Agency Problems
Part of the Agency Problems cluster
Should I Fire My Ecommerce Agency?
Fire the agency if the problem is execution: work visibly not done, an outcome metric that has not moved in 60 to 90 days, and no explanation of what was tried. Keep looking deeper if the problem is structural: margins, audience, or list size that no agency can change — because a new agency inherits the same wall, and you pay the transition cost for nothing. The entire decision reduces to telling those two apart, and this page is the framework for doing it before the termination email gets written.
What Does Firing an Agency Actually Cost?
By the time a founder types "should I fire my agency" into a search bar, the frustration is usually months old: another report that says things are fine, another month where the business does not feel it, and a creeping sense of being managed rather than served. That frustration is valid — and it is also exactly the state in which expensive decisions get made on feeling instead of data.
Firing has two real costs that rarely make it into the decision. The first is the transition gap. On Meta, a new agency restructuring campaigns sends them back into the learning phase — typically 7 to 14 days of unpredictable performance per ad set, which is why performance often dips right after a switch regardless of whether the new agency is better. On email, flows keep running, but the transition still consumes weeks of onboarding attention.
The second is lost institutional knowledge. A good agency accumulates things that never appear in a report: which creative angles this audience has already rejected, which subject lines work on this specific list, what was tested eighteen months ago and why it was abandoned. None of that transfers automatically. A new agency re-derives it — at your expense, over 30 to 60 days — unless the outgoing agency documents a proper handover, which a fired agency rarely does with enthusiasm.
The cost of switching is not a reason to stay. It is a reason to be certain you are switching for the right problem.
Which Problems Does Firing Actually Fix?
Execution problems — work the current agency should be doing and is not — are what a better agency genuinely fixes. The pattern that identifies them: the work is visibly absent from the account, and the agency cannot name what they tried or why the metric has not responded.
Stale creative, no testing
The same three to five ads running 90+ days, frequency climbing, ROAS declining, no new variants entering the account. A competent agency fixes this on day one.
Flows unbuilt or abandoned
Core email flows missing or sitting in draft months into a retainer, or set up in the first 60 days and never edited again. A new agency with a strong build process closes this in the first 30 days.
No structure to learn from
One consolidated campaign with no testing structure, or one email blast to the full list regardless of segment. There is no mechanism to find what works and scale it — which means the retainer is buying maintenance, not improvement.
Activity reporting with no hypothesis
Months of reports with no named test, no expected result, no explanation for a flat outcome metric. The clearest sign the engagement runs on inertia.
Which Problems Will a New Agency Hit Exactly the Same Way?
Structural problems live in the business, not the agency, and they follow you to the next agency intact. If a customer costs $60 to acquire and first-purchase margin is $40, the first order loses money by definition, no matter who runs the account. If the brand has advertised for two or more years in a narrow niche, most of the people likely to buy at the current price have been reached — the channel dependency pattern of rising CPMs and softening conversion that a fresh agency inherits on day one.
The same logic applies beyond acquisition. An email list too small to generate meaningful flow revenue is constrained by traffic and capture upstream of the agency. A high return rate quietly erasing margin makes every acquisition number look worse than the ad account deserves. These show up in the agency's results, which is why the agency gets blamed — but the fix is in the system, not the vendor.
The separating question: can the agency name what they tried and why it did not work? If they cannot name what they changed, it points to execution — fixable by switching. If they can name real, competent attempts and the metric still has not moved, the wall is probably structural, and the next dollar belongs in diagnosis, not in another agency running the same channel against the same economics.
Benchmarks to Know
7–14 days
Of unstable performance per ad set when Meta campaigns re-enter the learning phase after a switch
30–60 days
For a new email agency to rebuild list knowledge the old one accumulated
60–90 days
Fair window to judge an established engagement before deciding
45+ days
Without new ad creative: the execution failure a new agency fixes on day one
$60 vs $40
CAC above first-purchase margin: the structural math no agency switch fixes
2+ years
Advertising in a narrow niche: audience saturation a new agency inherits unchanged
Related Pages
Channel Dependency Risk
When one paid channel carries the business and its economics tighten, the agency running it takes the blame. The structural version of this problem — and the lever that actually addresses it.
LTV3High Return Rate
Returns quietly erase the margin acquisition spend depends on, making agency results look worse than the account deserves. Worth ruling out before blaming the vendor.
Frequently Asked Questions
What to Do Next
If you cannot yet tell whether your situation is execution or structural, that diagnosis is the decision. Having someone map the full revenue system before the termination email goes out is the right move — it either confirms the switch or saves you from paying transition costs to hit the same wall.
Free · 20 minutes · No pitch