes they're going to be short on capacity, the reflex is to hire.

Look at what that actually costs you right now. Thirty six percent of small business owners reported job openings they could not fill in July, the highest reading in a long while, and labor quality jumped eight points to become the single most cited problem on Main Street again. Meanwhile a net twenty percent of owners are planning to add people, the most since October 2022.

Translate that out of survey language. Everybody wants to hire, the good people are already taken, and the ones who are available are expensive and getting more so. If you start a search in September for a role you need in October, you are competing against a whole lot of other owners who had the exact same panic on the exact same week, and you will either wait three months or settle for someone you wouldn't have hired in a calmer season.

Then there's the money. A hire is a fixed cost you carry through January and February, which for most service businesses are the two thinnest months of the year. You are looking at borrowing rates around eight percent for short term money if cash gets tight. You'd be adding permanent overhead to solve a temporary spike, and then paying for that decision during your slowest quarter.

There's a better move and almost nobody makes it early enough.

What a bench actually is

A bench is a small, vetted, pre-negotiated group of people who don't work for you, have already done a real piece of work for you, know how you operate, and can be activated inside of a week.

Not a list of freelancers you found on a marketplace at eleven at night. Not a folder of resumes. People you have actually worked with, at a rate you've already agreed on, who have already signed the paperwork.

The difference between a bench and a list of names is the same as the difference between a fire extinguisher and knowing that fire extinguishers exist.

Three tiers, and you want at least one person in each.

Tier one, the overflow hands. Someone who can do the core delivery work of your business at maybe eighty five percent of your standard. Not as good as your best person. Good enough that the client never notices, with a review step in front of it. This is the tier you'll use most.

Tier two, the specialist. Someone who is better than anyone on your team at one specific thing you occasionally need and can't justify hiring for. The person who does the thing you always dread quoting.

Tier three, the operator. Someone who can run project management, client communication, and scheduling when volume spikes. This is the tier owners forget entirely, and then they end up personally coordinating four subcontractors at ten at night, which defeats the entire purpose.

Two people in tier one, one or two in tier two, one in tier three. That's the whole roster. Five people, max. Bigger than that and you can't keep any of them warm.

Finding them without wasting a month

Skip the big marketplaces for tier one and tier two. The signal to noise is brutal and you'll spend two weeks reading proposals.

Better sources, roughly in order of hit rate.

Your competitors' former people. Somebody who left a shop like yours in the last eighteen months and went independent. They already know the work, the client type, and the standards. Find them the boring way: ask around, look at who has gone quiet at the firms you compete with.

Your clients' vendors. If a client uses somebody adjacent to you and speaks well of them, that person is pre-vetted by someone whose judgment you already trust.

People who pitched you and you passed on. Go back through your inbox. There's usually somebody who reached out looking for work who wasn't right as an employee but is exactly right as a subcontractor.

Referrals from your own bench. Once you have one good person, ask them who else they'd vouch for. Good subcontractors know other good subcontractors and they will not send you someone who embarrasses them.

The audition project

This is the part that makes the whole system work and it's the part everybody skips because it costs money before it makes money.

Do not put a bench person on a real client engagement first. Give them a paid audition.

Real work, real money, small stakes. Two to six hours. Something that has a right answer so you can actually judge the output. Pay them your normal rate, not a discount, because you are buying information and information is worth full price.

Then score five things.

Did they ask clarifying questions before starting, or did they just charge ahead and guess? The ones who ask are the ones who won't cost you a client.

Did they deliver by the deadline, and if not, did you find out before the deadline or after?

Was the quality within range of your standard, or would it need to be redone?

How did they take feedback? Give them one real correction and watch what happens. Defensive is a no. Curious is a yes.

Would you let them talk to a client unsupervised? Answer that honestly, because tiers one and two sometimes have to.

Three yes answers out of five and they're tier one with review. Five out of five and they're tier two or three. Fewer than three and you just paid a few hundred dollars to avoid a disaster, which is the best money you'll spend this quarter.

Sort out the money before you need them

Never negotiate rate during a crisis. You will overpay and you will resent it, and resentment is a terrible foundation for a working relationship.

Agree on a rate card now. Standard rate, rush rate for anything inside seventy two hours, and a monthly retainer option if you want first call on their time.

Then check your margin at that rate before you shake on it. If you bill a client 150 an hour and your bench person costs you 95, you have 55 an hour of gross margin to cover your review time, your project management, and your profit. If your review process eats an hour for every four hours they work, your real margin is thinner than it looks. Do that arithmetic once, in writing, while nobody is panicking.

Paperwork, and this is short but not optional. Contractor agreement with clear IP assignment so the work belongs to you and you can hand it to the client. Mutual confidentiality. A non-solicit that says they won't take your clients directly for a reasonable period, which is standard and nobody good will object to. Proof of their own insurance if they're doing anything with liability attached. Get a W-9 in September so you're not chasing it in January.

Make them useful in a day, not a week

The reason most subcontractor relationships feel like more work than doing it yourself is that the handoff is verbal, and verbal means you explain it every single time.

Build one handoff packet. One document per service you'd hand off. What good looks like, with two real examples of finished work. The steps in order. The three things that always go wrong and how to handle them. Who they contact with questions and how fast they can expect an answer.

Record yourself doing the work once and talking through it. Use Fathom on your client calls too, so a bench person coming into a project mid-stream can read a transcript and know exactly what was promised instead of getting your ten minute recap from memory.

For routing, keep it stupid. A simple flow in Make that takes a new job, checks the type against your roster, and messages the right bench person with the packet attached will save you the daily coordination tax that makes owners give up on this. And if any part of your bench touches content or social, give them scheduled access through something like Buffer so they can queue work without you handing over the keys to your accounts.

Do you tell the client?

This question comes up every single time and owners tie themselves in knots over it.

Yes. Always. But how you say it decides whether it lands as strength or as an apology.

The wrong version sounds like a confession. "So, full transparency, we're a little slammed right now, so I'm going to have to bring in an outside guy to help with part of this." You just told your client you're overextended and they're getting the B team. Every problem for the next three months gets blamed on the outside guy, including the ones that aren't his fault.

The right version sounds like capability. "For this phase I'm bringing in Dana, who does this specific piece for us. She's worked on four projects in this exact space. I'm still the one accountable for the output and I review everything before it reaches you."

Same fact. Completely different message. One says you're stretched. The other says you have a team deep enough to bring the right person to the right problem, which is what larger firms sell as a feature.

Put it in the proposal, not in a phone call halfway through the job. If your standard scope document says you use vetted specialists for certain phases, it's a normal business practice by the time it happens instead of a surprise.

When it goes wrong, and it will

At some point a bench person misses a deadline or hands you work that isn't good enough. Here's the sequence, and the order matters.

You eat it with the client. Not "my subcontractor was late." You were late. You picked the person, you own the outcome. The second you deflect to a vendor, the client learns that your name on the job is not a guarantee of anything.

Then you fix the work, on your dime if you have to, and you do it fast.

Then you have a direct conversation with the bench person. One correction, specific, no lecture. Watch how they take it, exactly like the audition.

Then you make a call. Real people have bad weeks and one miss is not a pattern. Two misses is a pattern. Three and you're the problem for keeping them on the roster. Move them off, keep it friendly, and be glad you found out on a small job instead of the big one.

Keeping the bench warm

A bench you don't touch for five months is a list of names again. The people on it get busy, take a full time job, or forget who you are.

Three habits. One small paid project per person per quarter, even if you don't strictly need it, because it keeps the relationship and the standard alive. A short check-in message every six weeks, just a real one, no agenda. And thirty days of warning whenever you can see volume coming, because a heads up is what separates you from every other client they've got.

That's maybe two hours a quarter per person. Cheap insurance.

Know your trigger number

Here's what turns this from a good idea into a system.

Calculate your team's realistic weekly delivery capacity in hours. Not theoretical hours, actual ones, with meetings and admin and human beings subtracted out.

Set your trigger at eighty percent. When committed work crosses eighty percent of capacity for two weeks running, you call the bench. Not when you're at a hundred and ten percent and everyone's underwater. Eighty.

That twenty percent buffer is what lets you onboard a bench person properly instead of throwing them at a fire. It's also what lets you say yes to the big one when it comes, and mean it.

The whole point

You are not building a bench so you can grow forever. You're building one so that the single best opportunity of your year doesn't turn into the reason your fourth quarter was a mess.

The owners who take the big December job and come out the other side with a happy client and an intact team are almost never the ones who worked harder. They're the ones who made four phone calls in September.

It's September 4. You've got time. Go make the calls.

Talk Soon,

Dan

Dan Kaufman
Founder, Dead Simple Growth and Pinnacle Masters

P.S. If you already know you're going to be short on hands this fall and you'd rather not figure this out by trial and error, reply with the word BENCH and I'll send you the audition scorecard and the handoff packet template I use, both of which you can fill in for your own business in an afternoon.

Keep Reading