Revenue Concentration Limits: When to Refuse More Work

A regional property manager calls your office on a Tuesday. They want eight more rooftop units replaced before winter, and they want a crew on site in three weeks. They're already 40% of your revenue. The work is real, the margin is fine, and your dispatcher is looking at you through the glass.

Most owners treat this as a yes or no question. It isn't. It's a marginal decision with four possible answers, and the right one depends on what this specific job does to your crews, your calendar, your cash and your ability to walk away later.

This article skips the search for a magic concentration percentage. No number works for every HVAC, plumbing, electrical, roofing or landscaping shop. What works is a decision rule you apply to each new job from your dominant customer. If you want the exposure basics first, start with our guide to customer concentration risk, then come back here for the part where the phone is ringing.

Why a personal revenue concentration limit is a decision rule, not a percentage

Search for revenue concentration limits and you'll find rules of thumb: no customer above 20%, 25%, 30%. Lenders use versions of these. Buyers use them too. They work as warning lights. They're bad as operating rules.

Two shops can both have a customer at 35% of revenue and sit in completely different positions. Shop A bills that customer net 15, has a signed multi-year service agreement, and could move its crews to residential work within two weeks. Shop B bills net 75, works on purchase orders that can be cancelled with a phone call, and has trained its whole install team around one customer's specs. Same percentage. Very different risk.

A percentage describes where you are. It says nothing about whether the next job moves you toward safety or toward a cliff. That's the question you actually face when the customer asks for more.

So set your personal limit as a rule about behavior: "I accept a job from my largest customer only if it passes these four tests." A rule like that holds up when the customer is charming, when your pipeline is thin, and when you're tired. A percentage on a slide doesn't.

Treat any benchmark you find as a trigger for a harder look, not as a green light below the line. A customer at 18% can still wreck you if they pay slowly and your payroll depends on them.

The marginal question: what does this one job do to your business?

Stop asking "is this customer too big?" Start asking "what does this one job change?"

A marginal question forces specifics. It pulls the decision out of mood and into the schedule, the bank balance and the contract. You aren't judging the relationship. You're judging a block of crew hours and a block of receivables.

Four things change when you say yes to more work from a dominant customer:

1. Your crews get committed. Technicians and installers assigned to this job aren't available for anything else. If a better-paying or more diversified job shows up next week, you can't take it.

2. Other work gets displaced or delayed. Your small residential customers, your maintenance contracts and your walk-in repairs still need service. Something gives.

3. Your cash gets tied up. You buy materials and pay labor now. You collect on the customer's timeline.

4. Your options narrow or stay open. Some jobs leave you free to scale back. Others commit you to equipment purchases, new hires or subcontractor deals you can't unwind.

Those four effects map directly to the worksheet later in this article: crew capacity, displaced work, payment timing and reversible commitments.

The answer can differ job to job, even from the same customer in the same month. A small, fast, quickly paid job from your biggest account might be an easy yes. A large, slow, hard-to-reverse job from the same account might be a decline or a reprice. Treating the customer as one lump hides that.

Accept, defer, reprice or decline: the four outcomes and what triggers each

Every job from a dominant customer ends in one of four outcomes. Name them in advance so you aren't inventing the options while the customer waits.

Accept. Take the job as quoted. Triggers: crews have real slack, no higher-value work is displaced, the customer pays on terms your cash can carry, and you could stop after this job without damage. An accept is a clean pass on all four tests.

Defer. Say yes to the work, no to the date. Triggers: the job is fine on price and terms, but your crews are booked. You offer a start date four to six weeks out, or you split the job into phases. Defer is the most underused answer. Owners treat the customer's deadline as fixed when it often has more give than they assume.

Reprice. Take the job at a different price or on different terms. Triggers: the job passes on crew capacity but fails on payment timing or margin. You ask for a deposit, shorter payment terms, a mobilization fee, or a rate that reflects the displaced work. See our guide on customer deposits for how to cover materials before work starts, and our post on service business pricing strategy for raising rates without losing margin.

Decline. Say no. Triggers: the job would push you past what your crews can deliver well, displace higher-value work you can't recover, strain cash beyond your buffer, or commit you to something you can't reverse. Declining is a business decision, not a verdict on the customer.

Pick one outcome per job and write down why. "Defer because the second crew is committed through the 14th" is a decision. "Let's see how it goes" is drift.

The job acceptance worksheet: crew capacity, displaced work, payment timing, reversible commitments

Run every sizable request from your dominant customer through four questions. Put the answers on one page. If you can't fill in a line, you aren't ready to say yes.

1. Crew capacity. How many crew-days does this job need, and which crews? What's the schedule for those people over the same window? Count truck availability, licensed lead technicians and any certified roles the job requires. If you'd need overtime or a subcontractor to hit the date, write that down with the cost. Our post on service business capacity planning covers when it makes sense to add another crew instead of stretching the ones you have.

2. Displaced work. What will you push back, drop or rush to make room? List it by customer and dollar value. Include maintenance agreements, warranty callbacks and your highest-margin residential jobs. Displaced work is the cost most owners forget, because it never appears on the new job's quote.

3. Payment timing. When do you pay for materials and labor, and when does the customer pay you? Calculate the gap in dollars and in days. A job that takes 60 days to collect on a customer that's already your largest receivable adds to a pile you may already be carrying. Check it against your latest cash flow forecast.

4. Reversible commitments. What do you have to buy, hire or promise to do this job? A rented lift for two weeks is reversible. A new full-time hire, a financed truck or an exclusivity clause is not. The more irreversible the commitment, the higher the bar for acceptance.

Score each line green, yellow or red. All green is an accept. Yellow on capacity points to defer. Yellow on payment points to reprice. Red on displaced work or commitments points to decline or a serious renegotiation.

A hypothetical walk-through (illustrative numbers only)

These numbers are invented to show the mechanics. They aren't from a real customer.

A mechanical contractor with about $12M in annual revenue does 38% of its work for one regional property management group. The group asks for 10 rooftop unit replacements in a six-week window, worth roughly $180,000.

Crew capacity. The contractor has two install crews. One is booked on a school retrofit through week five. Taking the job on the requested dates means pulling the second crew and using a subcontractor for 8 crew-days at a premium. Yellow.

Displaced work. The second crew is also due to start four residential system replacements for new customers, worth about $64,000 at higher margin than the property group's work. Those would slide by three weeks, and two of the homeowners have said they'll shop around. Red-leaning yellow.

Payment timing. The property group pays net 60. Materials for 10 units cost about $95,000 up front. With payroll, the contractor would carry roughly $130,000 for two months. The cash buffer covers it, but barely. Yellow.

Reversible commitments. No new hires, no equipment purchases. Green.

A flat yes would pass on commitments and strain everything else. A flat no throws away $180,000 of real work from the customer that keeps the lights on. So the contractor picks a mix: defer and reprice. It offers to do the first five units on the requested start date and the remaining five three weeks later, asks for a 30% deposit on materials, and moves terms to net 30 on the second phase.

The customer accepts the phasing, pushes back on the deposit, and settles on a materials deposit of 20%. The contractor keeps its residential jobs, caps its cash exposure and keeps the relationship. None of that came from a concentration percentage. It came from testing one job four ways.

How to say no or not yet to a dominant customer without losing the relationship

Owners avoid the conversation because they assume a no ends the account. It rarely does. Large customers deal with vendors who overpromise all the time, and they trust the ones who give straight answers.

A few moves that work:

Lead with what you can do. "I can start five units on the 3rd and finish the rest by the 24th" lands better than "I can't do all ten." Offer the version of yes that your crews can deliver well.

Tie the answer to their outcome. Customers care about finished work, not your schedule. Frame a defer as protecting quality: "If I pull the second crew, the first-phase units get less supervision and your tenants feel it."

Explain a reprice in terms of cost. A deposit or shorter terms is a request to cover real money, not a favor. "Materials on this job run about $95,000 and I need to buy them before the 3rd. A 20% deposit lets me hold the schedule without squeezing other customers."

Give notice early. The worst time to say not yet is the week before the start date. When the request comes in, answer within a couple of business days with your outcome and the reason.

Say no cleanly when you mean it. If the job fails the tests, decline, name the constraint and offer a future window. "I can't take this one without hurting service for the rest of my accounts. I can take the next batch in January."

Be the vendor who's predictable. A customer that depends on you for a large share of its work has its own concentration problem. A supplier that says no sometimes is a supplier whose yes means something.

Keep a decision record and review it monthly with an operating partner

A decision rule only works if you apply it the same way each time. That takes a record. For every job from your dominant customer, write down the date, the request, the four worksheet scores, the outcome you chose and the reason. Two minutes per job.

The record does three things. It stops you from rationalizing a yes that your own worksheet said was red. It shows you patterns: maybe you defer every winter request and lose the customer's trust, or maybe you accept every job in the first quarter and run out of slack by March. And it gives you something to look back on when a decision goes badly, so you can tell a bad call from bad luck.

Then review the record monthly with someone who isn't inside the work. An outside operator will ask the question your dispatcher won't: "What happened to the residential jobs you pushed?" That's the purpose of a monthly business review: turn decisions into actions and check whether they held.

This is the kind of work Finnus is built for. Finnus gives owner-run businesses an experienced operating partner, monthly board meetings and a lasting decision record, so a call like "defer and reprice the property group's job" gets written down, reviewed and followed up on. Finnus Board costs $999 a month with a three-month minimum, then month to month, and it includes monthly board meetings, Business Memory and partner answers within two business days.

If your largest customer has more hold on your business than you have on theirs, you need a second set of eyes on every big yes.

Conclusion

Your revenue concentration limit isn't a number on a slide. It's the pattern of your last twenty decisions about your biggest customer. If the pattern is yes, yes, yes, you've already set a limit, and it's whatever your crews and cash can absorb before something breaks.

Pick the next job your dominant customer sends you and run the four-test worksheet before you answer. Choose accept, defer, reprice or decline, and write down why.

If you want an outside view on that call, book the free 30-minute intro call with Finnus at finnus.ai. Bring the job and the numbers, and leave with a second opinion from an operating partner who will help you record the decision and review it next month.