Competitor Buyer Guides: How to Read the Page That Outranks You and Write the One That Beats It
The page at position one for your keyword is the most useful document you'll read before writing your own, and most writers read it wrong: they skim it, note that it's long, and set out to write something longer.
What the page tells you is what the query needs, which reader it's for, how many products the buyer expects, which questions have to be answered, and what the page did to earn its verdict. That's the brief for yours. The rest of it, the sentences and the picks and the order, is the part you don't take.
This post covers a ten-minute read of a position-one guide and the ten lines to write down, how those lines become the brief for your page, what not to copy and why copying it loses, the one thing your page needs that theirs can't claim, and the "different enough" check before you publish.
The sitemap post covers finding which keywords a competitor targets; this is what to do once you've picked one and read their page.
The short answer
Open the top three results in a private window. For each, write down ten lines: who it's for, what it answers first, the H2s in order, product count, the picks, the table's columns, the FAQ questions, the date, its stated method, and how far the first button is.
Everything the best of them does well is your floor. Then find the thing none of them does (a reader, a question, a pick, a verdict, fresher data) and make it your page's first screen. Don't take their picks, their order or their sentences.
The ten-minute read
| Line | What to write down | What it tells you |
|---|---|---|
| Reader | Who the intro addresses: a beginner, a commuter, a parent, a pro | The reader Google thinks the query belongs to |
| First answer | What's in the first screen: a verdict, a list, a definition | What the buyer wants before scrolling |
| H2s in order | The section list, copied as a list of topics, not words | The questions the query needs answered |
| Product count | How many picks | The depth the buyer expects; ten, not three, not thirty |
| The picks | The products, and which is the top pick | The consensus; what you must at least consider |
| Table columns | What the comparison compares | The specs the buyer decides on |
| FAQ questions | The questions, as topics | The long-tail queries the page catches |
| Date and freshness | Published, updated, how old the picks are | Whether fresher data alone can beat it |
| Stated method | "We tested", "we compared specs", nothing | What claim you'll be measured against |
| Distance to the button | How many screens to the first buy link | How the page converts, and where yours can be faster |
Ten lines, three pages, half an hour. Where the three agree, the query has a shape and your page takes it. Where they differ, there's room.
From the sheet to the brief
The sheet becomes the brief almost line for line. The reader line becomes the reader in your brief. The H2 list, merged across the three pages, becomes the question list your page must answer, in the order the buyer asks them. The product count sets yours.
The table columns become the specs your picks are described by. The FAQ topics become candidate FAQ questions, reworded. The prompting post covers the six inputs a brief needs; the competitor sheet fills four of them before you've thought about your own angle.
The picks line is used differently: it's the list your research has to consider and, mostly, confirm or reject on your own evidence, not the list your page publishes. If your research lands on the same top pick, say why in your own terms. If it lands elsewhere, that's a page with something to say.
What not to copy
- Their picks in their order. The same ten products ranked the same way is a copy, however the sentences differ. Research the category; arrive at your own list, which will overlap theirs by half and disagree on the rest.
- Their sentences. Obviously, but also their paraphrased sentences: a section that makes the same three points in the same order with different words is a rewrite, and the duplicate content post covers why that pattern ranks below the original.
- Their table's rows. The columns are the genre; the rows are their research.
- Their FAQ word for word. The topics are fair; the questions as phrased are their long-tail targets, and the exact phrasing just competes with them for the same snippet.
- Their method claim. "We tested all twelve" is theirs to make and yours to make only if true. A page can't borrow a method; it can state its own.
- Their mistakes. A discontinued pick, a wrong spec, a 2023 price: the sheet should note them, because each is a place your page is right where theirs is wrong.
The one thing theirs can't claim
Matching the floor gets you a page as good as theirs, which ranks below theirs, because theirs has the links and the age. The page that beats it has one thing the sheet showed was missing from all three. Usually one of these:
- A reader they ignore. Three guides for enthusiasts and none for the person buying their first one, or the reverse.
- A question they dodge. The H2 the buyer needs that none of the three has, often "which one should I skip" or "is the expensive one worth it".
- A pick they missed, or a pick they all kept that's been discontinued.
- A verdict they hedge. Three pages that end "it depends" lose to one that says which, for whom.
- Fresher data. Prices, availability and models a year newer than theirs, stated with the date.
Whichever it is goes in the first screen, where the buyer sees it before they'd see the sameness.
The different-enough check
Before publishing, put your draft beside the position-one page and answer four questions. Would a reader who'd read theirs learn anything from yours? Is your top pick yours, with a reason in your words? Is there a section they don't have?
Could someone tell, from the first screen alone, which site wrote it? Two noes means the page is a rewrite; it will rank below theirs and it's worth another hour. Four yeses is a page that can take the position.
Doing it with AffSERP
The sheet is yours to make; what the tools do is the page it leads to. AffSERP's Competitor Sitemap Explorer finds the keywords a competitor targets so you can pick the one to go after.
The X Best Review tool writes the guide from Amazon's current product data, so the picks are selected on specs, reviews and availability today, not copied from the page you read, and the fresher-data edge comes built in.
The brief you give it carries the reader, the question list and the product count from the sheet; the one thing theirs can't claim is a sentence in that brief, and the first screen the tool writes is where it lands.
FAQ
How do I analyze a top ranking page?
Read it as a buyer, then as an editor, in ten minutes. As a buyer: what did it answer first, which pick did it push, how fast did I reach a button. As an editor: the H2 list in order, the product count, the picks, the comparison table's columns, which questions the FAQ answers, the word count, the date, and what it says it did to earn the verdict. Write each down in a line. The sheet of ten lines is the analysis; everything it did well is now the floor for your page.
How do I outrank a competitor article?
Match what the page does well and add what it doesn't do: a reader it ignores, a question it dodges, a pick it missed, a verdict it hedges, a table it doesn't have, data a year fresher. Then be clearer in the first screen than it is. You don't beat a position-one page by being longer; you beat it by answering the same query better for the same reader, with one thing it can't claim. Links and site authority decide ties, so on a new site pick the competitor pages that are weakest, not the biggest keyword.
Is it OK to copy a competitor's structure?
The structure of a buyer guide (intro, picks, table, how-to-choose, FAQ) is a genre, not their property, and every guide at position one shares it. What you don't copy is their picks in their order, their sentences, their table's exact rows, their FAQ questions word for word, or the claim they made about testing. A page that follows theirs section for section with the same picks reads as a copy to Google and to the reader, and ranks below the original.
What is competitor content analysis for affiliate sites?
Reading the pages that rank for a keyword you want, to learn what the query needs (the reader, the questions, the product count, the depth) before writing your own. On an affiliate site it has a narrower purpose than on a content site: the pages that rank tell you what a buyer at that query expects to see before they click a button. The output is a brief, not a report.
AffSERP's X Best Review selects products on current specs, reviews and availability, so the guide you brief from the competitor sheet starts with the fresher-data edge built in.
Start with free wordsKeep reading
