Find Amazon competitor keywords: build a conquest permission ledger before PPC spends

Finding Amazon competitor keywords is easy now. Almost too easy. Put a rival ASIN into a reverse-ASIN tool, export the phrases they rank for, sort by search volume, add the biggest terms to a Sponsored Products campaign and wait for market share to move. It feels decisive.

It is also how many self-service brand owners turn useful competitor research into expensive PPC noise.

The named mistake I see is conquest-by-export: treating every competitor keyword as a campaign target simply because a tool found it. The keyword may be real. The competitor may be relevant. The search volume may look attractive. But none of that proves your product deserves to buy that click today.

My stance: competitor keywords should pass a conquest permission ledger before they receive ad spend. That ledger asks whether your offer can win the comparison, whether the margin can afford the higher click cost, whether stock can support extra demand, whether the keyword role is clear, and whether your advertising software can stop the test before curiosity becomes waste.

This guide is for brand owners managing Amazon Ads themselves, usually from around €1.5K monthly marketplace ad spend. At that level, competitor keyword research can absolutely unlock growth. But the win rarely comes from stealing every rival keyword. It comes from choosing the few competitor terms where your product has a believable reason to convert.

What the existing advice gets right

The public guidance around competitor keyword research is useful. Helium 10 positions Cerebro as a reverse-ASIN workflow for discovering competitor keywords and finding search terms for your own listings. SmartScout describes competitor analysis as a repeatable workflow that should look beyond keywords into listings, reviews, ads and pricing. Perpetua makes an important strategic point: competitor conquesting can become expensive because Amazon’s ad auction cares about relevance, click-through rate, conversion rate, price point and margin. BidX, Quartile and Teikametrics all show where the market is going: faster campaign creation, AI bidding, marketplace intelligence and automation that can act much faster than a human operator.

YouTube tutorials tend to focus on the mechanics: paste ASINs into a tool, filter keywords, build PPC campaigns, target rival product pages. Reddit and seller-forum conversations are more nervous. Operators worry that competitor targeting burns budget, that big brands are impossible to outbid, or that tools surface thousands of phrases without telling them which ones deserve action.

The gap is not keyword discovery. The gap is keyword permission. Most advice helps you find competitor keywords. Less advice helps you decide which of those keywords should be allowed to spend, which belong in listing copy, which should become product-targeting tests, and which should be parked until the product, price or review base improves.

Start with the competitor set, not the export

A reverse-ASIN export is only as good as the ASINs you feed into it. The first operator habit is to separate competitors into three sets:

• True substitutes: products a shopper could buy instead of yours today.
• Aspirational leaders: category winners with more reviews, stronger content or better authority than you currently have.
• Bad-fit giants: huge products that share a broad category but attract shoppers with different intent, price expectations or use cases.

Only the first set deserves immediate PPC attention. Aspirational leaders are useful for learning. Bad-fit giants are useful for humility.

Example: LunaKettle sells a €36.95 stainless steel gooseneck kettle. The obvious competitor export contains keywords from a €119 smart temperature-control kettle, a €24 travel kettle and a €42 pour-over kettle with 4,800 reviews. All three are “kettles”. Only one is a true substitute. The smart kettle teaches content gaps. The travel kettle attracts a different buyer. The pour-over competitor is the only ASIN whose keywords deserve a near-term ad test.

That distinction prevents the first budget leak: buying traffic from shoppers who were never making your comparison.

Score each keyword by comparison advantage

Competitor keyword research should not start with search volume. It should start with the comparison a shopper will make after the click.

I like a simple five-point score:

• Price advantage: are you at least 5-15% cheaper, or clearly premium for a reason?
• Review position: are you close enough in rating and review count to feel credible?
• Content fit: does your title, image stack and A+ content answer the same use case?
• Delivery promise: can you match Prime or the marketplace’s expected delivery speed?
• Margin headroom: can the SKU survive the likely CPC and lower conversion rate?

Give each keyword a green, amber or red status. Green means the shopper comparison helps you. Amber means the keyword belongs in a small learning test. Red means it should not receive paid traffic yet.

Scenario one: LunaKettle finds “electric gooseneck kettle” at 18,000 monthly searches. Its true substitute competitor sells at €42.95, has 4.4 stars and 1,200 reviews. LunaKettle sells at €36.95, has 4.5 stars and 340 reviews, contributes €11.80 before ads and has 54 days of stock. If the expected CPC is €0.68 and the cold competitor-keyword conversion rate is 7%, the ad cost per order is roughly €9.71. That leaves about €2.09 contribution before any returns or coupon cost. The term is not a scale target yet. It is a €15-per-day learning test with a bid ceiling and a stop rule.

Scenario two: NorthPeak sells a 12-pack protein bar for €22.90 with €6.40 contribution before ads. A competitor keyword export surfaces a large rival brand term with 41,000 searches. NorthPeak has 85 reviews; the rival has 9,000. CPC estimates sit around €1.10, and expected conversion on a competitor brand query may be only 3%. That is about €36.67 ad cost per order. The keyword is commercially red, even if the tool ranks it as “high opportunity”. The right move is not PPC conquest. It is content learning: study the rival’s claims, flavours and image angles, then use that insight in listing optimization and maybe defensive category targeting later.

Give every competitor keyword a campaign role

A messy competitor campaign usually has one label: “competitor keywords”. That is not enough. Your advertising software needs a role, because the role determines the bid, budget, success metric and patience level.

Use four buckets:

• Conquest: competitor terms where your offer has a clear advantage and you want to win incremental demand.
• Defense: your own brand, product and ASIN-adjacent terms where competitors may appear.
• Substitute learning: terms that may work but need proof before scaling.
• Listing intelligence: competitor phrases that belong in copy, backend search terms or content tests, not paid campaigns yet.

The role matters because a 30% ACOS can be fine for a launch conquest keyword and unacceptable for a mature defensive term. A substitute-learning keyword may need 40-60 clicks before judgement. A defensive keyword might need a lower bid but near-continuous visibility. If your software treats them as one pool, it will either overcut learning or overspend on curiosity.

This is where FiveX should sit in the workflow. Use FiveX marketplace advertising views to connect campaign performance with product profitability instead of judging only campaign ACOS. Use product-level margin and cost fields to calculate whether a competitor keyword has headroom. Use stock and inventory signals to stop a conquest test from accelerating a SKU with only 12 days of cover. The keyword export starts the conversation; the operating data decides permission.

Translate competitor keywords into ASIN tests carefully

Many competitor keywords point to a better tactic: product targeting. If a shopper is comparing specific products, an ASIN-targeting campaign can be more precise than fighting the search auction. But product targeting needs the same permission logic.

For each competitor ASIN, record the reason you deserve the placement. “They sell a similar product” is weak. Better reasons are: you are 12% cheaper, your rating is 0.3 stars higher, you include an accessory they charge extra for, your pack size fits a different use case, or your delivery promise is stronger.

Example: UrbanNest sells a storage basket at €28.50 with €9.20 contribution before ads. A competitor ASIN has 3.9 stars, 620 reviews and recurring complaints about weak handles. UrbanNest has 4.6 stars, 210 reviews and reinforced handles shown in image two. A product-targeting test at €0.42 CPC with a 10% expected conversion rate costs about €4.20 per order. That still leaves €5.00 contribution before returns. This ASIN deserves a controlled test. A rival at 4.8 stars, 6,000 reviews and a lower price does not.

The same rule applies to bol.com and other marketplaces when competitor or category targeting is available. The channel mechanics differ, but the commercial question is identical: do we have a comparison advantage strong enough to buy this attention?

Build stop rules before launch

Competitor keywords are slower to judge than exact non-brand terms because the shopper did not start with you in mind. That does not mean they deserve unlimited patience.

Before launch, define three stop rules:

• Spend stop: pause when spend reaches the smaller of €35 or three times expected contribution without an order.
• Click stop: pause or reprice after 50 clicks with no sale, unless the keyword is explicitly a launch-learning term.
• Profit stop: reduce bids when actual ACOS exceeds the SKU’s break-even ACOS by more than 20% after two orders.

These numbers are not universal. They are a starting point. The important part is that the rule exists before the keyword goes live. A bad competitor test feels interesting while it is running. A ledger makes it boring enough to stop.

FiveX ad automation rules can help here by turning the stop logic into repeatable controls: pause targets when spend crosses the learning cap, reduce bids when ACOS breaches the product’s threshold, or flag products where stock cover no longer supports conquesting. FiveX Ads AI recommendations can then work inside those guardrails instead of optimizing toward a target that finance would never approve.

The practical workflow

1. Pick 3-5 true substitute ASINs. Ignore broad category giants unless you are only learning from them.
2. Export competitor keywords. Use reverse-ASIN tools, Brand Analytics, Search Query Performance, ad search-term reports and manual SERP checks.
3. Remove weak-intent phrases. Exclude terms where the shopper wants a different size, feature, brand tier or price bracket.
4. Score comparison advantage. Price, reviews, content, delivery and margin decide permission.
5. Assign a role. Conquest, defense, substitute learning or listing intelligence.
6. Set bid ceilings and stop rules. Do this before launch, not after the spend feels uncomfortable.
7. Review with profit, not ACOS alone. Bring in contribution margin, stock cover and channel role.

The trade-off is obvious: this is slower than dumping 2,000 competitor keywords into campaigns. Good. Speed is not the scarce resource. Clean permission is.

How FiveX helps

FiveX is useful because competitor keyword decisions do not live only inside Amazon Ads. They need product margin, stock, channel performance and automation control in the same operating view.

With FiveX, a brand owner can connect Amazon Ads and marketplace data, see which SKUs have enough contribution margin for conquest tests, monitor whether stock still supports demand acceleration, review ad recommendations before applying them, and build rules that pause or reprice targets when performance leaves the approved range. That is the difference between “we found competitor keywords” and “we know which competitor keywords are allowed to spend”.

My operator rule is simple: never let a keyword tool decide budget. Let it discover the battlefield. Then let margin, offer strength, stock and campaign role decide whether you should enter.