Judge a marketing agency by the questions they ask you

The proposal arrives before the diagnosis

Most agency proposals are written before anyone understands the business they are for. Six pages of deliverables, a monthly number at the back, a list of channels. Nothing in it could only be true of your company. Swap the logo and it would fit any contractor, dental practice or shop within fifty miles.

That is usually not dishonesty. It is a sales process built for volume. Somebody has fourteen calls booked this week and a template that closes a respectable share of them. The plan comes out generic because nothing in the process that produced it ever touched anything specific about you.

Which means the deck is the wrong thing to grade. Grade the conversation that came before it.

What a competent shop asks in the first call

A firm that expects to be judged on results has to know a few numbers before it can promise anything. Expect questions like these, in the first or second conversation:

• What is your average job value, and how far does it swing between your best work and your worst?

• Of the people who call you, what share end up as booked work?

• Who answers the phone between eight and six, and what happens to a call at seven-thirty on a Saturday?

• Which jobs do you want more of, and which ones do you take because you have to?

• How far will your crew drive before the job stops being worth the windshield time?

• What did the last person who did this for you actually do, and what came of it?

• Can we get read-only access to your ad account, your Business Profile and Search Console before we quote?

The last one matters more than it sounds. An hour inside your existing accounts tells an experienced person more than any questionnaire, and a shop that will not look before it prices is not planning to look closely afterward either.

Why those questions and not others

Booking rate multiplied by average job value is the revenue sitting behind a single lead. That product sets the most you can afford to pay for one. Without both numbers, nobody can tell you whether a ninety dollar lead is a bargain or a slow way to go broke, and every target in the proposal is decoration.

Capacity is the other one people skip. A two-truck plumbing outfit already turning work away does not need forty more calls a month. It needs better ones, or higher prices, or a third truck first. We have talked New Jersey clients out of ad budgets for exactly that reason. An agency that never asks what you can absorb is selling volume into a business that cannot take it.

Four questions to ask them back

Who does the work, and can I speak to them now? "Our team" is an answer that means a junior you will meet in month three. Ask for a name and get that person on the call.

Tell me about a client where it did not work. Everyone has one. A firm that cannot produce a specific failure and say what it learned has either not been doing this long or is not being straight with you.

What happens in the first thirty days? Good answers are unglamorous. Tracking, account access, a look at real search terms, one page rewritten. If month one is a brand workshop, you are paying for their onboarding.

What do I own if we part ways? This is the one that costs people real money.

Ownership matters more than the monthly price

Before you sign, make sure these sit in your name and not theirs: the domain registration, the website files or an export of them, the Google Ads account, the Business Profile with you as primary owner, the analytics property, and any tracking numbers routed through a call system. The agency gets added to your accounts as a manager. It should never be the other way around.

The usual version of this going wrong is not theft. It is an agency that quietly built everything inside its own umbrella account, then goes quiet, and now the phone number on two hundred directory listings belongs to a company that has stopped answering email. Untangling that takes months, and your rankings do not wait politely while you do it.

Ask for a thirty-day termination clause after the first ninety days, and one page of reporting each month showing cost per lead and, wherever it can be tracked, cost per booked job.

Where this test falls down

A good discovery call is not proof of good execution. Some genuinely skilled operators are awkward on sales calls, and a polished salesperson can ask every right question and then hand the account to somebody who has never seen your industry. These questions filter out the worst third of the market. They do not identify the best.

So protect yourself the boring way. Start smaller than they suggest. Agree in advance on the one number you will both look at in ninety days. Keep the logins in your name. Then judge the work rather than the pitch, and give it long enough to be judged fairly.

If you want to run the test on us, those are the questions we expect to be asked, and we answer them before quoting anything. Start there.