Write for the Question, Not the Keyword

For twenty years, Amazon trained all of us to think in keywords. Customers typed fragments, “running shoes men size 10 waterproof”, and we wrote listings to match those fragments. Fair enough; that's how search worked. But the fragment was always a compressed, slightly mangled version of a real question the customer had in their head. And now, increasingly, they're asking the real question out loud — to an AI that reads your listing for the answer.

So the instruction for this era is simple to say and surprisingly hard to do: write for the question, not the keyword.

Overhead view of two hands typing on a laptop with a large question mark graphic over the keyboard, next to a phone, notebook and pencil on a wooden desk.

The keyword was always a proxy for a question

Here's the thing I want you to see: the keyword was never the real thing. It was a proxy. "Two way radio long range" was never what the customer wanted — it was the crude search-box compression of "I need radios that'll actually work across a noisy building site without cutting out." The intent was the question. The keyword was just what survived being forced through a search box.

This matters because everything I believe about Amazon starts from consumer psychology: you're never really optimising words, you're optimising how a specific person, with a specific need, makes a decision. The keyword mattered only as a clue to that person's intent. We tolerated the crude proxy because the search engine was crude. Now the interface can take the actual question — and reward the listing that actually answers it. The proxy layer is thinning, and what's underneath is the real customer intent we should have been serving all along.

What "writing for the question" looks like

It's a genuine shift in how you draft, not just a tone change.

Start from the buyer's real questions. Before you write a word, list the actual questions a customer in your category asks on the way to buying — the pre-purchase worries, the "will it work for my situation?" queries. And use the sources Amazon gives you for free: the customer Q&A section posted on most listings, the questions buried in reviews, the pre-sale messages. The Q&A section in particular is a gift — real customers posting the exact things they need to know before buying — though bear in mind the answers there come from other customers and aren't always right, so trust the questions and check the answers yourself. That combined list is your brief. Not a keyword tool's export — the real human questions.

Answer them plainly in the listing. For each important question, make sure the listing contains a clear, specific answer. Not a keyword that vaguely gestures at the topic — an actual answer. If people ask "will this fit a standard car cup holder?", the listing should state the base diameter and, ideally, just say it fits standard cup holders. You're writing responses, not scattering terms.

Use natural language, because the reader is now natural-language. Both the human and the AI reading over their shoulder handle plain, specific sentences far better than compressed keyword strings. A bullet that reads like a helpful answer from a knowledgeable shop assistant will out-communicate — and now out-*index* in AI-mediated discovery — a bullet that reads like a search query.

Match the depth of intent. Some questions are top-of-funnel ("what even is a good X?") and some are bottom ("does this specific X do Y?"). A listing that answers both the orientation questions and the precise deciding ones serves the customer wherever they are in their thinking — and the AI can pull the right answer for the right shopper.

Why this doesn't mean abandoning keywords

Let me head off the obvious objection, because I'm not telling you to throw keywords in the bin. Traditional keyword search still runs in parallel, so the terms still need to be present — a listing that answers questions beautifully but never contains the words people search will still struggle in the classic results. The point isn't keywords versus questions. It's that you get the keywords in by answering questions well, rather than answering questions as an afterthought to stuffing keywords. Write the genuine answer, in natural language, and the relevant terms tend to fall out of it naturally — because the words people use to search are, mostly, the words that appear in a good plain-language answer to their question.

That's the resolution: write for the question first, and you get the keywords for free, in context, in a form both the human and the machine reward. Write for the keyword first, and you get a stuffed listing that answers nobody's actual question — and, increasingly, wins nobody's recommendation.

The customer was always asking a question. The search box just hid it from us for two decades. Now it's coming back into view. Write for it.


FAQ

  1. What does it mean to write an Amazon listing for the question instead of the keyword?

    It means starting from the real questions customers ask before buying, drawn from the Q&A section, reviews and pre-sale messages, and answering them plainly in the listing, rather than just working keywords into the copy.

  2. Does writing for the question mean I should stop using keywords?

    No. Traditional keyword search still matters, so the terms still need to appear. The shift is in sequence: answer the real question in natural language first, and the relevant keywords tend to appear naturally within that answer.

  3. Where can I find the real questions customers are asking about my product?

    The customer Q&A section on your listing, questions embedded in reviews, and pre-sale messages. Trust the questions as genuine customer intent, but check the answers yourself, since Q&A answers come from other customers and aren't always accurate.


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.


Want your listing rewritten to actually answer what customers are asking? 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. 

Next
Next

Every Keyword Is Its Own Little Economy