Why I turn down work I could easily take.

Every no I send costs me money that day. I'm one person. There's no partner to talk me out of it, no sales manager watching a number, no pipeline anybody but me has to explain.

I still send them.

Not because I'm booked to the ceiling, and not because I've reached some level where being picky is a flex. It's simpler than that. The projects I turn down are the ones where I can already see the ending — somebody pays me, nobody's happy, and six months later there's a site nobody touches and a working relationship neither of us wants to revisit.

I'd rather skip to the end and save us both the trip.

Here's what makes me say it.

When the website isn't the problem

Here's how it goes.

Say somebody calls because business is slow and a new site is the lever they can actually reach. Twenty minutes into the conversation, the real problem walks in: they miss half their calls. Their Google Business Profile still lists hours from two years ago. They take four days to get a quote out and the guy across town takes four hours.

A $1,500 website fixes none of that. It's a nicer front door on a house where nobody answers the door.

So I say so. Sometimes that turns into a smaller job: a one-page site instead of four pages, plus a list of Google Business Profile fixes they can make themselves. Sometimes it turns into nothing and they go fix the phone problem, which was the thing that needed fixing.

That's a conversation worth having, even when it costs me the job.

When they want the AI, not the outcome

Somebody read something, or a competitor mentioned it at a chamber lunch, and now they want AI. Ask what they want it to do and the answer is a shape, not a task. Streamline things. Be more efficient. Not fall behind.

That's not a project. That's a mood.

Every AI setup worth building starts with an actual sentence: I spend two hours every Friday retyping job notes into invoices. Every new inquiry takes me fifteen minutes to figure out what they're even asking. Something with a clock attached to it.

Without that, there's nothing to point the tool at. I could build something impressive and they'd stop opening it in three weeks, because it wasn't replacing anything they hated doing.

I don't turn this one down outright. I ask them to spend a week writing down what eats their time, and to call me back. That list makes a much better project than the one they asked for. (The free AI Time-Leak Checklist is a good way to make it.)

When price is the entire conversation

My prices are on my website on purpose, so most of this gets filtered before anyone calls. But it still happens: three emails in and every one of them is about the number.

Negotiating isn't the issue. Budgets are real, and there's a real version of this where somebody says "I've got $800, what can I get" — and the answer is a one-page site, and that's a fine project. I like those.

What I'm watching for is different. It's when nothing else about the work has come up at all. Not who they're trying to reach. Not what the site has to do. Just: can you do it cheaper.

That tells me what the whole engagement will be. Every decision becomes a cost conversation, every round of revisions is a negotiation, and the thing we build ends up shaped by what we cut instead of what it needed. I'd be taking their money to make something neither of us is proud of.

When I'd be the fourth voice in a chain

A business owner brings me in, but the decisions run through a partner who isn't on the call, a spouse with strong opinions about the color green, and a nephew who "does computer stuff."

I'm not being precious about it. Small businesses are family businesses and everybody's got a stake. But feedback that arrives secondhand, from four people who never talked to each other, isn't feedback — it's a game of telephone that I'm getting paid to lose.

What I ask for is one person who can say yes. Not one person with opinions — one person who can end a conversation. If nobody can, the project doesn't end. It just eventually stops.

Sometimes naming this out loud fixes it. They think about it and say, okay, it's me, I'll collect everyone's notes. Great. Now we've got a project.

When the sales process is already unpleasant

This is the one I trust most.

How somebody treats you before they've paid you is the best version of the relationship you're ever going to get. If the emails are curt now, if a Tuesday question needs an answer by Tuesday night, if someone's already telling me what my last quote should have been — that's the honeymoon.

It's tempting to talk yourself past that feeling. Don't. People rarely warm up after the deposit clears.

There's no diplomatic way to decline this one, so I don't try to explain it. I just say I'm not the right fit and wish them well.

What the no is protecting

Here's the part that makes this less noble than it sounds: I'm not turning work down for their benefit. I'm doing it for the clients whose work is already on my desk.

When you're one person, a bad project doesn't stay in its lane. It takes the Tuesday morning that belonged to somebody else's launch. It's the email you're still composing in your head at dinner. It's the reason a good client's small request sits for three days.

An agency can absorb that — bad accounts get handed to whoever's newest. I can't. Every hour a wrong-fit project takes comes directly out of a right-fit one.

So the honest math is that I'm not saying no to a client. I'm saying yes to the ones I already have.

And a no costs me one invoice. The wrong yes costs me a quarter.

---

If you're not sure your project is the kind I'd take, that's usually a sign it's worth a conversation rather than a form. Tell me what's actually going on and I'll tell you straight whether I'm the right person — including if the answer is no.