Amazon PPC campaign structure looks like an account-tidiness problem. Auto campaign here, exact campaign there, product targeting in its own corner, a naming convention if everyone behaves nicely. Very satisfying. Also dangerously incomplete.
The named mistake I see with self-service brand owners is building campaigns around match types before building them around commercial permission. The account becomes neat enough to audit, but not smart enough to stop bad spend. A €4,000 monthly Amazon Ads budget gets split by auto, broad, phrase, exact and ASIN targeting. Bids are adjusted every Friday. Search terms are harvested. Yet the best-looking campaign may still be pushing a low-margin SKU with 11 days of stock while a higher-margin variant is buried in a shared ad group.
My stance: Amazon PPC campaign structure should be designed backward from the decisions your advertising software needs to make. If the software cannot see which SKU has margin, stock, Buy Box ownership, launch priority and budget permission, the structure is not “simple”. It is just hiding the business logic.
This guide is for brand owners managing their own Amazon Ads from roughly €1.5K per month and up. The goal is not to create the most granular structure possible. The goal is to create a structure that lets you scale winners, contain tests and pause spend before it damages contribution margin.
What the existing advice gets right
The research landscape is useful, but it mostly stops at the media layer.
Amazon’s own documentation explains the hierarchy clearly: campaigns contain ad groups, ad groups contain product ads and targets, and portfolios help group campaigns for reporting and budget control. Amazon also describes portfolios as a way to track performance and share unspent campaign budget inside a portfolio. That matters because budget lives above the keyword level, not inside your tidy spreadsheet.
Capconvert adds a strong operational point: design the account backward from the bulksheet model and portfolio budget engine. If a campaign name cannot be split by a spreadsheet or API, it becomes expensive to manage.
AdLabs frames the core trade-off well: every structure sits on a spectrum between aggregation and segmentation. Too aggregated, and you cannot optimize precisely. Too segmented, and the account becomes an operational nightmare. They also nail the mental model that campaigns have settings, ad groups have targeting, and products have performance.
Perpetua covers the classic workflow: define a goal, choose auto versus manual, harvest search terms, add negatives and optimize bids. BidX is more prescriptive, recommending four campaigns per product when volume allows: one automatic and three manual campaigns, separated because budgets are controlled at campaign level.
All useful. The gap is what most brand owners actually need on Monday morning: which products are allowed to receive more spend today, and which campaigns should be capped even if ACOS looks fine?
The missing angle: structure by profit permission
Most Amazon PPC structure advice starts with targeting mechanics. I prefer starting with four permissions:
- Margin permission: can this SKU afford the target ACOS after marketplace fees, fulfilment, returns and COGS?
- Inventory permission: can we sell more units without creating a stockout or starving another channel?
- Offer permission: do we own the Buy Box, have competitive delivery and maintain enough review strength?
- Strategy permission: is this campaign defending the brand, launching a product, harvesting queries, liquidating stock or scaling profitable demand?
A campaign structure that ignores these permissions forces your team to manage exceptions manually. That is where self-service advertisers lose time. Worse, it creates the illusion that automation is working because bids move, while budget still flows into SKUs that should not be scaled.
In FiveX, this is exactly why advertising data should sit next to product profitability, inventory insights and marketplace analytics. A bid rule is much more useful when it knows the SKU has €7.80 contribution margin, 42 days of stock and stable Buy Box ownership. Without that context, the rule is just optimizing an ad metric.
The FiveX campaign structure model
Here is the structure I recommend for most self-service brand owners. It is not the only possible model, but it gives enough control without turning the account into a museum of tiny campaigns nobody wants to touch.
1. Portfolio level: budget by business purpose
Use portfolios as commercial budget containers, not as decorative labels. A good portfolio tells the team why money exists there.
- DEFENSE | Brand protection: branded search, own ASIN defense and competitor conquest defense.
- SCALE | Proven profit: SKUs with healthy margin, stock cover and conversion history.
- LEARN | Search term discovery: auto and broad campaigns with controlled daily budgets.
- LAUNCH | New products: products where you accept temporarily higher ACOS for ranking and review velocity.
- CLEARANCE | Stock exit: overstocked or end-of-season SKUs where margin rules are different.
This matters because a €3,000 budget should not behave the same way everywhere. In SCALE, you expect contribution profit. In LEARN, you accept controlled waste for search term data. In CLEARANCE, you may tolerate lower margin to release cash from inventory.
FiveX hook: use the FiveX profitability dashboard to set portfolio-level budget caps from retained contribution margin, not last month’s spend. If SCALE produced €9,600 contribution margin after ads and LEARN produced negative €420 while creating three useful exact targets, the next budget conversation becomes much calmer.
2. Campaign level: one goal, one budget, one guardrail set
For a compact catalogue, one product or one tightly related parent ASIN per campaign often works well. For larger catalogues, group variants only when margin, price point, conversion rate, stock cover and target ACOS are similar.
A practical naming convention might look like this:
DE|AMZ|SP|SCALE|B0ABC123|Exact|HeroShelf|ACOS24|M32
That name tells your software and your team the marketplace, ad type, business purpose, ASIN or product family, targeting type, product group, ACOS ceiling and margin band. Is it glamorous? Not at all. But it is machine-readable, and machine-readable beats “Lisa test campaign 2 final final” every single day.
The operator trade-off: more campaigns create more control, but also more budget fragmentation. At €1.5K monthly spend, you do not need 180 campaigns. At €40K across Amazon.de, Amazon.nl and Amazon.com, segmentation by marketplace, purpose and margin band becomes useful.
3. Ad group level: keep product-to-target logic clean
Ad groups should keep products and targets aligned. If five products share the same keywords but have different conversion rates and margins, putting them in one ad group makes reporting easier and decisions worse.
For most brands, I like these ad group rules:
- One parent product family per ad group when variants behave similarly.
- Separate ad groups for different price tiers, even if the keywords overlap.
- Separate ad groups for hero SKU versus accessories or bundles.
- No mixed-margin ad groups unless you are deliberately accepting averaged optimization.
That last line is the little monster. Averages are comfortable, but profit hides there. A 22% campaign ACOS can contain one SKU at 12% and another at 46%. If the second SKU has thinner margin, cash quietly leaks out.
4. Target level: separate discovery from control
Auto and broad campaigns are useful for discovery. Exact and product targeting campaigns are useful for control. Do not ask one campaign to do both jobs equally.
A clean setup for one proven product family could be:
- Auto discovery: low daily budget, harvest weekly, strict negative rules for irrelevant terms.
- Broad research: controlled bids, query mining, no heroic budgets.
- Exact winners: proven search terms with target ACOS by margin band.
- ASIN conquest: competitor or complementary product targets, measured separately.
- Brand defense: own brand and own ASIN protection, separated from non-brand growth.
The mistake is letting discovery campaigns become permanent spend sinks. If an auto campaign finds a search term that converts 18 orders at 19% ACOS, promote it. If it spends €74 over 30 days with no order and no strategic reason, negate or lower the bid. Discovery should create decisions, not just more rows.
Scenario 1: the hero SKU that should not scale
Imagine a Dutch home brand selling a bamboo drawer organizer on Amazon.de.
- Price: €29.95
- Marketplace and fulfilment fees: €8.40
- COGS and packaging: €10.20
- Return allowance: €1.10
- Contribution margin before ads: €10.25, or 34.2%
- Minimum retained profit target: 8%
- Break-even ACOS: 34.2%
- Operating ACOS cap: 26%
The campaign has a lovely 21% ACOS over the last 14 days. Many advertisers would increase budget. But FiveX shows only 9 days of FBA stock cover and inbound inventory arriving in 16 days. Scaling now risks a stockout, lost ranking momentum and a messy relaunch.
The right structure makes the action obvious. Because the SKU sits in SCALE with an inventory guardrail, advertising software reduces the daily budget from €42 to €18, keeps brand defense live at €6 per day and pauses non-brand broad discovery until stock cover is above 21 days again.
Notice the point: ACOS did not fail. The campaign did not fail. The supply constraint changed the permission to spend.
Scenario 2: the “bad” ACOS campaign that is doing its job
Now take a Spanish skincare brand launching a refill pack on Amazon.es.
- Price: €18.50
- Contribution margin before ads: €6.10, or 33%
- Launch budget: €900 for 30 days
- Target launch ACOS: up to 45% for the first 60 orders
- Normal operating ACOS after launch: 24%
After two weeks, the launch exact campaign shows 39% ACOS. In a generic dashboard, someone flags it red. But the product moved from zero to 44 paid orders, generated 19 organic orders, gained 7 reviews and now converts at 11.8% instead of 6.4%. The campaign is expensive, yes. It is also performing the launch job it was given.
That is why the campaign belongs in LAUNCH, not SCALE. After the 60th paid order or once review velocity stabilizes, it can graduate into SCALE with a 24% ACOS cap. Until then, mature-product rules would kill the launch too early.
FiveX hook: FiveX lets teams compare paid orders, organic lift, contribution margin and inventory movement at SKU level, so a launch campaign is judged by launch economics instead of being punished for not yet behaving like a mature bestseller.
Scenario 3: the shared ad group hiding a margin leak
A Belgian electronics accessory brand advertises three USB-C hubs in one exact-match ad group because the keywords overlap.
- Hub A: €39.95 price, €13.20 contribution before ads, 31% ACOS cap.
- Hub B: €29.95 price, €6.40 contribution before ads, 21% ACOS cap.
- Hub C bundle: €49.95 price, €17.80 contribution before ads, 35% ACOS cap.
The ad group reports 24% ACOS. Looks acceptable. But the search term “usb c hub hdmi ethernet” sends most clicks to Hub B, which cannot afford 24% ACOS if the team wants to keep 6% retained profit. Hub C could afford the traffic and has a better conversion rate, but it is not getting budget priority.
The fix is structure, not a clever bid tweak. Split Hub B into its own exact campaign with a 21% cap, put Hub C in a higher-permission SCALE campaign, and let software allocate bids by margin band.
Rules for smaller advertisers spending €1.5K to €5K per month
If your budget is modest, do not copy an enterprise account with thousands of campaigns. You need clarity, not complexity.
- Start with your top 5 to 15 SKUs, not the full catalogue.
- Create separate portfolios for LEARN, SCALE and DEFENSE first.
- Use one auto discovery campaign and one exact winner campaign per hero product family.
- Keep daily discovery budgets intentionally small: often €3 to €10 per product is enough to learn.
- Review search terms weekly, not hourly. Panic is not a strategy, even when the CPC graph looks spicy.
- Set ACOS caps from contribution margin before ads, not from a category benchmark you found online.
For a €2,500 monthly budget, a practical split could be €1,250 SCALE, €500 DEFENSE, €500 LEARN and €250 LAUNCH or seasonal tests. The point is to stop discovery from accidentally eating the month.
Rules for larger accounts spending €10K+ per month
Once spend grows, the bottleneck changes. The question becomes less “can we create campaigns?” and more “can we govern them without drowning?”
- Use naming fields your software can parse: country, marketplace, ad type, purpose, product family, targeting, margin band and ACOS cap.
- Separate brand, category, competitor and product targeting because incrementality and intent differ.
- Create margin bands, for example M15, M25, M35, so bid rules can react differently.
- Connect inventory thresholds to budget rules: reduce discovery below 21 days of stock, pause scale below 14 days unless replenishment is confirmed.
- Audit search term duplication. The same query competing across five campaigns is not a strategy; it is internal auction confusion with a nicer name.
FiveX hook: FiveX can combine Amazon Ads, marketplace profitability, stock and repricing context so teams see which campaigns deserve human attention. Instead of opening every campaign, operators can filter for “spend increased, margin below target, stock below 21 days” and fix the exceptions first.
The weekly operating rhythm
Campaign structure only works if the team uses it. My preferred weekly rhythm is simple:
- Monday: check stock cover, Buy Box issues and margin changes before touching bids.
- Tuesday: harvest search terms from LEARN into exact campaigns where conversion and margin allow it.
- Wednesday: review SCALE campaigns against contribution margin and TACOS, not ACOS alone.
- Thursday: check launch and seasonal campaigns against their specific success criteria.
- Friday: clean negatives, duplicate targets and budget caps before the weekend.
This rhythm is deliberately boring. Boring is good. Marketplace advertising gets expensive when every week becomes an improvisation session.
How to know your structure is working
Your campaign structure is healthy when these things are true:
- You can explain every portfolio in one sentence.
- You know which campaigns are allowed to scale and why.
- Search term harvesting has a destination, not just a spreadsheet.
- Budget caps reflect product economics and stock reality.
- Launch campaigns are not judged by mature-product rules.
- Automation rules can act on margin, inventory and offer status, not only clicks and orders.
If you cannot answer those points, do not add more automation yet. First fix the structure. Then let software accelerate the right operating model.
Final take
The best Amazon PPC campaign structure is not the prettiest. It is the one that makes profitable decisions easier.
Auto, broad, exact and ASIN campaigns still matter. Naming conventions still matter. Portfolios still matter. But they matter because they help you control budget by business purpose, SKU economics and operational readiness.
For self-service brand owners, this is where advertising software earns its place. FiveX brings marketplace ads, profitability, inventory, repricing and analytics into one view, so your campaigns are not optimized in isolation. The software should not simply ask, “Which keyword deserves a higher bid?” It should ask, “Is this product commercially allowed to grow today?”
That is the structure worth building.