Competitor Content Analysis: Find SERP Rivals and Compare

To analyze competitor content, list the pages that actually appear for your target queries, read each one as a visitor would, and record how it handles depth, originality, evidence and format. Then decide, feature by feature, whether a rival's choice helps the visitor or only fills space. The rivals that matter are the ones sharing your search results, and they may not be the companies you compete with for customers. Every step here uses public pages and your own notes. The examples follow an online language school that wants its "learn Spanish online" page to hold up against what searchers see.

Your rivals are decided query by query

An online language school may think of other language schools as its competition. For the query "learn Spanish online", though, the first results page could also contain a course directory, a learner forum thread and a public library resource page. Those are the pages a visitor compares your page with, whether or not they sell lessons.

Three kinds of rival

A rival can change from one query to the next

A large school may share the results for "learn Spanish online" and vanish from "Spanish subjunctive exercises". Record the rival as a pair, a query and a URL, instead of as a company name. Otherwise the table says who is big, not who is relevant.

Site strength is not page strength

Google's guide to ranking systems (Opens in a new tab) says the systems are designed to work at the page level, using a variety of signals and systems. It adds that having some good site-wide signals does not mean that every piece of content from a site ranks highly. A well-known domain therefore does not make each of its pages a benchmark. Compare pages, one at a time.

Building the rival list from your own search terms

Start with a list of queries that describe what the page is for. Keep the set small and coherent, so the rival list reflects a single visitor task.

Choosing five to eight queries

Take the queries from your own Search Console export for the page, plus a few from the questions learners ask your support team. Remove queries that contain your school name, because those results describe people who already know you. Remove one-word queries too, since their results mix several purposes.

Recording the first page

For each query, write down the organic results on the first page: the URL, the page type and the title. Results can differ by person and setting, so practical suggestion: check the query in two separate browser sessions and give extra weight to the pages that appear in both. Date the record. A rival list is a snapshot, not a fact.

The overlap grid

Put queries in rows and sites in columns. Each cell holds the page type or stays empty. Example scenario: a language school records five queries.

Reading the grid

Site C turns up for four of five queries, as a guide in three of the four, so it is the closest thing to a main search rival for the topic and deserves a careful read. Site A shows up with course pages, a business rival for commercial queries. Site D appears with forum threads, which tells you what learners say in their own words. The grid decides how long you spend on each site.

Setting up the comparison sheet

One sheet holds everything. Rows are rival pages, columns are the things you will compare. Write short observations instead of scores, such as "free level test, no explanation of results". A sentence tells the next reader what you saw. A number only suggests precision.

Reading a rival page as a visitor first, then as an editor

Read the page twice. The first pass is the visitor's: you have the query in mind and you want an answer. Note where you first found it. The second pass is the editor's: you look at structure, sources and choices.

First pass: how long until the answer

Time yourself. If the page answers within the first screen, mark it. If the answer sits behind a long introduction or a sign-up form, mark that too. This single note explains more about a page's value than its length does.

Second pass: structure

Copy the heading outline into your sheet in your own words, listing each section's purpose rather than its title. Look at the tables and ask what they compare. Check which pages the rival links to and which pages link back. A guide that links onward to a course page shows a path the school wants visitors to take. The outline is a map of someone else's decisions, so use it for contrast and not as a template.

Who, how and why

Google's page on helpful, reliable, people-first content (Opens in a new tab) poses three questions. Is it self-evident to visitors who authored the content? Is automation or AI use self-evident to visitors through disclosures or in other ways? Why does the content exist: the page says it should be created primarily to help people who visit the site directly, not mainly to attract search engine visits. Add three short lines for these to your sheet, and answer them for your own page as well.

Four lenses: depth, originality, evidence and format

The same Google page offers self-assessment questions that map onto these lenses. They are the basis for the lenses; the scoring approach is a working method of your own.

Depth

The page asks whether, after reading, someone will leave feeling they have learned enough about a topic to help achieve their goal. For the Spanish page, a rival that explains levels, weekly time commitment and what a first lesson looks like is deeper than a rival that lists prices.

Originality

Google's questions include whether the content provides original information, reporting, research or analysis, and whether it offers insightful analysis or interesting information beyond the obvious. A rival that publishes its own placement quiz or its own sample lesson is original. A rival that paraphrases a textbook syllabus is not. Write "paraphrased syllabus, no original contribution" when you see it.

Evidence

The page asks whether content clearly demonstrates first-hand expertise and a depth of knowledge, and whether it is written or reviewed by an expert or enthusiast who demonstrably knows the topic well. A rival claiming "fluent in three months" should show a method, a learner's path or a teacher's credentials. Record what support the claim has and where the support is missing.

Format

Format is how the information reaches the visitor. A comparison of study plans is easier to use as a table than as four paragraphs. When the first results page shows videos or images for the query, that is a sign the searcher expects those formats too.

Weighting the lenses by query

Practical suggestion: lean on evidence and depth for decision queries, on format for how-to queries, on originality where the results are crowded with similar pages.

Example scenario: the "learn Spanish online" page

Example scenario: the school compares its page with three results. The table holds observations only.

What the table says

The school page sits next to Rival A. It lacks the level guidance and study plan that Rival B offers, and it has no learner voice like Rival C. None of these rows says to copy a rival. They say which parts of the searcher's question go unanswered on the school's page.

From finding to task

Write one task per finding. "No level guidance" becomes a short section on how a placement works. "No teacher information" becomes a staff block with qualifications the school can document. "No learner voice" becomes a request to students for permission to quote their feedback. Add an owner and a date to each.

Separating what rivals have from what is actually useful

A feature on a rival page is not proof it helps the visitor. Some pages are long to look thorough, and some are split into several sections to show more advertising. Copying every feature would inflate your page in the same way.

Decision rule

Decision rule: add a feature only when all three answers are yes. If one is missing, record the reason in the sheet. Closing a content gap means answering a question, not mirroring a competitor's layout. For a wider method on finding missing topics, see content gap analysis.

Staying clear of copying

Do not lift a sentence, table or image from a rival page. Answer the same question with your own expertise, examples and evidence. When two pages cover one topic, the difference comes from sources, experience and presentation, not from word overlap. A scoring table built on Google's helpfulness questions can help you test your own draft; see the article quality scoring table.

Public pages only: limits of competitor research

Limit the work to what a visitor can open in a browser. Do not try to reach pages behind a login, a membership wall or accounts that are not yours.

Reading method

Open pages in the browser, note headings, tables and links, and keep the sample small. Do not run bulk or automated collection that a site's terms of use prohibit. If the terms restrict automated access, read by hand.

Notes without pasting

Write observations in your own sentences. Keep the URL and the date you looked. If the page changes or disappears, your note still describes what you saw.

Data you may legitimately use

Your own Search Console data and public pages are the two sources here. Another site's Search Console, server logs or private dashboards are out of bounds. If you use estimates from a third-party tool, label them as estimates in the sheet.

What ranking systems mean for your reading

The ranking systems guide describes several systems that shape a result, and a few of them change how you read a rival page.

Meaning of the query

The guide says BERT helps Google understand how combinations of words express different meanings and intent, and that RankBrain can return relevant content even if it does not contain all the exact words used in a search. A rival page can therefore be relevant without repeating your query. Judge the answer, not the vocabulary.

Passages

Passage ranking looks at sections of a page to understand how relevant the page is to a search. When one section of a rival page answers your query well, study that section and what it does, not the whole page.

Reviews and freshness

The guide also describes the reviews system, which aims to reward high quality reviews, and freshness systems that show fresher content for queries where it would be expected. If your query is a product comparison and a rival page reads like a review, add a column for original testing. If freshness matters for the query, note each rival's update information.

Quality signals without a ranking promise

The guide says reliable-information systems help surface more authoritative pages and demote low-quality content. The documentation has no statement on how much each signal weighs. Record quality markers such as authorship, sources and update dates as trust signals for visitors, not as ranking explanations.

How the page type changes what you read

Not every rival page deserves the same kind of reading. A directory entry, a guide, a course page and a forum thread each carry value in a different place, so the checklist you run should change with the type.

Reading a course page

Ask what a visitor could decide on the spot. Look for level descriptions, lesson length, who teaches, what a trial involves and what happens after sign-up. If the page hides these behind a form, write that down. A course page that answers fewer questions than yours is a business rival, not a content benchmark.

Reading a forum thread

A forum thread is not a benchmark for presentation, but it is a record of the questions learners ask in their own words. Copy no one's post. Instead, list the recurring questions in your sheet and check whether your page answers each of them. Unanswered ones are gap candidates.

Reading a directory

Look at the fields a directory compares: price, level, teacher type, time zone. Each field is a decision criterion that the directory's authors believed matters. You do not need all of them, but a field your page lacks and your visitors care about belongs on the gap list.

Reporting the findings to the person who decides

The comparison sheet is a working document. The person approving the work needs a page, not a spreadsheet. Give them three columns: what we found, why it matters to the visitor, and what we will do.

Keep it to one page

A decision maker will not read the sheet, but may want to see which observation each recommendation rests on. End the summary with a link to the sheet and a date.

Say what you did not check

List the queries you did not cover and the rival pages you skipped. A summary that admits its limits is easier to trust, and it tells the next person where to look.

A two hour competitor analysis workflow

Practical suggestion: the times are approximate and vary by job. Run the work in four blocks.

Checklist

• Queries describe the page's real job and exclude brand terms.
• Each query's first page was recorded with a date.
• Rivals are stored as query and URL pairs.
• Every cell in the sheet holds an observation sentence.
• Your own page appears as a row, filled in first.
• Each rival feature passed the three-question test.
• No text, table or image was copied from a rival.
• Each finding has an owner and a due date.

Keeping the list short

Reading ten pages per query stretches two hours into a week. Pick three to five pages that appear for more than one query and represent different page types. Write one line explaining why each is in the sheet, and mark the others "not read".

Turning findings into a plan and tracking it

Three lists

Keep additions to the existing page, ideas for new pages and deliberate non-actions apart. They move at different speeds and answer to different owners. Record non-actions too, so a later colleague sees why the school did not add a feature a rival has.

Prioritizing

Score each finding with two questions: how much will it help the visitor, and how easy is it to ship. High help and easy work goes first. High help and hard work needs its own plan. Low help drops off the list.

Follow-up

After a task ships, compare the page's queries and impressions in Search Console with the earlier period. The comparison does not prove the change caused any movement, because demand and rival pages shift too. It does show whether the page has started to appear for new queries. Re-run the sheet when results change visibly.

Mistakes in rival analysis

Starting from a list of famous brands

A list built from industry names skips the results page the visitor sees. Build the list from queries.

Reading only the top result

The first result is not automatically the best answer. Read the whole first page and choose the page that answers best, using the sheet as evidence.

Treating every rival feature as an advantage

Long intros, repeated sections and heavy advertising on a rival page are not strengths. Mark them as examples to avoid.

Expecting a ranking from a finished sheet

Adding a section that a rival lacks does not promise a higher position. The sheet exists to improve your answer. You measure positions afterward.

Frequently Asked Questions

What is the difference between a search rival and a business rival?

A business rival sells what you sell. A search rival appears for the same query, and it can be a directory, a forum or a guide site that sells nothing.

How large should the query set be for one page?

Practical suggestion: five to eight related queries give an overlap grid with enough variety to show which rivals recur.

Can I use a rival's page as a template?

Do not copy a rival's text, table or image. Answer the same question with your own expertise and evidence. The difference between two pages on one topic comes from sources, experience and presentation.

Does a large site make every one of its pages a benchmark?

No. Google's ranking systems guide says the systems are designed to work at the page level and that some good site-wide signals do not mean that every piece of content from a site ranks highly. Compare pages.

Is automated scraping part of rival analysis?

The documentation has no statement on it. Practical suggestion: open a few pages by hand, and if a site's terms of use restrict automated access, do not go around the restriction.

When should I rebuild the comparison sheet?

Practical suggestion: rebuild it when you change the target page or when the first results page for the query changes visibly.

Conclusion

Competitor content analysis starts from the results page, not from a list of companies. You find the pages that share your queries, read them as a visitor and then as an editor, compare depth, originality, evidence and format, and test every rival feature against the visitor's need, your ability to produce it and the page's single job. The work stays within public pages and your own words, and it makes no promise about rankings.

When the comparison shows that a page also needs stronger external references, you can look at the backlink packages.