Writing a Buyer Guide for a Product Category You Don't Know: The Two-Hour Research Routine Before the Brief
The keyword is good, the competition is thin, and you've never owned, used or thought about the product. Every affiliate site gets here, and the guide that comes out depends on what happens in the next two hours.
Skip them and brief the writer with the keyword alone, and you get a competent page about nothing in particular: ten bestsellers, the listing's headline spec explained, a verdict that could be any of them. Spend them, and the brief carries what a buyer in that category actually weighs, and the guide reads as if someone knew.
This post covers the six things you need to know before briefing a writer on an unfamiliar category and why each one changes the guide, the two-hour routine that gets them in a fixed order (Amazon's own category page, the top guides' headings, one-star reviews, a forum thread, a spec sheet), what you still won't know at the end and how the guide should say so, and the brief that comes out the other end.
The prompting post covers the six inputs a brief needs; this is where four of them come from when you don't already have them.
The short answer
Before the brief, learn six things: reader types, deciding specs, price tiers, brands, complaints, vocabulary. Get them in two hours: Amazon's category filters (30 min), the top three guides' H2s (20), the one-star reviews on three bestsellers (30), one forum thread (20), one spec sheet (20).
Write the method section to say you compared data, not tested. Then write several guides in the category, because the two hours amortise.
The six things
| You need to know | Why the guide needs it | Where it comes from |
|---|---|---|
| Reader types | Which readers the picks serve; the "pick this if" lines | The top guides' section names; the forum's recurring "I'm a ... and I need" |
| Deciding specs | The three to five things the comparison turns on; the table's columns; what to explain | Amazon's filter sidebar; the specs the one-star reviews complain about |
| Price tiers | Where the steps are and what each buys; the budget pick and the stretch pick | Amazon's price filter and the clustering of bestsellers |
| Brands | Which to include, which buyers avoid, which are rebadged imports | The brand filter's order; the forum's opinions; repeated names across the top guides |
| Complaints | The caveats; the cons that aren't pros inverted; the listing claims to distrust | One-star and three-star reviews on the three bestsellers |
| Vocabulary | So the guide says "cut-off beam" not "shaped light"; so the FAQ asks what buyers ask | The forum thread; the top guides; the spec sheet's terms |
The two hours
- Amazon's category page and filters (30 minutes). Open the category, sort by bestsellers, and read the left sidebar: the filters Amazon offers are the specs the store has learned buyers sort on, and their order roughly follows importance. Note the price ranges in the filter and where the bestsellers cluster; those are the tiers. Note the brands in the brand filter and which appear in the top twenty. Click into three bestsellers at different prices and read the specs tables side by side; the rows that differ are deciding-spec candidates.
- The top three guides' headings (20 minutes). Search the keyword, open the top three guides, and copy their H2s as a list of topics. The competitor guides post covers the fuller ten-line read; here you want the section names (the reader types and the questions) and the picks that appear in all three (the consensus to check, not to copy).
- One-star reviews on three bestsellers (30 minutes). Filter each to one and three stars, read until the complaints repeat, and write down the three or four patterns per product. The one-star reviews post covers telling a pattern from a grudge. These become the caveats and the specs that need explaining ("runtime is quoted at the lowest setting").
- One forum or subreddit thread (20 minutes). Search "[category] recommendation" on the relevant subreddit or forum and read one long thread. You're not after picks; you're after the words people use, the brands they dismiss and why, and the reader types that keep appearing ("I'm a beginner and", "for a small flat").
- One spec sheet (20 minutes). A manufacturer's product page or manual for one mid-tier product, read for what the numbers mean: what the unit is, what a high and low value does in use, what the marketing names translate to. This is where "IPX4" becomes "splash-proof, not submersible" in the guide.
Write it on one page as you go: six headings, the notes under each. That page is the brief's raw material and it's reusable for every guide in the category.
What you still won't know
Two hours of reading doesn't tell you how the product feels, whether a complaint that's common online is rare in practice, or which of two close specs matters more in a year of use. The guide should say so, in the method section: that it compared current products on specifications, owner review patterns and price for a stated reader, and didn't test them.
That sentence is honest, and it's more credible than a borrowed claim of expertise. Where the research left a real gap (two specs you can't rank), the guide says which reader should care and points them to where the answer is, rather than guessing.
The brief that comes out
- Reader: the two or three types from the forum and the top guides, with the one this guide is for named first.
- Deciding specs: the three to five from the filter sidebar and the complaints, with the headline spec to dismiss in a sentence.
- Constraints: the price tier this guide covers, from the filter clusters.
- Brands: the ones to include and any to exclude, with the reason from the forum.
- Banned claims: the listing claims the one-star reviews contradict ("waterproof", "all-day battery").
- Vocabulary: five to ten terms the guide should use, as buyers use them.
- Method line: "compared on specs, review patterns and price; not tested".
That's the six-input brief from the prompting post, filled from the two hours. The second guide in the category reuses five of the seven lines.
Doing it with AffSERP
AffSERP's Keyword Explorer shows the category's query shapes, which is the reader-types list before you've read a forum, and the X Best Review tool pulls the products' current specs, prices and review patterns from Amazon's data, which covers the tiers and the complaints without the manual read.
What the two hours still add is judgement: which specs decide, which brands buyers trust, which words they use, and those go into the brief as the deciding specs, the brand list and the vocabulary. The tool writes the method section honestly by default, which is the right line for a category you learned this morning.
FAQ
How do I research a product category I know nothing about?
Two hours, in a fixed order: Amazon's category page and its filters for the price tiers, the brands and the specs the store itself sorts by; the top three ranking guides' headings for the questions buyers ask; the one-star reviews on the three bestsellers for what goes wrong; one enthusiast forum or subreddit thread for the vocabulary and the opinions; and a manufacturer's spec sheet to learn what the numbers mean. At the end you know the reader types, the deciding specs, the tiers, the brands, the complaints and the words, which is what a brief needs.
Can you write a buyer guide without expertise in the category?
You can write an honest one that compares current products on specs, review patterns and price for a stated reader, as long as it says that's what it did and doesn't claim hands-on testing or expert judgement it doesn't have. What you can't do without the two hours is know which specs matter, which brands are respected and which complaints are common, and a guide written without that reads as a reshuffled listing. The research doesn't make you an expert; it makes you an informed comparer, which is what the method section should say.
What should I know before briefing an AI writer on a product?
Six things: who buys this and for what (the reader types), which three to five specs separate the products (the deciding specs), where the price steps are and what each buys (the tiers), which brands buyers trust and which they avoid (the brands), what owners complain about (the complaints), and the words buyers use for things (the vocabulary). A brief with those six produces a guide for that category; a brief with just the keyword produces a guide for any category with the nouns swapped.
How long does it take to research a niche for a buyer guide?
About two hours for the first guide in a category, and twenty minutes for each guide after that, because the reader types, tiers, brands and vocabulary carry over and only the deciding specs and complaints change per sub-category. The first two hours are the expensive part and the reason to write several guides in a category once you've paid it, rather than one guide each in ten categories.
AffSERP pulls current specs, prices and review patterns from Amazon; the two hours give you the deciding specs, brands and vocabulary to put in the brief, and the method line stays honest by default.
Start with free words