Budget one month of double fees for the overlap, plus four to eight weeks of reduced momentum
while a new team diagnoses and ad learning recalibrates. Against that, the cost of staying
compounds. If cost of customer acquisition has drifted for two quarters, staying is almost
always the more expensive choice.
The short version
- One-time cost: two weeks of double fees plus four to eight weeks of softer performance.
- Ongoing cost of staying: compounding, and usually larger.
- Quantify it with your own acquisition cost trend rather than by feel.
- Payback usually lands in month three to four when the switch was warranted.
- Do not switch on one bad quarter. Switch on a bad diagnosis.
Building the actual number
I run Flapen with 50 operators managing about 70 brands, and clients arrive from
other agencies regularly. This is the arithmetic I would run before deciding.
| Cost line | Typical size | Notes |
|---|---|---|
| Overlap fees | Two weeks of both fees | Worth paying. Buys knowledge transfer |
| Your time | 10 to 20 hours | Collection, selection, onboarding |
| Reduced performance | 4 to 8 weeks, partial | Ad learning and diagnosis period |
| Rebuilt creative | Sometimes zero | Only if you hold flattened exports |
| Rebuilt campaign learning | Sometimes significant | If campaigns lived in their tooling |
| One-time total | About one to two months of fee, plus a soft quarter |
The two variable lines are creative and campaign learning, and both depend on decisions made
when you hired the outgoing agency rather than on anything happening now. If you own your
Seller Central account, hold source files, and campaigns were built in your own account, both
lines are close to zero.
The cost of staying, which nobody quantifies
Take your cost of customer acquisition twelve months ago and today, at comparable volume.
If it has risen 20 percent and nobody has explained why, that gap is your monthly cost of
staying, and it compounds. On a brand spending $10,000 a month on acquisition, a 20 percent
drift is $2,000 every month, indefinitely, against a one-time switching cost of about one
to two months of fee.
That comparison is why sellers on month-to-month terms switch decisively and sellers on
twelve-month contracts talk themselves into waiting. The transition cost is known and
one-time. The staying cost is invisible and recurring, so it loses the argument despite being
larger.
When the ROI is not there
Three cases where switching will not pay.
The problem is your product economics. If landed cost leaves no margin or the market is
below the about $2 million a year floor where there is enough revenue to capture profitably,
no agency fixes that. Switching buys a new opinion on an unwinnable position.
You have not raised it once. A direct conversation naming the numbers, the unshipped
initiatives, and what you need in sixty days frequently produces a reassignment or a scope
change. That costs nothing and often resolves a caseload problem.
One bad quarter with a good diagnosis. Numbers fall for reasons nobody controls. If the
cause was identified early, explained specifically, and met with a proposal, the partnership
is working even though the results are not.
Expected payback
When a switch was warranted, expect the following shape.
Weeks one to two are diagnosis. Weeks three to six are execution with performance still soft.
Most accounts show measurable ACoS improvement inside the first 30 days of new management,
though the full picture takes longer. Payback on the one-time cost typically lands somewhere
in month three or four.
If nothing has moved by day 90 against the baseline you recorded before giving notice, the
problem was probably not the agency. That is worth knowing too, and it is why recording the
baseline matters.
What most agencies will not tell you
The incoming agency has an incentive to make switching sound cheap, and the outgoing one has
no incentive left at all. Neither will give you the honest arithmetic, so do it yourself with
your own acquisition cost trend.
The other quiet point: most of the switching cost was determined when you signed with the
previous agency. Account ownership, creative source files, and where campaigns physically
live decide whether the transition line is near zero or painful. If you are
currently choosing a first agency, that is the moment to make your future switch cheap.
Related answers
- Switching Amazon agency without losing momentum
- When to switch Amazon agency
- Amazon agency transition timeline and risks
- Month-to-month vs annual Amazon contracts
- Hiring an Amazon agency: the complete guide
Our terms are built so leaving is cheap. That is deliberate, at Flapen.
