When to Fire Your Marketing Agency (and When Not To)

You should fire your marketing agency when the relationship stops producing progress you can see and they cannot explain why, not the first month a metric dips. Most agency firings happen at the wrong time for the wrong reason. Companies cut a good agency during a normal slow patch, or they cling to a bad one out of sunk cost and switching fatigue. The skill is telling a real failure apart from a rough month, because the two look identical from the outside and demand opposite responses.

What are the real signs it is time to leave?

Leave when the agency has stopped delivering value and cannot give you a straight account of what is happening. Concrete signals, and you want more than one before you act:

• You cannot tell what you are paying for. Invoices arrive, work is vaguely happening, and you could not name what the last three months produced. This is the rotted-retainer failure I described in the marketing retainer is dead. If nobody can point to output, the fee is buying nothing.
• They have no explanation for bad results. Numbers can dip for real reasons, and a good agency will name them and adjust. An agency that shrugs, blames the market for everything, and changes nothing is not steering. It is coasting.
• The senior people vanished. You hired experts and now deal only with juniors who escalate everything and decide nothing. The expertise you bought quietly left the account.
• They stopped bringing ideas. A good agency pushes you, proposes, argues. One that only executes exactly what you ask and never initiates has become an expensive pair of hands you could replace with cheaper hands.

Two or more of these together, sustained, is a real signal. Act on it.

What looks like a reason to fire but isn't?

Plenty of things feel like grounds to fire an agency and are actually just how marketing works. Firing over these is how companies destroy their own momentum.

A single bad month is not a reason. Marketing is noisy, results fluctuate, and one weak period tells you almost nothing. If you fire on every dip, you will churn agencies forever and never let anything compound. The compounding work especially takes quarters, not weeks, to pay off, which I covered in how to measure agency ROI. Cutting a brand program at month three because revenue has not spiked means killing it right before it works.

A strategy problem is not an execution problem. If your positioning is broken or your product does not sell, the agency executing that broken plan is not the villain, and firing them changes nothing. You will hire a new agency and get the same result, because the flaw was upstream. Diagnose whether the gap is strategy or delivery first, the same split I used in fractional CMO vs agency. If it is strategy, no new agency fixes it.

And boredom is not a reason. Wanting a fresh vendor because the current one feels familiar is expensive novelty-seeking. Switching costs are real: onboarding, lost context, months before the new agency knows your account as well as the old one did.

How do you diagnose before you decide?

Before firing, have the direct conversation you have probably been avoiding. Tell them plainly what is not working and give them a defined window to fix it. A good agency responds with a real plan and visible change. A bad one gets defensive or promises vaguely and does nothing. Their response to honest pressure is itself the diagnostic. You often learn more in that one conversation than in six months of quiet dissatisfaction.

Separate the controllable from the uncontrollable while you are at it. Look at the channels the agency owns end to end. Are those improving? If the parts they fully control are working and the overall number is not, the problem may be your product, pricing, or sales, not the agency. If the parts they control are also failing and they have no answer, that is a genuine agency failure.

Write down what "working" would look like, specifically, before you judge. Vague dissatisfaction leads to firing at random. A clear standard lets you judge fairly and gives the agency a real target to hit or miss.

How do you leave well if you do?

If you decide to go, protect your own assets on the way out. Get the accounts, the creative, the data, and the documentation transferred before the relationship ends, not after, when goodwill is gone. An agency holding your ad accounts or brand files hostage is a lock-in you should have avoided when you signed.

Then apply what you learned to the next hire. Whatever went wrong, encode it into how you evaluate the replacement, using the criteria in how to evaluate a modern marketing agency. Most bad agency relationships trace back to a fuzzy hire and a fuzzier scope. Fix that at the start of the next one.

Fire for absence of progress and honesty, not for noise. Stay through slow-compounding work and normal fluctuation. And always diagnose whether the real problem is the agency or your own strategy before you swing the axe. That discipline is exactly what I want clients to bring to their relationship with Girard Media, because a clear-eyed buyer makes for a better partnership on both sides.