Amazon PPC advertising sells placements at auction across search results and product pages, and you pay only when a shopper clicks. Three families exist, Sponsored Products, Sponsored Brands, and Sponsored Display, and each one fails for a different reason, so diagnose the symptom first. Thin impressions, thin clicks, and thin orders have three separate owners.
The short version
- Three families, three jobs. Sponsored Products wins the search term, Sponsored Brands buys the top row, Sponsored Display retargets shoppers who already looked.
- The symptom names the owner. Thin impressions is a bid problem, thin clicks is an image problem, thin conversion is a product page problem.
- Break-even sits at your contribution margin before advertising. Every published category average describes somebody else's landed cost.
- About $1,000 a month in media is where optimization starts to mean something, and we hold no hard minimum under it.
- Caseload decides whether the fix happens. Our 50 operators run about 70 brands, about 1.4 each, so every ad account carries one name.
What you are buying at the auction
Amazon sells each placement by auction, and you pay when a shopper clicks rather than when the ad appears. Sponsored Products puts one ASIN into search results and onto competitor product pages.
Sponsored Brands buys the headline row above the results with your logo. Sponsored Display follows shoppers who viewed your page or a rival's.
Most sellers run all three and then read them as a single number. A search term problem, a creative problem, and a retargeting problem all surface as one worse ratio at the account level. Separate them by campaign type before anyone opens the bid column.
Budget sets the speed at which you learn anything. We hold no hard minimum on ad spend, and about $1,000 a month is the point where optimization becomes meaningful. Under that, a week of clicks is too thin to read.
Read the symptom before you touch a bid
Weak advertising results land in one of seven symptoms. The cause sits a layer under the symptom, and the owner sits a layer under the cause. Diagnose in that order, and the bid column is often the last thing you open.
| Symptom | Cause underneath it | Who fixes it |
|---|---|---|
| Impressions barely register | The bid sits below the auction floor, or the term carries no volume | Whoever owns bids and keyword harvesting |
| Impressions high, clicks low | Main image, price, or review count loses the comparison in the results row | Creative, with whoever sets price |
| Clicks high, orders low | The product page, not the campaign | Listing and product, never the ad manager alone |
| Orders healthy, ACoS above break-even | Landed cost and fees, not the auction | Sourcing and finance, with the ad manager |
| ACoS flat, TACoS climbing | Advertising is buying the sales the organic base used to deliver | Whoever owns the traffic plan |
| Spend pooled in a few loose terms | No negatives, and no structure between exact, phrase, and broad | Whoever reads the search term report every week |
| Sales steady, returns rising | A product defect that advertising is amplifying | Quality control, and stop the spend first |
Two rows are often true at once. Work them in the order the money moves, and re-measure after a full week rather than the next morning.
When two rows stay true after a full cycle, the question stops being advertising. Read the four signals instead: rating trend, return rate, conversion rate, and cost of customer acquisition trajectory. If none of the four improves inside 60 to 90 days, the honest answer is to kill the product rather than rebuild the campaigns.
The third column is the one buyers never test
Symptom and cause are published knowledge, available on any blog including this one. The third column is where accounts live or die, because a fix with no owner is a line in a slide deck. So test the column that names a person.
Ask any provider how many accounts each person carries, and ask for the number rather than the reassurance. Fifty operators run about 70 brands here, which works out to about 1.4 brands per operator. That ratio is what makes a weekly search term report per account possible instead of aspirational.
The reporting cadence follows the same logic: a written update in Slack every week, a live review every two weeks, and access in between. Onboarding runs in order, audit first, then a brand manager assigned, then blockers identified, then execution. Measurable ACoS improvement typically lands inside 30 days of that sequence.
The audit itself is written, free, and back within 48 hours, across seven areas: listing quality, primary image click-through, conversion rate, ad performance, traffic channel activation, pricing, and return rate. Advertising is one of the seven. A report that arrives as campaign screenshots alone has skipped most of the list.
Geography changes the answer too. We operate all 23 Amazon marketplaces and produce content in English, German, Spanish, and French. German shoppers do not type translated American search terms, so keywords and negatives are built per country.
Hold us to the same test you run on anyone else. If we cannot name the operator on your account and the number of brands that person carries, do not hire us.
What an ads-only provider will not tell you
Three things stay out of the monthly reporting call, and on a careless month that includes ours. Each one is the same diagnostic in reverse: the number you are shown, the thing underneath it, and the person who has to move.
| What the report shows | What sits underneath it | Who has to fix it |
|---|---|---|
| Spend and sales both up | Budget grew, efficiency did not | You, by asking for profit per ASIN beside ad cost per ASIN |
| ACoS improved sharply | Rank-building spend was cut, and the organic base pays for it later | The person who agreed the target, in writing |
| Every test in the summary worked | The losing variants disappeared quietly | The provider, by publishing the tests that failed |
The first row is the common one. Spending more lifts revenue and impressions together, which is why those two metrics headline so many decks. Ask for profit per product beside advertising cost per product instead.
The free audit is a sales instrument, ours included. Judge it by whether the fixes name your ASINs, quantify the impact, and could be executed by someone you already employ.
Related answers
- ACoS meaning explained
- PPC on Amazon
- Amazon suspension appeal help in the UAE
- Compare Amazon SEO services for FR DE IT ES
- Amazon account measurement and audits: the complete guide
One free thing to do this week. Pull 60 days of search term data for your strongest product, sort by spend with zero orders, and total that column. That figure is what a negatives list would have handed back, and the list costs an hour to write.
For a written read of all seven audit areas on your own account before you pay anyone, request the free 48-hour audit from Flapen.
