How to Do an SEO Competitor Analysis: Learn From Your Rivals and Strengthen Your Site

What is an SEO competitor analysis and how do you do one?

An SEO competitor analysis is the process of studying the sites that outrank you for your target keywords and working out which content, structure and technical advantages put them there. You identify your real search competitors, measure the keyword, content, technical and link gaps, and turn those gaps into a prioritised task list.

That sounds simple. In practice, though, most analyses I review are either a screenshot of a tool export or a plan to copy a rival's website. Neither works. I have been doing this work for clients since 2012, and I have learned that a good analysis answers one question: what does Google reward for this query, and why are we not there?

In this guide I walk you through how I answer that question, step by step. I focus on method rather than on tools, because tools change while the way of looking at a results page stays useful.

Why is your business competitor not always your search competitor?

The first thing I ask a new client is who their competitors are. The answer is almost always two or three well known firms in the same industry. However, the sites that compete with you for the same query on Google are often not those firms at all. You may face a directory, a marketplace, a forum or a small blog you have never heard of.

For example, think of a manufacturer of commercial kitchen equipment. In sales, its rivals are other manufacturers. On the results page for "commercial dishwasher prices", though, marketplaces and comparison articles may fill the first page. Those sites do not look like they steal customers, but they take the clicks.

So I build the analysis in two layers:

• Business competitors: firms that sell the same product to the same buyer. They matter for positioning and offer comparison.
• Search competitors: every domain that shares the first page with you for your target queries. They are the real input for SEO decisions.
• The grey zone: sites that both sell and publish content. These are usually the most dangerous rivals.

In short, you start the analysis with a list of queries, not a list of companies. Your competitor is whoever appears on the results page for those queries.

How do you find your real search competitors?

I start with a list of 20 to 50 target queries. I pull them from existing Search Console queries, from phrases the sales team hears every week and from keyword research focused on sales. The list should reflect the topics that make you money. Generic traffic terms alone will mislead you, because they rarely turn into customers.

Next, I search each query in a private window with the target country and language set. I write down the domain of every one of the first ten organic results. Doing this by hand feels slow. Still, seeing the results page with your own eyes during the first round is worth the time.

Then I count the domains. The three to five domains that keep showing up across a large share of your queries are your real search competitors. A huge news site that appears for a single query is not a competitor; it is noise.

The count also tells you something else: the page type Google prefers. If product pages rank for half of your list and guides rank for the other half, you are fighting two battles against two different groups. Making that split early simplifies every step that follows.

What data and tools do you need for an SEO competitor analysis?

You do not need an expensive tool for an SEO competitor analysis, but you do need to know the right data sources. Free sources give you a solid first round. Paid tools, however, add scale and save time.

These are the core sources I use:

• Google Search Console: your own query, impression and average position data. You cannot see competitors here, but it gives you your side of the comparison.
• Google Keyword Planner: according to the Google Ads help page, the "start with a website" option lets you enter any site and get keyword ideas related to its content.
• PageSpeed Insights: shows lab and field performance for a competitor page.
• Rich Results Test: reads the structured data on any URL, so you can see which schema types a rival uses.
• Third party SEO tools: show a competitor's estimated organic keywords and backlinks. These are estimates, so do not treat them as exact figures.

There is one more source that matters as much as any tool: your own eyes. Reading a competitor page the way a customer would gives you insight no tool provides. That is why I read at least five competitor pages from top to bottom in every analysis.

How many competitors should you analyse?

For a deep review, three to five search competitors are enough for most businesses. That range is a starting point from my field experience, not a fixed rule. In other words, add more competitors and the analysis grows wider, but your decisions get blurrier.

When I choose, I look at three things. How many of the target queries does the site appear for? Does it serve the same search intent as you? Is it within reach in terms of scale? Reach is an important filter. For instance, competing head on with a national marketplace is not a realistic first goal for a young business.

Therefore, I usually pick one "leader", two "similar size" rivals and one "fast riser". First, the leader shows you the ceiling. Similar size rivals show you the sites you can overtake in the near term. Finally, the fast riser gives you the freshest clue about what Google has started to reward.

On the other hand, if you run very different product lines, each line needs its own competitor set. Reading a whole site through one list leads to wrong conclusions, especially in e-commerce.

In practice I open a separate tab for each product line. Each tab holds that line's queries, competitors and notes. As a result, you can see clearly where you are strong and where you are weak, and you can move resources to the line with the biggest opportunity.

How do you extract the keywords your competitors rank for?

There are two ways to see a competitor's keywords: tool estimates and your own observation. Third party tools list the terms a domain probably ranks for. Keyword Planner generates ideas from the content of a competitor URL. So I use both together.

That said, I never take the raw list as it is. First, I split it into four groups:

1. Shared keywords: terms where both you and the competitor rank. Here you look at the position gap.
2. Gap keywords: terms where the competitor ranks and you do not appear at all. This list is usually the most valuable.
3. Advantage keywords: terms where you lead. You need to protect these.
4. Irrelevant keywords: terms that help the competitor but do not match your offer. You drop them.

Next, I tag the gap keywords by intent: people looking for information, people comparing options and people ready to buy. This shows you which term needs a new page and which one needs a new section on an existing page. I cover that assignment step in detail in my keyword mapping guide. Splitting one keyword across two pages causes cannibalisation.

How do you run a content gap analysis?

A content gap is a user question that a competitor answers and your site does not. The difference from a keyword gap is simple: a keyword describes a phrase, while a content gap describes a need. Specifically, people can search for one need with dozens of different phrases.

In practice I break the competitors' blogs, guides, category pages and FAQ sections into topics. Then I look for the matching page on my client's site for each topic. If there is no match, the topic goes on the gap list. If the match is weak, it goes on the improvement list instead.

For example, at a software company you may find that competitors have integration pages, pricing comparisons and setup guides, while your site only has product descriptions. In that case, then, the problem is not a keyword. It is a missing content type.

One warning here, because this is where many teams go wrong. Finding a gap does not mean rewriting the competitor's page. Google's guidance on helpful content asks whether content that draws on other sources avoids simply copying or rewriting them and instead adds substantial value and originality. In other words, you need to bring your own data, experience and examples. Otherwise you have added one more page to the pile without making a difference.

What do search features on the results page tell you?

A results page is not just ten blue links anymore. Featured snippets, the "People also ask" box, video carousels, the local pack, shopping results and AI summaries all share the organic clicks. For that reason, I note which features appear for every target query.

This observation tells you two things. First, click potential: if a large local pack and an ad block sit at the top, even the first organic position may not bring the traffic you expect. Second, format: if Google shows videos for a query, users probably want a visual explanation.

I also check who holds each feature. For example, if a featured snippet comes from a competitor's table, study that page's structure. Looking at which sources AI summaries cite is now part of the job as well; my article on writing content for AI Overviews covers that side.

Finally, I check how your own title and description look next to your rivals. The Google SERP preview tool speeds up that comparison. If every competitor leads with price, a different message may help you win the click.

How do you judge the content quality of competitor pages?

Measuring content quality by word count is the most common mistake I see. A competitor page with 3,000 words does not mean you need 3,500. Instead, the real question is how completely and reliably that page answers the user's question.

When I read competitor pages, I use a checklist:

• Does it give a direct answer on the first screen, or does it open with a long introduction?
• Does it contain original data, photos, calculations or worked examples?
• Is it clear who wrote it and why they can speak on the topic?
• Can you scan it through tables, lists and subheadings?
• Is it current, or does outdated information still sit on the page?
• Does it guide the reader to a natural next step at the end?

I write short notes against each point for every competitor. This way, instead of a vague verdict like "their page is better", you get a concrete finding such as "they show the price range in a table and we do not". As a result, a concrete finding turns straight into a task. Moreover, these notes make it far easier to brief a writer.

What should you compare on the technical side?

A technical comparison relies on what you can observe from the outside, without access to the competitor's systems. The aim is not to audit their stack. Instead, it is to see whether your own site has a difference that holds its rankings back.

The main points I check are indexability, canonical tags, the mobile view, structured data, page speed and how clean the URL structure is. For structured data, you enter a competitor URL into the Rich Results Test and see which types it uses. If rivals mark up their product pages with Product and Review data, you can prepare suitable types with a schema generator.

Even so, do not overrate technical differences. If both sites meet the basic technical requirements, content and authority usually decide the ranking. What I really look for in the technical review are the problems a site creates for itself: pages that carry a noindex tag by mistake, redirect chains, or main content that only loads late through JavaScript.

Put simply, looking at a competitor helps you notice your own weak spots. You do not copy a rival just because they do something well. But if you have a basic problem, the comparison makes it visible.

How do you compare competitors with Core Web Vitals data?

When you compare speed, keep lab scores and field data apart. According to the PageSpeed Insights documentation, the tool reports real users' LCP, INP and CLS experience over the previous 28 day collection period. The "good" thresholds are 2.5 seconds for LCP, 200 milliseconds for INP and 0.1 for CLS.

This field data comes from the Chrome User Experience Report. Also, it only appears for pages or origins with enough visitors. So a small competitor may show no field data at all. In that case you only have the lab test, and you should not read it as a definitive comparison.

To keep the comparison fair, I compare like with like: home page with home page, category with category, product with product. Otherwise, comparing a rival's blog post with your heavy product page tells you nothing.

Also, a speed gap alone rarely explains a ranking. If a competitor is slower than you and still ranks higher, the cause probably lies elsewhere. I explain how I read a speed test step by step in my Google Lighthouse guide, and my article on SEO and UX covers the wider page experience picture.

How do you read differences in site structure and internal linking?

A competitor's menu, category tree and URL pattern show which topics it gives weight to. For example, a topic in the main navigation is strategic for that site. In contrast, a topic that is missing from the menu but covered in depth on the blog usually exists to earn traffic.

In this review I look for answers to a few questions. Did the competitor create a dedicated page for the target query, or does it cover the topic inside a general page? Do category pages include explanatory copy and a clear filter structure? How many clicks does it take to reach important pages from the home page?

For instance, a competitor may have one page per service, with links to related guides under each one. If all your services sit on one long page, you are not offering Google a strong candidate for each topic.

On large sites this difference becomes even more visible. I describe how I build category architecture in my article on category structure for large websites. When you combine the structural findings from your competitor review with that framework, it becomes clear which pages to create and which ones to merge.

How much does a backlink profile comparison matter?

Links are still an important signal, but comparing raw counts misleads you. A competitor with 5,000 links does not mean you need 5,000 links too. After all, a large share of those links may come from directories, automated sites or irrelevant pages.

That is why I look at sources rather than totals. Which industry associations, news sites, suppliers and partners link to the competitor? How many of those sources could link to you naturally? The most valuable finding is usually a shared source. An industry publication that links to two of your rivals but not to you is an opportunity within reach.

I also check which competitor pages attract the most links. In practice, these are often research reports, free tools or in depth guides. This tells you which content type you need to produce if you want to earn links.

Let me be direct on one point: do not try to catch up by buying links. Google's spam policies clearly prohibit link schemes meant to manipulate rankings. As a result, a short term gain can turn into a long term penalty.

How can Google Ads data support organic competitor research?

Paid search data adds a surprising amount to organic research. According to Google Ads Help, the auction insights report for Search campaigns offers six statistics: impression share, overlap rate, outranking share, position above rate, top of page rate and absolute top of page rate. The report shows which advertisers enter the same auctions as you.

So why does this matter for SEO? A competitor that spends ad budget on a term is effectively telling you that the term pays off for them. Therefore, checking which organic pages that advertiser shows up with helps you prioritise queries with high commercial value.

In addition, the Google Ads Transparency Center lets you see the ads an advertiser runs. The benefits your rivals stress in their ad copy can also give you ideas for your organic titles.

Reading paid and organic data together also clarifies how much weight each channel should carry. In my Google Ads management work I always plan this balance alongside SEO. When both channels feed from the same keyword list, you waste less budget.

How is competitor analysis different for local search?

For local businesses, the competitor set differs from classic organic results. For queries like "dentist near me", the businesses in the map pack are your real rivals. What is more, that set changes with the searcher's location; a search in one district can return different businesses from a search a few streets away.

In a local review, these are the points I check:

• The primary and secondary categories chosen on the business profile.
• Review count, rating and how consistently the business replies to reviews.
• Photos, a service list and up to date opening hours on the profile.
• Whether the website has a separate page for each location or service area.
• Whether name, address and phone number stay consistent across platforms.

In this comparison the biggest gap often shows up in category choice and review flow. If a competitor picked the right primary category and you stayed with a generic one, you can fall behind in the map pack no matter how good your content is. I cover local visibility in depth in my Google Maps SEO guide.

How do you turn findings into an action plan?

An SEO competitor analysis stays in a folder unless it becomes a task list. I score every finding on impact, effort and confidence. First, impact is the likely contribution to traffic or leads. Effort is the time and resources the task needs. Finally, confidence reflects how solid the data behind the finding is.

The table below is a simplified version of the comparison template I hand to clients at the end of an analysis. The entries are illustrative; you need to fill them in with your own site's data.

After that, I put every finding on the calendar with an owner and a date. This way the analysis stops being a diagnosis and becomes the roadmap for your SEO consulting work.

What should an SEO competitor analysis report include?

A good SEO competitor analysis report should be short enough for a manager to grasp the big picture in five minutes, and detailed enough for the team to do the work. I prepare it in two layers: a one page summary and appendices.

The summary page covers these points:

• Competitor set: which domains you chose, based on which query list, and why.
• Visibility comparison: how many target queries you and each rival hold on page one.
• Top three gaps: the clearest differences on the keyword, content or technical side.
• Priority task list: tasks for the first 90 days, with owners and dates.
• Assumptions and limits: which figures are estimates and which are exact measurements.

The appendices hold the raw query table, notes on competitor pages, speed measurements and link sources. As a result, the summary stays clean while every claim still has its evidence in the report.

Above all, the last point matters most. An SEO competitor analysis often relies on estimates, and saying so openly builds trust. When you can answer "where did this number come from?" without hesitation, the report stops being a slide deck and becomes a decision tool. I explain which pages of a report deserve attention in my guide on how to read a digital marketing report.

What are the most common SEO competitor analysis mistakes?

Over the years I have seen the same mistakes again and again. These are the ones I run into most often:

1. Picking the wrong competitors: choosing the sales rival and missing the rival on the results page.
2. Treating tool estimates as facts: third party traffic figures are estimates, and nobody outside a company knows its real traffic.
3. Copying: imitating a rival's headings, structure or even text. At best, that makes you the second best copy.
4. Ignoring intent: trying to win a query with a sales page when the competitor ranks with a guide.
5. Doing it once: running the analysis one time and working from the same assumptions for a whole year.
6. Skipping prioritisation: producing a list of a hundred items without deciding what comes first.

What these mistakes share is that they forget the purpose. In short, the goal is not to collect facts about a rival. It is to find which change on your own site will make a difference. Writing "so what will we do?" next to every finding is the simplest way to avoid them.

How often should you repeat a competitor analysis?

I recommend a full analysis once or twice a year and a light check every month. That rhythm is a starting point from my field experience, not a guarantee. In markets where competition shifts quickly, you may need shorter intervals.

In the monthly check I look at page one for the target queries. Has a new domain entered? Also, has a rival climbed fast? Has Google added a new feature to the results page? Also, if you see a sudden drop in average position in Search Console, check the results page first. Very often the cause is not on your site; it is a move a competitor made.

Some events call for a new analysis without waiting for the calendar: a major Google update, a new player entering the market, a redesign of your site or a change in your product range. In those periods, therefore, your old competitor set may no longer hold.

The healthiest approach is to build this tracking into your regular reporting. I describe the metrics I follow in my guide to digital marketing KPIs; competitor visibility makes a good addition to that list.

Conclusion: watch your rivals, but draw your own route

A good competitor analysis does not tell you whom to imitate. Instead, it tells you where to stand apart. You find your real search competitors, measure the keyword and content gaps, read the results page and note the technical differences. Then you turn all of it into a prioritised task list.

If this is your first time, start small: 20 queries, three competitors and one table. After the first round you will see which information actually changed your decisions, and next time you can focus on that alone.

If you would like to run this analysis for your own site together, you can reach me through the SEO consulting page. We start by discussing where you stand and which queries matter, and then we define your competitor set together.