How to Say No to Work That Does Not Fit

Why this matters

The work that eats an owner's week is rarely the work they won. It is the work they chased. The site visit for a job three towns out, the quote for a scope nobody in the shop is licensed to do, the third revision for a customer who was always going to pick the cheapest number. Each one felt like it might turn into something, and collectively they are the hours that were supposed to go into pricing, hiring, and getting home.

Saying no is usually taught as a wording problem, and a sibling article handles the wording well. This is the step before that: deciding, on defined criteria, at intake, whether the inquiry is worth an hour of anybody's time. Get that right and the wording matters much less, because you are declining early, when nobody has invested anything and nobody is offended.

The screen runs at intake, not at the quote

Almost all the waste happens between the phone call and the quote: the drive, the walkthrough, the write-up, the follow-up. So the screen has to run before any of that, on the phone, by whoever answers, using questions they can ask without your judgment.

That means the criteria have to be written down and answerable by facts. "Does this feel like a good customer" is not a screen. "Is the address inside the service area, is this work we are licensed for, and can we get to the equipment" is a screen, and a part-time office person can run it on the first call.

If your screen requires you personally to apply it, you have not built a screen, you have built one more thing that queues on you.

The three hard gates

Any one of these fails and the answer is no. They are not scored, weighted, or traded off, and they do not flex when the schedule is thin.

Gate 1: competence and licensing. Is this work your shop is licensed, trained, and equipped to do? If the inquiry involves gas appliance work, service-entrance or utility-side electrical, confined-space entry, or anything else your license and your training do not cover, decline and refer to a licensed trade. Do not attempt it with a helper and a video, and do not agree to "take a look" at it, because a look becomes a diagnosis and a diagnosis becomes an implied competence. The referral is the correct professional answer and it protects the customer, your license, and your insurance position.

Gate 2: site conditions you cannot control. Can you get to the equipment, isolate it, and make the work area safe? A customer who will not permit a shutoff on a shared system, will not clear an access path, will not confine an aggressive dog, or insists the work happen around occupants in a space that must be cleared is describing a job you cannot execute correctly. Decline before the visit rather than discovering it on arrival with a tech standing in the driveway.

Gate 3: geography. Is the address inside your defined service area? Define it once, as a drive time rather than a mileage radius, because drive time is what your day actually pays. Write down the number, and note that "we sometimes go further for the right job" is not a gate, it is the absence of one.

The four soft factors

Everything past the hard gates gets scored. One point each, and any inquiry accumulating 2 or more points is declined, unless it takes a documented exception under the rule further down.

1. Below your minimum viable ticket. A job whose expected size does not cover the trip plus the setup plus a fair share of the overhead it consumes. Express the minimum in your own terms - a number of billable hours, a job type, a category in the price book - and write it down.
2. Priced and rejected before. This customer, or this exact scope, has already been quoted by you and turned down on price. A second quote at the same price wastes both of you; a second quote at a lower price teaches the market what your first number was worth.
3. Comparison shopping with no history and no timeline. Not simply getting multiple quotes, which is reasonable. The specific pattern of no prior relationship, no decision date, and an opening question about price before scope.
4. Schedule shape you cannot meet without breaking a commitment. A window you can only hit by moving an already-booked customer. That is not a scheduling challenge, it is a trade of a confirmed customer for an unconfirmed one.

Two points, not one, because any single factor has enough legitimate exceptions that declining on one alone will cost you real work. Two independent factors on the same inquiry is a pattern.

Set your decline speed and hold it

Decline within one business day of having enough information to decide. State it as a standard and hold your intake to it.

A slow no is worse than a fast no in three separate ways. It costs the customer time they could have spent hiring somebody else, which is the part that damages the relationship and the referral. It costs your intake person the mental overhead of tracking a live inquiry that is not live. And it makes the eventual decline read as a rejection after consideration rather than a scope mismatch identified promptly, which is exactly the reading that makes people take it personally.

The wording of the decline is a separate skill and a sibling article covers it. The one thing worth stating here: the reason you give should be the gate that actually failed. Inventing a softer reason gets you a customer who calls back next month having fixed the wrong thing.

Back-test before you run it live

Do not deploy a screen you have not tested against your own history. A screen that is too tight is far more expensive than no screen at all, and you will not find out for a quarter.

Run the screen retrospectively against the last 20 jobs you actually won. If it would have declined 2 or more of those 20, which is 10% or more, the screen is too tight. Find the factor that caught them, loosen that factor specifically, and re-run the back-test. Below 2 of 20, deploy it.

Note the unit: won jobs, not inquiries. Testing against inquiries tells you how much the screen declines, which you already know it does. Testing against wins tells you what it would have cost you, which is the only question that matters.

One month of inquiries, worked through

A shop taking about 40 inquiries a month. Before the screen, the owner or his estimator quoted 36 of the 40 - the 4 that never got quoted were people who stopped answering the phone - and won 11 of those 36 quotes. That is a 30.6% win rate on quotes issued. Each quote cost about 1.6 hours across the drive, the walkthrough, the write-up and one follow-up, so 36 quotes was about 57.6 hours in the month.

He wrote the gates and the factors, then back-tested against his last 20 wins. The screen would have declined 1 of the 20, a job that scored on both the minimum-ticket factor and the schedule-shape factor and which he had won and, on reflection, had not enjoyed. That is 1 of 20, or 5%, which is below the 2-of-20 threshold, so the screen was cleared to run live.

The following month, on 40 inquiries:

The 6 hard-gate failures were 2 outside the drive-time boundary, 2 involving work the shop is not licensed for, and 2 where the customer would not permit the isolation the job required.

Of the 29 quoted, he won 11 - the same count as the previous month. Win rate on quotes issued went from 11 of 36, or 30.6%, to 11 of 29, or 37.9%. Quoting effort went from about 57.6 hours to 29 quotes at the same 1.6 hours each, or about 46.4 hours, so about 11.2 fewer hours spent quoting in the month.

Be careful what that 11.2 hours is. It is unbilled effort you stopped absorbing, not revenue and not billable time. Nothing about it is a return; it is 11.2 hours of your intake and estimating capacity that went back on the shelf, and whether it turns into anything depends entirely on what you put in it. An owner who spends the recovered hours doing more of the same low-fit chasing has gained nothing measurable.

Also be careful with the win-rate figure. It rose because the denominator fell, not because anything about the shop's selling improved. That is the intended effect and it is a real improvement in where effort goes, but calling it "we got better at closing" would be a misreading of your own number.

Reading the results honestly over a longer window

One month proves nothing. Three things to watch across a quarter:

Won-job count, not win rate. If the count falls while the rate rises, the screen is declining work you were winning. That is the failure this whole procedure is designed to avoid, and the win rate will happily hide it, since declining everything produces a perfect rate on zero jobs.

Which gate fires most. A gate firing on a large share of inquiries is telling you something about your marketing, not about your customers. If geography is your biggest decline reason, you are advertising outside your service area. If licensing is, you are being found for work you do not do. Both are cheaper to fix upstream than to decline forty times a month.

Referral flow from declined inquiries. A well-run decline that includes a real referral tends to come back. If none of your declines ever return or refer anyone, look at how the decline is being delivered, because the screen is doing its job and the conversation is not.

What flexes and what does not

In a slow stretch, the soft factors can be relaxed - raise the decline threshold from 2 points to 3, or suspend the minimum-ticket factor for a defined period. The hard gates never flex. A gate that moves when the schedule gets thin is not a gate, and the work you take on the far side of it is exactly the work that produces the callback, the license exposure, or the injury.

The documented exception. Occasionally you should take a job the screen declines: a builder relationship you are trying to open, a job type you want the shop to learn, a long-standing customer's odd request. Take it deliberately, write down the reason and the end date, and count the hours against it. The difference between a strategic exception and simply ignoring your own screen is that the exception is written down, bounded, and reviewed. Unwritten exceptions accrete until the screen is decoration.

A gate you fail every week is a scope decision, not an intake problem. If licensing catches a meaningful share of inquiries every month, the honest question is whether to add that capability, not how to keep declining it. The screen surfaces that question; it does not answer it.

References

• See related: Saying No Without Burning the Bridge
• See related: What an Owner Should Stop Doing First
• See related: How to Design an Owner's Week That Survives Contact
• See related: The Cushion That Lets You Say No
• Trade-standard practice for lead qualification and intake screening in field service