Every Keyword Is Its Own Little Economy
Most keyword lists are treated like a shopping list. A column of words. Bulk them into a campaign, set a bid, and off you go, as if the only difference between one keyword and the next is the letters in it.
That's the single most expensive misunderstanding in Amazon advertising, and it comes from treating a keyword as a word rather than what it actually is: a market. A distinct commercial environment, with its own customers, its own competitors, its own prices, its own weather. Two keywords sitting next to each other in your list can be as different as two shops on opposite sides of a city.
What actually lives inside a search term
Take any single term and look at what it really contains. It has its own search volume — the size of the market. Its own customer intent — are these browsers or buyers? Its own competitor set — who else fights here, and how hard? Its own price expectations — what the customer walking in through this term expects to pay. Its own review thresholds — how much social proof you need before anyone trusts you here. Its own conversion dynamics, its own advertising cost, its own level of brand dominance.
None of that transfers to the next term on your list. Which is why treating keywords as interchangeable units in a campaign is like treating "a shop in Mayfair" and "a shop in a retail park off the M6" as the same business because they both sell things. Same activity. Wildly different economics.
Cheap and expensive each mean several opposite things
Here's where it gets properly interesting, and where the market lens earns its keep. When a term is expensive, that could mean several completely different things. It might be expensive because it's fiercely competitive — lots of sellers, hard to win. It might be expensive because the customers are right at the point of purchase and everyone knows that click is worth a lot. Or — and this is the one people miss — it might be expensive for you specifically because Amazon doesn't see your product as relevant to that term.
That last point is worth sitting with, because it's the mechanism underneath the whole thing. Amazon gets paid for the click, not the impression. So they'll happily charge you less when your listing gets clicked more, and charge you more when it doesn't — because a listing people actually click is worth more to them per showing than one they scroll past. On top of that, Amazon earns a commission on the eventual sale, so a product that converts and sells is doing them two favours, not one. Put those together and relevance becomes a price. Be hugely relevant to a term — clicked, converting, selling — and that term gets cheaper for you. Be seen as a poor fit, and the same term gets punishingly expensive, because you're the click Amazon expects to lose money showing.
So "cheap" and "expensive" cut both ways. A term can be cheap because the intent is weak and nobody wants those tyre-kickers — or cheap because you are exactly what those searchers want and Amazon is rewarding your relevance. It can be expensive because the whole market is brutal — or expensive purely because you don't belong there yet. You cannot tell these apart by looking at the price alone. A £2 CPC and a 20p CPC each hide several opposite realities. You have to read the market and your own relevance to it to know whether expensive means "worth it," "brutal," or "you're not welcome here yet" — and whether cheap means "opportunity" or "junk."
Winnability isn't a property of the term. It's a property of you.
Now the part that most keyword advice completely skips. Whether you can win a given market isn't a fixed feature of that market — it depends on your economics.
Two sellers can look at the exact same term, facing the exact same clearing price for a click, and be in completely different businesses. Because the real question was never "what's the CPC?" It's "how many clicks does it take me to make a sale here, what's that sale worth to me, and what margin does it leave me?" A high-converting, healthy-margin product can happily pay a CPC that would ruin a low-converting, thin-margin one on the identical term.
So "is this keyword too expensive?" is an unanswerable question in the abstract. Too expensive for whom? The term that's a graveyard for your competitor might be a goldmine for you, purely because your listing converts better and your margins are fatter. Winnability is yours, not the market's.
The practical shift
What does this change on a Monday morning? It changes how you build and cull. Instead of "here's my list of keywords, bid on them all and see what sticks," you assess each term as a market you're deciding whether to enter: How big is it? Who's already here? What do these customers want and expect? Can my economics win it profitably, or am I about to pay to lose? Is it underpriced right now — an opportunity — or fairly priced and brutal?
The single most common keyword mistake I see is targeting far too many terms — spreading budget across dozens of markets you never seriously assessed, most of which you can't win and shouldn't have entered. The fix isn't "bid on everything" and it isn't "bid on as little as possible" either. It's this: bid on the terms that should convert best for you, and then on as many of those as you can afford while staying within budget. Pick for conversion potential first, then let budget set how wide you go. That way every term you're in is one you had a real reason to enter, and none of them is starved of the money it needs to actually compete — instead of a sprawling list funded so thinly you lose everywhere at once.
A keyword is a market. Treat each one like a business decision to enter or not, judged against your own economics, and your whole approach to advertising quietly gets sharper — and cheaper.
FAQ
Why should I treat each Amazon keyword as its own market instead of just a word to rank for?
Because every keyword has its own search volume, customer intent, competitor set, price expectations and conversion dynamics. None of that transfers to the next keyword, so bidding on them as interchangeable units ignores real differences in economics.
Why is an expensive keyword not automatically a bad one?
A high CPC can mean a term is genuinely competitive, that the customer is close to purchase, or simply that Amazon sees your specific listing as a poor fit for that term. The same price can hide very different realities, so cost alone doesn't tell you whether a term is worth it.
What's the most common keyword targeting mistake?
Spreading budget across too many keywords that were never properly assessed, most of which can't be won profitably. The fix is picking terms based on your own conversion potential first, then funding as many of those as your budget allows, rather than bidding thinly across everything.
About the author
Zamir Cajee is co-founder of This Way Up, a UK business specialising in Amazon marketplace strategy, and co-host of The Upside Podcast, where he and the team break down how Amazon actually works and how it lies to you. Zamir has built multi-million dollar businesses from scratch and has been selling into the EU since 2016.
Not sure which of your keywords are actually winnable at your margins? Reach out to us for tailored guidance.
Subscribe to the This Way Up newsletter for impactful strategies and insights, or tune in to The Upside podcast for expert advice on successfully managing your e-commerce efforts.