Most sellers don’t fire a bad Amazon PPC agency for wasting ad spend. They fire it eight months later, after the wasted spend has already dragged down organic rank, eaten into margin, and shown up in a dozen reports that all said “ACOS is down” while profit quietly wasn’t. By then, switching costs more than staying would have. This guide covers the Amazon PPC agency red flags that actually predict that outcome, not the surface-level ones every agency already knows how to avoid saying out loud.
Quick summary
Why Ignoring Amazon PPC Agency Red Flags Costs More Than the Invoice
The monthly fee is the smallest number in this relationship. The bigger cost is what happens to an account that gets managed passively for six months: wasted spend that never gets recovered, search term data that never gets acted on, and organic rank that slips because nobody noticed sales velocity dropping while ACOS looked fine.
Here’s the mechanic. An agency under pressure to show improvement can cut non-brand spend, let branded campaigns (which almost always run cheap, high-converting traffic from people already searching your name) carry the account’s blended ACOS, and send a report with a green arrow next to it. ACOS genuinely goes down. Total sales and organic rank, which the report doesn’t show, quietly go down with it. The diagnostic question that exposes this every time: what happened to total sales and organic rank while ACOS was falling? That question is worth asking about your own account right now, whether or not you suspect a problem.
Organic rank doesn’t fall and recover on the same timeline, which is why the question matters so much. Rank is slow to lose and slower to rebuild, because it’s built on sustained sales velocity, not a single good week. An account that’s been coasting for two quarters isn’t two quarters away from fixed. It’s closer to six months away, because part of that time gets spent simply rebuilding the velocity the account already had before anyone noticed it slipping. That asymmetry, fast to lose and slow to earn back, is the real reason a passive agency is more expensive than its invoice suggests.
The signs of a bad Amazon PPC agency rarely show up as one dramatic failure. They show up as a pattern of small, defensible-sounding decisions that all happen to make the agency’s own job easier: less time explaining changes, less exposure to a bad quarter, less risk of an uncomfortable question on the monthly call. Nobody sets out to run an account passively. It happens gradually, and the sections below cover exactly where to look for it.
The Amazon PPC Agency Red Flag Scorecard
Grade a current or prospective Amazon PPC agency against this red flag scorecard before you sign anything, or before you decide whether to stay.
One flag on its own might be a fixable conversation. Two or more Critical flags together are the clearest warning signs of a bad Amazon PPC agency, and a reason to walk.
Red Flag #1: They Want Your Login, Not Delegated Access
An agency asking for your Amazon password, or insisting you run everything through their master account instead of yours, is the single clearest red flag in this list, and one of the most common Amazon advertising agency red flags precisely because it’s the easiest one to ask for and the easiest one for a seller to miss. Amazon’s own account guidance is built around a different model entirely: the brand stays the account Owner and grants outside partners scoped access through Seller Central’s user permission system, not a shared login.
The practical risk isn’t abstract. If your ad account, campaign history, and Brand Registry enrollment live under an agency’s login instead of yours, leaving them becomes a negotiation instead of a decision.
A legitimate agency asks for exactly the access it needs (usually Advertising, and sometimes Inventory or Reports) through a proper delegated user, keeps your business as the account Owner, and can explain the permission scope in one sentence. If they can’t do that, ask why not before you ask anything else.
It’s worth checking this periodically even with an agency you trust. Access has a way of outliving relationships. A freelancer or a previous agency you stopped working with two years ago can still show up in your Authorized Partners or User Permissions list, quietly able to see campaign and sales data they have no current reason to have. Reviewing that list once a quarter costs almost nothing and closes a gap most sellers never think to check.
Red Flag #2: Amazon PPC Agency Pricing That Rewards Spend, Not Profit
Amazon PPC agency pricing typically works one of four ways: a flat monthly retainer, a percentage of ad spend, a percentage of total revenue, or a hybrid of a base fee plus performance terms. Percentage-of-spend is the most common model in the industry, and it’s also the one with the built-in conflict of interest. If an agency earns more every time you spend more, “should we increase this budget” and “will this budget increase your profit” stop being the same question, and there’s no natural incentive to ever say no to more spend.
Run the math on your own account. On $30,000 a month in ad spend, a 12% fee is $3,600, whether that spend is compounding profit or quietly leaking it. A flat retainer or hybrid model (base fee plus a bonus tied to profit or TACOS improvement, not raw spend) removes that incentive mismatch entirely.
None of these models is automatically wrong. A percentage-of-spend agency can still do good work. The point is to ask, directly, how the pricing model shapes what they’ll recommend before you ever get to the number itself.
This is also where wasted spend hides in plain sight. Across accounts we’ve taken over, a focused 10-minute review of the search term report is usually enough to surface real, recoverable waste. Our wasted ad spend audit walks through exactly where to look first. If your current agency has never shown you that number, ask for it before you ask about pricing at all.
Red Flag #3: Guaranteed ACOS, Rankings, or Sales Numbers
Amazon Ads runs an auction where your bid sets a ceiling, and relevance, competition, and buyer behavior decide what you actually pay and where you actually land. Amazon’s own ACOS guide is explicit that there’s no single “good” ACOS.
A credible agency talks in ranges and assumptions instead: your historical conversion rate, your margin, your current CPCs, and what’s realistic given all three. “Based on your numbers, we’d expect ACOS somewhere between X and Y once campaigns mature” is a forecast. “We guarantee 15% ACOS” is a sales tactic wearing a forecast’s clothes.
The same logic applies to timelines. Amazon’s own algorithm gives new listings a temporary testing window where the system is still learning how shoppers respond, and new or restructured campaigns need real click and conversion volume before their data means anything. An agency promising dramatic results inside the first two or three weeks is either misunderstanding how the algorithm actually settles, or hoping you won’t check back until the number looks better than it currently does.
Red Flag #4: They Report ACOS and Hide TACOS (and Profit)
ACOS measures ad spend against ad-attributed sales only. TACOS (Total Advertising Cost of Sale) measures ad spend against your entire Amazon revenue, ad-driven and organic combined, which is the number that actually reflects whether advertising is helping or hurting the business. An agency that only ever shows you ACOS is showing you the number that’s easiest to manipulate and leaving out the one that matters more.
The gap between the two numbers can be enormous. Across three brands we’ve worked with in the same category, the one running the lowest ACOS, 15%, generated $42,000 in revenue and $6,100 in profit. The middle brand, running a 28% ACOS, generated $118,000 in revenue and $16,800 in profit, already ahead on both counts. The brand running a 40% ACOS generated $260,000 in revenue and $31,200 in profit, more than five times the profit of the “better” ACOS. Lower ACOS is not automatically better. Profit is the number that decides that, and profit depends on your breakeven ACOS, which is a real, calculable figure, not a benchmark someone quotes from memory.
Most sellers get their own breakeven ACOS wrong by a wide margin. We regularly meet sellers who believe their margin sits around 45% and, once Amazon fees, product cost, freight, storage, and returns are all accounted for, are actually netting closer to 28%. That’s not a rounding error. It’s 17 points of margin the seller didn’t know they’d already spent, and it changes every campaign decision built on the wrong number. Returns matter here too: a product with a 12% return rate has a worse true ACOS than the console shows, because ads paid to acquire sales that later got refunded.
If your agency can’t tell you your breakeven ACOS and explain how they calculated it, they’re not managing to your profit. They’re managing to a ratio.
There’s a category problem hiding underneath this too. A 35% ACOS is a genuine crisis for a product someone buys once and never again. That same 35% is often a bargain for a consumable with a 40% repeat-purchase rate, because the first sale is only the entry point to months of organic reorders that never touch the ad budget again. An agency applying one ACOS target across your whole catalog, regardless of what each product actually is, is using a benchmark that doesn’t fit half of what it’s being applied to.
The reporting itself can also lie politely, and not on purpose. Amazon attributes a sale back to an ad click over a window of time, so “yesterday’s ACOS” is an incomplete number that keeps improving for days as delayed orders trickle in. There’s a second layer to this called the halo effect: a shopper clicks your Sponsored Products ad, leaves without buying, comes back two days later, and purchases through your organic listing instead. That sale happened because of the ad. The ad gets zero ACOS credit for it. An agency that reacts to a single day’s numbers, or treats weekend data as final before the attribution window closes, is often solving a problem that attribution lag created, not one that actually exists in the account.
For more on why a falling ACOS alone proves almost nothing, see the lowered ACOS claim.
Red Flag #5: Vague Change Logs and Vanity-Metric Reports
An Amazon Ads Console dashboard full of impressions, clicks, and click-through rate can look busy and say almost nothing. Ask a simple question of any report you’re handed: what are the three most important changes made to this account this month, and what specific data justified each one? A real answer names search terms, bids, budgets, and dates. “We optimized the campaigns” is not an answer. It’s a placeholder for one.
This pattern has a name in our own account reviews: the report-driven agency, one that spends its energy making the report look good instead of making the account perform better. Cutting spend improves ACOS on a dashboard instantly. It doesn’t show up as a problem until sales and organic rank fall a few weeks later, by which point the report has already moved on to the next month’s green arrows.
A useful test is to compare two consecutive monthly reports side by side. If the format, the metrics highlighted, and even the language are nearly identical except for the numbers, the report is a template, not an analysis. A real monthly update reads differently every month, because the account looked different every month, and the write-up should reflect whatever actually happened, good or bad.
We report profit dollars, the brand-versus-non-brand spend split, and wasted spend recovered, every week, specifically because a monthly summary is too slow to catch this pattern before it compounds. For the full list, see signs of a report-driven agency.
Red Flag #6: Unverifiable “Certifications” and Secret Techniques
Watch for language like “proprietary algorithm,” “secret ranking method,” or “technology that guarantees lower CPCs,” all common Amazon PPC agency red flags in the sales pitch. Amazon PPC has real technical depth, but there is no secret button that outperforms a properly structured account and consistent optimization. If an agency won’t explain its actual process, even at a high level, that’s usually because the process is thinner than the pitch.
Certifications are easier to check than most sellers realize. Amazon runs an official Partner Network directory sellers can search directly, so a claimed “Verified Partner” or “Advanced Partner” status doesn’t have to be taken on faith. If an agency’s site displays a partner badge, look them up in Amazon’s own directory before assuming it’s current.
The alternative to secrecy is a repeatable process an agency is willing to describe in plain terms: tear down the account structure, fix what’s broken, remove waste, then scale what’s proven. Across accounts we’ve run this process on, it’s produced an average 34% reduction in ACOS, and it’s repeatable precisely because it isn’t a trick. It’s closer to a checklist than a secret.
Match type strategy is a good example of the difference between real methodology and a vague promise. Broad match is a research tool, spending controlled budget to discover language buyers actually use. Exact match is where proven, high-intent terms get their own tuned bid and carry the account’s margin. Phrase sits between the two. A search term that proves itself inside broad or phrase gets promoted into its own exact match campaign and then negated at the source, so the account never pays twice to rediscover something it already knows converts. That’s a process an agency can walk you through step by step. A “proprietary algorithm” is not.
Red Flag #7: Fake or Unverifiable Testimonials
The FTC’s Consumer Reviews and Testimonials Rule, in effect since October 2024, explicitly covers agencies that create or sell fake reviews and testimonials, not just the businesses that display them. That’s a meaningful detail. An agency’s marketing page is not exempt from the same fake-review rules that apply to product listings.
Watch for testimonials with no identifiable client, generic stock photos standing in for real people, case studies with no starting point or timeframe, and results quoted as a percentage with nothing underneath it (“increased sales 400%,” from what baseline, over what period, at what ad spend). Real proof looks more specific and more boring: starting ad spend, what changed, the timeframe, and the result, in that order. That’s much harder to fabricate than a glowing quote next to a stock photo.
A related but subtler version of this is the single best case study, polished and repeated everywhere, standing in for typical results. An agency with 100 client relationships and ten genuinely great outcomes can build an entire marketing site around those ten and never mention the other ninety. Ask what share of their accounts actually hit the result they’re showing you, and ask for a reference at a size and category close to yours rather than their single best story. A reference call is harder to stage than a testimonial page.
Red Flag #8: Contracts That Trap You
A 12-month minimum with no performance exit clause puts all the risk on you. So does an auto-renewal buried in the fine print, a large early-termination fee, or language that gives the agency any claim over campaign data, creative assets, or Brand Registry enrollment after you leave.
Ask directly: if this isn’t working after 90 days, how exactly do I leave, and what happens to my account, my data, and my ad history when I do? Get the answer in writing before you sign, not after you’re unhappy. A seller-friendly contract has a defined ramp-up period (commonly 60 to 90 days), a real exit clause after that, and no ambiguity about who owns what once the relationship ends.
Pay particular attention to any clause touching intellectual property or data ownership. Some contracts quietly claim ownership over campaign structures, creative assets, or even historical performance data, which can make a clean handoff to a new agency or an in-house team far messier than it needs to be. If you keep Owner-level Seller Central access the way Red Flag #1 describes, most of this risk disappears on its own, since the account and its history stay yours no matter who’s managing it. Everything else in the contract is a term you’re negotiating against yourself later, under worse conditions than you have right now.
Red Flag #9: Junior Staff Swapped In After the Pitch
The person running the sales call is often a founder or senior strategist. The person who ends up managing your account week to week is sometimes someone you’ve never spoken to, managing 30, 40, or more accounts at once. That’s not automatically a problem. Junior staff can do excellent work with the right oversight. The problem is being sold one thing and handed another without being told.
Ask who specifically manages the account day to day, how many other accounts that person handles, and who reviews their work before it reaches you. If those questions get vague answers, or the senior person from the pitch is never available again, that’s the bait-and-switch pattern showing up early, while you can still walk away from it.
There’s no universal right answer for how many accounts one manager should handle. It depends on account complexity and how much of the work is automated versus hands-on. But a manager juggling 40 or 50 accounts has roughly an hour a week for each one after internal meetings and admin work, which is enough time for basic bid tweaks and not much else. If your account needs more than that and isn’t getting it, the account manager’s workload is the reason, whether or not anyone says so directly.
Red Flag #10: Cookie-Cutter Strategy, No Account Audit
Most accounts we take over aren’t underperforming because of bad bids. They’re structurally broken. The same keyword often lives inside three or four different campaigns at once, which splits Amazon’s performance data four ways instead of letting any single campaign accumulate enough signal to actually optimize against. Mixing broad, phrase, and exact match inside one ad group routes budget toward whichever keyword spends fastest, not whichever one converts best.
An agency that proposes a strategy before looking at your actual account is guessing. A real pre-signature audit should surface something specific, like “your branded campaign is carrying 30% of total spend” or “this search term has spent $400 with zero orders,” not a generic “you need better optimization.” The fix, when an account needs it, isn’t a personality overhaul. It’s giving every keyword and campaign a single, clear job (research, proven performance, ranking, or defense) so the data has somewhere consistent to accumulate. That structural work, done once, tends to keep paying off for months without constant hand-holding.
Watch for the same symptom showing up as identical package tiers across every client regardless of catalog size or margin. A 20-SKU supplement brand and a 400-SKU home goods catalog do not need the same campaign template, the same budget split, or the same bidding cadence, and an agency selling both the same “Growth Package” is optimizing for its own delivery process, not for either business.
10 Questions to Ask an Amazon PPC Agency Before You Sign
Knowing how to vet an Amazon PPC agency comes down to asking questions that require a specific answer instead of a reassuring one. Bring this list to the first call. For the extended version, see questions that make agencies uncomfortable. The ten below are the short version, built for that first conversation.
If the answer to question 10 is vague, that alone tells you whether an audit actually happened before the pitch.
Already Spotted Amazon PPC Agency Red Flags? Do This Next
Start with a 15-minute self-audit of your own account before you decide whether to have the hard conversation or start the switch.
Pull 60 days of your search term report and sort by spend. Flag any term with meaningful clicks and zero orders, that’s money leaving with nothing to show for it. Check the placement report next: spend defaulting into placements nobody deliberately chose is a quiet, common leak. Then count how many campaigns share the same top keyword. If the answer is more than one or two, your data is split and nothing is optimizing cleanly. Most accounts confess their problems after just those three checks, and the number that turns up is rarely small. Sellers who run this audit on their own typically find somewhere around $1,800 a month in recoverable waste, sitting in an account that looked fine on the dashboard the whole time. For the full step-by-step version, see finding wasted Amazon PPC spend.
If the numbers back up what you suspected, review your contract’s exit clause and notice period before you say anything to the agency. Know what you’re obligated to before you raise the issue, not after.
Once the relationship ends, audit the Authorized Partners and User Permissions tabs in Seller Central and remove any access that shouldn’t still be there. This includes agencies you left years ago. That step gets skipped constantly, and it’s the one most likely to come back and bite you later, since old, forgotten access is exactly the kind of thing that turns into a security problem nobody’s watching for. If you’re evaluating a replacement at the same time, run their answers to the ten questions above against what your self-audit just showed you. A new agency that can already speak to the specific waste you found is a very different conversation than one still talking in generalities.
What a Good Amazon PPC Agency Actually Looks Like
Strip away the marketing language and it comes down to five things. You keep Owner-level control of your Amazon account. You can see real search term, placement, and profit data, not just a summary someone built for you. Changes get made and explained on a weekly cadence, not a monthly one. Waiting a month between adjustments means acting on data that’s already stale; see weekly bid changes vs. meetings for why the cadence itself matters. Pricing doesn’t reward spend for its own sake. And forecasts come with ranges and assumptions attached, not guarantees.
A sixth thing matters just as much, even though it’s harder to put in a checklist: honesty about what advertising can’t fix. Roughly half the accounts we’re brought in to “fix” turn out to have a pricing, listing, or inventory problem wearing a PPC costume. A losing main image, a price sitting above what the category will bear, or a stockout that reset the listing’s rank twice in a year will drag down ad performance no matter how the campaigns are structured, because ads amplify whatever they’re pointed at. An agency willing to say “your PPC is fine, your listing is the problem” before taking your budget is telling you something a purely sales-driven agency won’t.
None of that is complicated to ask for. It’s just rarely offered upfront, which is exactly why it’s worth asking for directly.
There’s also a difference between automation and management, and a good agency treats them as separate jobs rather than substitutes for each other. A rule like “if ACOS climbs above 30%, cut the bid” sounds disciplined, but it can’t tell the difference between a keyword that’s genuinely losing money and one that’s still proving itself, and it treats every new keyword’s early, unproven weeks as a failure instead of a normal part of the process. Left alone, accounts run on rules like that tend to drift smaller, safer, and slightly worse every quarter, while still looking fine on a weekly report. Machines are good at dayparting, harvesting search terms overnight, and flagging anomalies at three in the morning. Judging launch timing, reading cash and inventory constraints, and telling the difference between spend that’s bleeding and spend that’s building rank still take a person who actually knows the account.
Across the 600-plus accounts we’ve managed since 2021, spanning more than 3,000 products in 40-plus categories, the agencies (including us) worth staying with are the ones a seller can audit at any time, not just the ones with the best-looking monthly PDF. That’s also the standard behind our own 97% client retention. Sellers don’t stay because a report looked good once. They stay because the account kept getting better after the first quarter, too.
If your current agency can’t answer the questions in this Amazon PPC agency red flags guide, that’s your answer. See how Selouse’s PPC management team runs weekly, transparent optimization on every account.
