Almost every business owner who calls us has already fired one marketing agency, and most of them waited too long to do it. Not because the signs were hidden. Because they could not tell the difference between a rough month and a real pattern, and nobody wants to blow up a relationship over what might just be bad luck. This is the honest version: what actually separates a normal dip from a reason to walk, what a real report looks like versus one built to flatter the agency, and how to make the switch without losing three months of momentum in the process.
The short answer.
Fire your agency when a bad pattern repeats with no correction plan, not when a single month underperforms. That distinction sounds obvious written down. It gets a lot blurrier at 9pm when you are staring at a dashboard that looks worse than last month and wondering if you are being paranoid or if you finally have proof.
Here is the filter we tell every prospective client to run before they call us: is this a signal, or is this noise that feels like a signal because you are the one paying for it?
| Feels Like A Red Flag | What It Actually Means |
|---|---|
| Results dipped for two weeks | Normal. Platform algorithm updates and seasonality cause real dips even in accounts run well. Judge trends over 60 to 90 days, not two weeks. |
| They didn't reply within the hour | Not ideal, but not fatal if a real answer and a plan follow within a day. |
| They pushed back on your idea | Good agencies disagree with clients sometimes. That is expertise doing its job, not disrespect. |
| Reports only show impressions and reach | Actual red flag. Reports should tie to leads, cost per lead, and booked revenue. |
| Same excuse three months running | Actual red flag. A pattern with no correction is a decision, not an accident. |
One slow month is data. Three slow months with the same excuse and no adjustment is a verdict. The agency has already told you what it is going to keep doing.
Most owners who call us have been sitting on that verdict for longer than they realized. They knew something was off around month two. By month four they had built a private list of grievances in their head but never said any of it out loud, because confronting an agency feels like a bigger disruption than quietly tolerating flat numbers. It rarely is. The businesses that switch early lose a few weeks of transition. The ones that wait a year lose a year of growth they can never get back, plus the transition anyway.
The reporting tells on them.
You can learn almost everything you need to know about an agency from the report they hand you before they say a single word. Reporting is the one document an agency fully controls, top to bottom, from what gets measured to what gets left out. If a report is vague, it is vague on purpose or out of laziness, and neither one is a reason for confidence.
A report built to inform you leads with the numbers that matter to your bank account: leads generated, cost per lead, and how many of those leads turned into actual booked work. A report built to flatter the agency leads with the numbers that make any campaign look busy: impressions, reach, engagement, followers gained. None of those numbers pay your bills. They are the easiest metrics on the internet to inflate and the easiest to hide behind when the real numbers are flat.
- The report changes shape every month. A consistent format that gets worse is honest. A format that keeps shifting to whichever metric looks good this month is not.
- You have to ask for the numbers that matter. If cost per lead and booked revenue are not on the report by default, and you have to request them, the agency is not building the report for you. They are building it to survive the call.
- Every dip gets blamed on something external. The algorithm, the season, the economy, your industry being "just tough right now." External factors are real and worth naming once. When they are the answer every single month, the agency has stopped diagnosing and started narrating.
- There is no comparison to a plan. A number without a target next to it is just a number. Good reporting shows what was projected against what happened, and explains the gap either way.
- The report arrives late, or not at all without a reminder. A monthly report that shows up on the fifteenth instead of the third is a small thing until you notice it is the same small thing every month. Reporting cadence is usually the first discipline to slip when an account has stopped being a priority internally.
None of this requires you to become a marketing expert. It requires you to ask one question every month: "Did this turn into leads, and did those leads turn into money?" Watch how quickly and how directly that question gets answered. The speed of the answer tells you almost as much as the answer itself.
If your current reports genuinely do not include cost per lead or booked revenue, do not assume the worst before you ask for it directly. Send one email: "Starting next month, I want cost per lead and how many of those leads booked, alongside whatever else you're already sending." A good agency adds it without friction, usually within a cycle. An agency that stalls, gets vague, or tries to explain why that number is hard to track is showing you something worth paying attention to. Cost per lead and bookings are not hard to track. Every platform and every CRM on the market can produce them.
Free Marketing Audit
Want a second opinion on your current numbers?
Send us your last three months of reports. We'll tell you straight whether the results are actually good, actually bad, or somewhere honest in between. No pitch required.
Get My Free Audit →The relationship signs that matter more than the metrics.
Some of the clearest warning signs never show up in a report at all. They show up in how the relationship actually feels to be inside of, month over month.
Your account keeps changing hands. A new account manager every quarter means every new person spends their first month relearning your business instead of improving it. If you have lost count of how many people have "owned" your account this year, the agency has a staffing problem it is quietly passing on to you.
Strategy stopped showing up. Early on, a good agency brings you ideas you did not ask for: a new offer to test, a channel worth trying, a competitor move worth reacting to. When that stops and every call becomes a status update instead of a working session, you are paying retainer prices for maintenance mode.
The moment an agency stops bringing you ideas, they have quietly downgraded you from a client they are trying to grow into an account they are trying to keep.
They still push channels that cannot be measured. If your agency's answer to slowing growth is a direct mail campaign, a radio spot, or a run of print ads, be honest about what that suggests. Direct mail, radio, and print are the old playbook. They are untrackable, and in 2026 there is no version of "we think it worked" that should satisfy you when Meta and Google can tell you exactly what a dollar produced. An agency reaching for untrackable channels is an agency that does not want to be measured, whether they realize it or not.
Conversations get defensive instead of accountable. There is a real difference between an agency that says "here is what went wrong and here is the fix" and one that spends the call explaining why the numbers do not actually mean what they look like they mean. The first is a partner. The second is managing you, not your marketing.
You are the one bringing them the ideas. If you are the person suggesting new promotions, new pages, or new offers, and the agency's job has quietly become executing your homework instead of generating their own, ask yourself what you are actually paying the retainer for. Execution has value. It should not be priced like strategy.
Scope creeps but the invoice does not shrink to match. Services get quietly dropped, a dedicated copywriter becomes a shared one, a weekly call becomes a monthly one, and the retainer stays exactly where it was. Watch what you are actually receiving today against what was promised in the original proposal, not just what the invoice says.
Any one of these on its own is worth a conversation, not necessarily an exit. Two or three of them stacked together, sustained over a full quarter, is the pattern this whole article is about.
What a good agency actually does.
It helps to know what the alternative looks like, because "not this" is not a strategy. Here is the standard we hold ourselves to, and the one we think any agency worth paying should meet.
Reporting ties straight to revenue. Every client gets numbers on leads, cost per lead, and what those leads turned into, not a highlight reel of engagement. If a number is soft that month, it says so, with a reason and a next step attached.
Speed matches the channel. Paid direct response on Meta and Google should show you real signal inside 24 to 48 hours of launch, not a vague promise of results "somewhere down the road." SEO and content compound over months, and a good agency tells you that difference up front instead of letting you assume every channel moves at the same speed.
Client history is honest, not inflated. We run local content and SEO for Network Automotive, a multi-location auto repair group in Arizona, and we run Google Search, Local Services Ads, and content for Network Collision, a Gilbert-based collision shop. Across our client base we have published more than 230 articles. We do not need to round those numbers up, and neither should any agency you are evaluating. If a pitch leans on stats that feel suspiciously round or impossible to verify, ask for the source.
The communication cadence is set and kept, not improvised. A good agency tells you upfront exactly when you will hear from them: a monthly report on a fixed date, a standing call, and a direct line for anything urgent in between. You should never have to wonder whether silence means nothing is happening or everything is on fire. If a proposal does not name a specific cadence, ask for one before you sign, not after the first quiet month makes you nervous.
Timelines get set honestly before the contract is signed, not renegotiated after. This is where a lot of agencies quietly set clients up to be disappointed. A shop expecting Meta or Google leads within the first week and a shop expecting six months of patient SEO growth are dealing with two entirely different products, and a good agency draws that line clearly on day one instead of letting every channel sound equally instant on the sales call.
A good agency wants to be measured. If the reporting, the timelines, and the client history all hold up to a direct question, that is the strongest signal you have found the right one.
If you run an auto repair shop specifically and want the fuller picture of what a real 90-day marketing engine looks like for that industry, our auto repair shop marketing guide breaks down the exact channels and sequencing we use with shops like Network Automotive and Network Collision.
How to make the switch without losing momentum.
Once you have decided, the goal shifts from "was this the right call" to "how do I make this transition boring." A clean exit protects the leads still in your pipeline and keeps your next agency from starting three steps behind. Most of the anxiety around firing an agency is not really about the agency. It is about the gap in the calendar between "old agency stops" and "new agency starts producing." Close that gap on paper before you send a single message, and the emotional weight of the decision drops considerably.
- Confirm you own everything first. Ad accounts, domain registrar, Google Business Profile, analytics, CRM integrations, and every creative file should sit under access you control, not the agency's. If you are not sure, check today, before you say a word to them.
- Read the notice clause. Most retainer contracts require 30 days written notice. Know the number before you plan your timeline so you are not caught paying for a month of work you have already mentally ended.
- Line up the next agency before you give notice. A gap in publishing or ad spend does more damage than a mediocre agency does. Have the next team ready to pick up the accounts the same week the old one exits.
- Keep the notice short and professional. You do not owe a long explanation, and a defensive back-and-forth rarely changes anything. State the end date, request the account handoff, and move on.
- Export everything before access closes. Historical reporting, creative assets, and past campaign data are worth having even after the relationship ends. Ask for a full export in the notice, not after.
The switch itself is rarely the hard part. The hard part is admitting three months ago what the reports were already telling you.
Momentum is the thing most owners underestimate losing. A shop or clinic that goes even three weeks without active ads or fresh content usually feels it in the calendar a month later. Plan the handoff like a relay, not a stop and restart, and the switch barely registers on your revenue.
Free Marketing Audit
Ready to talk to someone who will actually show you the math?
We'll review your current results, tell you honestly if a switch is warranted, and hand you a 90-day roadmap either way. No pressure, and no contract required to find out.
Get My Free Audit & Roadmap →Frequently asked
questions.
How long should I give a new marketing agency before judging results?
Give SEO and content six months minimum, since organic growth compounds slowly. Give paid ads on Meta or Google 60 to 90 days to exit the learning phase before making a verdict. What you should not tolerate at any point is silence, excuses without a plan, or reporting that will not tie back to leads.
Is it normal for marketing results to dip sometimes?
Yes. Platform algorithm updates, seasonality, and market shifts cause real dips even in well-run accounts. The question is never whether a dip happened. It is whether your agency can explain why and show you the correction plan within days, not months.
What should a marketing report actually show me?
Leads, cost per lead, and how many of those leads turned into booked revenue. Impressions, reach, and engagement can be included as context, but if they are the headline numbers instead of a footnote, the report is built to flatter the agency, not inform you.
How do I fire my marketing agency without losing my momentum?
Confirm you own your ad accounts, domain, analytics, and creative files before you give notice. Check your contract for the required notice period. Line up the next agency so there is no dead gap in spend or publishing, then send a short, professional notice.
Should I fire an agency after one bad month?
No. One bad month is data, not a verdict. Fire an agency when a bad pattern repeats for three months or more with no correction plan, when reporting will not tie to revenue, or when your account has changed hands more times than you can count.
