Skip to content
You’re spending $30,000 a month on Amazon ads. TACOS is sitting at 18%. You hired an agency six months ago because managing PPC yourself wasn’t scaling. But the numbers haven’t moved, and every monthly call sounds like a vocabulary lesson, not a diagnosis.
So you’re back to the same question: is the problem the agency model itself, or did you just hire the wrong one? And would a consultant do better?
Here’s how to think through it without getting sold to.
What an Amazon PPC Agency Actually Does (vs. What You Think You’re Buying)
Most Amazon PPC agencies operate with one account manager handling 15–30 accounts simultaneously. They use software automation—Perpetua, Pacvue, Skai—to execute bid adjustments at scale. The software isn’t the problem. The problem is that software optimizes for what it can measure, and it measures backwards-looking data: last week’s ACoS, last month’s impression share.
It does not optimize for your inventory position, your seasonal CVR curve, or the fact that three of your top five keywords are being cannibalized by a competitor running 80% share of voice. Those decisions require judgment. Judgment requires time. And time is the one thing an agency with 25 accounts can’t give you.
What you’re often paying for at $3,000–$6,000/month is competent execution of a generic playbook. That playbook works fine for accounts that just need maintenance. It doesn’t work for accounts with a TACOS problem, a launch problem, or a listing CVR problem—because those require diagnosis first, execution second.
Where a Consultant Works Better
An Amazon consultant with a limited client roster—typically 5–12 active accounts—can do something an agency structure rarely allows: spend three hours inside your Search Query Performance data before making a single bid change.
The value isn’t cheaper PPC management. A good consultant often costs similar to or more than a mid-tier agency on a per-hour basis. The value is that they’re diagnosing before prescribing. When TACOS is high, there are at least five distinct causes:
- Broad match waste on irrelevant ASINs
- Organic rank loss on top keywords (so PPC is propping up what used to be free traffic)
- CVR drop from listing deterioration (more clicks needed per sale = higher spend)
- Competitor conquesting eating your impression share on brand terms
- Budget allocation skewed toward awareness when your conversion funnel is the actual problem
An agency running your account inside a dashboard sees ACoS by campaign. A consultant with 12 clients and six hours to work your account this week sees which of those five problems is actually driving your number.
The Structural Question: Who Owns the Strategy?
The clearest way to decide: does your current setup have someone who owns the question ‘why is this happening’ rather than ‘what should we change next’?
If your agency sends monthly reports with metrics and quarterly strategy calls, they own execution. You own strategy by default, which means you’re doing two jobs.
If you’re an operator at $100K–$500K/month, you probably don’t want to own both. But hiring a consultant doesn’t automatically solve it either if the consultant is also just executing—taking your brief, running your campaigns, and reporting back. The question to ask any prospective consultant: ‘Walk me through the last time you told a client their TACOS problem wasn’t a PPC problem.’ If they can answer that with a specific mechanic, you’re talking to someone who diagnoses. If they pivot to their process deck, you’re not.
A Practical Framework for the Decision
Use an Amazon PPC agency when: your account is at steady state ($20K+/month spend, established rankings, stable CVR), you need execution bandwidth rather than strategic input, and you have internal capacity to set direction and review results critically.
Use an Amazon consultant when: you’re launching a new product line and the honeymoon window matters, your TACOS has drifted more than 4–5 points in 90 days without a clear external cause, your organic rankings have declined and you suspect the PPC strategy is a symptom rather than the fix, or you’ve cycled through one agency already.
The burned-by-agency seller is real. About a third of established operators who reach out have paid an agency for 6–12 months and have data that looks similar at the end as at the start. That’s not always agency malpractice—sometimes it’s a listing problem, sometimes it’s a catalog architecture problem. But it is a diagnostic failure, and diagnostic failures are what the consultant model is built to fix.
What to Look for in Either Option
Whether you’re evaluating an agency or a consultant, four things matter more than the pitch deck:
- SQP fluency: Can they read a Search Query Performance report and tell you something you don’t already know within 10 minutes?
- Organic-first thinking: Do they understand the relationship between PPC velocity, CVR, and organic rank movement—or do they manage PPC in isolation?
- TACOS as the primary metric: Anyone optimizing primarily to ACoS is optimizing for the wrong thing if you care about profitability at the business level.
- Clear scope on what they don’t do: The right partner tells you what’s outside their wheelhouse. The wrong one says yes to everything.
If you’re at the point where you’re evaluating both options and want a candid read on which fits your situation, request a free audit. I’ll look at your actual account data and tell you whether the problem is structural, executional, or something else entirely before we talk about next steps.
For more on vetting specifically, read the 9 questions that separate operators from order-takers. And if you’ve already identified agency red flags in your current setup, this post on what those look like in practice is worth reading before you make a move.
