How to analyse your competitors in Google and AI answers

Competitor analysis in search and AI answers means examining the sites that lead Google's results and get chosen as sources inside AI engines' answers to your customers' questions, then comparing their pages with yours to find the gaps that keep you from leading. Your business needs it now because working on your site in isolation from what the buyer actually sees in the market means spending your time and budget on general pages, while a competing site wins the customers who are ready to buy, simply because it answered the questions they ask before paying.

Many business owners and marketing managers start by improving their sites from the inside only. They redesign the home page or write introductory articles they assume the reader cares about, and then wonder why buyer visits from search do not grow. The plain reason is that a search engine and an AI engine do not judge your page in a vacuum. For every question they weigh your page against the other pages published in your field. If the competing page answers the question in its first sentence and gives the details, conditions and documented figures the buyer is looking for, the engine picks it and passes over yours, even when your service is better and older in the real world.

Modern competitor analysis combines two paths that are no longer separate in the buyer's journey. The first is competitor analysis in traditional SEO, which focuses on keywords and the usual results pages, as explained in What is SEO? A guide to search engine optimisation. The second is competitor analysis in optimising for AI answers (GEO), which is the work on your site's content, data and presence so that AI engines choose it as a source and cite it when a customer asks a full question. When you combine the two in one study of your competitors, you stop guessing. You know which question to start with, which page on your site needs an update now, and which new topic deserves a page of its own.

Here is an example. Picture a furniture store in Riyadh whose site shows photos of bedrooms with one short line of description, while another store in Jeddah has detailed pages that answer buyers' questions about comparing types of wood, how long made-to-order work and delivery to Riyadh take, and the warranty and exchange terms. When a buyer asks Google or an AI engine for the best store that makes bedrooms to order with a warranty and delivery to Riyadh, the second store leads the results and the answers, because its pages gave the engine and the reader the information that settles a purchase. Competitor analysis shows the first store owner this concrete reason, so they put their effort into completing the details of their product pages instead of publishing general articles about colour coordination.

Your real competitors in Google and AI answers are the sites that lead the search results page and that the engines cite for your customers' buying questions. Very often they differ from the competitors you know in the street or in your own neighbourhood. The buyer trusts the source that gives the clearest answer at the moment of searching, so any site that holds that place on the first page or inside an AI answer competes with you directly for the customer's attention and visit.

The first distinction to make at the start of the analysis is between the street competitor and the search competitor. A street competitor is the shop, office or clinic that sells the same service in your area and whose name your customers mention every day. Its website may be weak or neglected and may not appear in search results at all. A search competitor, on the other hand, is the site that leads search results and AI answers for your customers' questions, even if it is a smaller business, a store in another city that ships to yours, or a comparison platform that does not sell the service itself. If you watch only the familiar names you already know, the real search competitors take your visitors without you noticing.

When a buyer types a question into Google, the engine shows a search results page (SERP). That is the whole page, which includes the AI summary at the top when there is one, followed by the organic links and a section of related questions that other searchers ask. When you examine this page, or read an AI engine's answer to your customers' questions in Saudi Arabia, the UAE or Egypt, you will see that the sites the engines build their answers from, and that compete with you for the reader, fall into four main types.

• Stores and direct service sites sell the same service or product, compete in the same activity and target the same buyer, whether they are in your city or in another city that serves your customers.
• Specialised directories gather lists of businesses and service providers by sector and city, and the engines rely on them to establish the options available in the market.
• Comparison platforms compare prices, features and ratings across several options on one page. Engines often choose them because they sum up the buyer's options in one place.
• Blogs and informational sites publish buying guides, reviews and detailed explanations that answer comparison and choice questions, so they attract the customer in the research stage before the purchase.

You can see everything that is public about a competing site, such as its published pages, what Google shows in search results and AI summaries, and what the engines cite. But nobody outside that site can see its real visit or click numbers. The reason is firm: the data in the site's internal analytics accounts belongs to its owner alone, and any figure that claims to reveal a competitor's traffic is an estimate by a third party, not an official number.

Separating fact from estimate is the most important step in protecting your marketing decisions from being misled or wasting money. Google gives every site owner two free tools to read the data of their own site. The first is Google Search Console, Google's free tool that shows a site owner the data of their own site only in search, including clicks, impressions, click-through rate and average position by query, page, country and device, as described in the Search Console Help documentation for the Performance report. The second is Google Analytics, a tool that analyses the visits to a site and how visitors behave inside your own pages after they arrive. Because this data belongs to the owner of each site, no tool or agency can show you the Search Console or Google Analytics numbers of your competitor. Every figure published about a competing site's traffic is a third party's estimate, so do not build your plan or your budget on estimated visit numbers for competitors.

On the other hand, the real data you can examine yourself, with full confidence, for any competing site is the public data that the reader and the engine actually see. It covers three sides, and they are enough to build a strong competitive plan.

The public pages published on the competing site. You can open its service, product and guide pages, read their headings and paragraphs, and see how it answers customers' questions and which facts, statistics, sources and details it mentions.

What Google search shows for your customer's question. You can search with any word or question and see the leading sites in the results, the related questions the engine shows, the AI summary at the top of the page and the sites it links to.

The answers of AI engines and their sources. You can put the buyer's question to AI answer engines, read the text the engine writes, and see which sites and which specific pages it chooses and cites in its answer.

The table below sums up the clear difference between confirmed real data and the estimates and hidden data when you study competitors.

When you analyse a site that leads search results or AI answers before you, you compare four specific things. These are the buying questions that site leads and you do not, the actual page that won the ranking or the citation, the depth of facts, statistics and documented sources inside that page, and the mentions and links the site has earned from outside itself. Comparing these four moves you from a general impression that the competitor's site is better to a precise list of what your own page lacks in order to overtake it.

The first thing to compare is the question and content gap. By this we mean the questions buyers ask in your market that a competing site leads, while your site has no page that answers them in depth, or mentions them only in passing in one line on a general page. When your competitor dedicates a separate page to a comparison question, or to a question about the terms of a service in a particular city, it wins all the visits for that question because your site offers no alternative. The second thing is to identify the leading page itself. Sites do not lead under their general names. They lead through a specific page that met what the engine asks for, so you always compare page with page, not site with site.

The third thing, and the one with the greatest effect in AI answers and modern search, is the depth of facts, statistics, quotations and documented sources in the page text, along with how clearly the answer opens each paragraph, as explained in the guide on how to write a paragraph an AI engine quotes in its answer. This point explains the finding of the research paper on GEO, "GEO: Generative Engine Optimization", published on arXiv in November 2023 and presented at the KDD conference in 2024. In their experiments, the researchers found that adding quotations, statistics and cited sources to a page's text raised its presence in generative engines' answers by about 30% to 40%, while keyword stuffing added little. This is a result of academic experiments, not a promise for your site, but it shows why comparing the facts, figures and sources on your competitor's cited page with yours is a useful, practical step. It also agrees with the SEO Starter Guide from Google Search Central, updated in December 2025, which states that Google's core recommendation is helpful, reliable content that serves the reader and is kept up to date, while keyword stuffing is against Google's spam policies.

The fourth thing is off-site mentions and links. These are mentions of your business name, or links to your site, on other trustworthy sites, specialised directories and articles in your field. The SEO Starter Guide from Google Search Central confirms that Google finds most new pages through links, and that links add context that helps the engine understand a page's subject and value. So when you see a competing site beating you even though the quality of your content is close, the reason is often that it is mentioned and referred to in sector directories, news sites and trusted platforms where your business has not yet appeared. The guide on how AI engines know your business from outside your site explains how to build that presence the right way.

You can use the competitor analysis table below as a practical template when you examine any competing page that leads before you on an important buying question.

You analyse your competitors in Google search with free sources in five steps. You begin with your site's real data in Google Search Console, then examine Google's results and AI summaries for your customers' questions, then open the leading pages, compare them with yours, and record the differences in a clear working table. This free method gives you a realistic reading of the market, because it relies on what the engine actually shows the buyer, and you do not need to pay monthly subscriptions before you know where your pages stand.

Imagine an accounting office in Cairo that wants to know why other offices beat it in search when people ask about setting up a business and filing tax returns. Instead of searching at random, it can apply these five steps in order, and the same order works for any service sector or online store.

1. Pull your search phrases and pages from your Google Search Console account. Open the Performance report in your site's Google Search Console account and review the queries and pages in your target country. You are looking for the words and questions where your site gets impressions but has a weak average position or a low click-through rate. At the same time, check the page indexing report in Google Search Console to make sure your main pages are indexed and technically sound, because a page that is not indexed does not even enter the competition.
2. Collect your customers' most important buying questions and search for them in Google. Add to the phrases from Search Console the full questions your customers put to your sales team before buying, then search for each question in Google from your customer's country. Watch the whole results page. Note whether the engine shows a Google AI Overview at the top, which sites it put links to inside the summary, which sites lead the first links, and what extra questions appear in the related questions section.
3. Classify the leading sites and identify the pages that compete with you directly. In your list, separate the sites that offer the same service or product as yours from the directories, comparison platforms and informational blogs. For every competing site that appears across more than one question, copy the link of the specific page that led the result, and do not stop at noting the main domain name, because winning and losing happen at the level of the page.
4. Open the leading pages and examine their content and sources. Read the leading page the way the buyer reads it and the way the engine examines it. Check whether it answered the question in the first paragraph, which conditions, details, figures, statistics, quotations and sources it gave, and whether the content is up to date and divided under sub-headings that answer the related questions on the subject. Then search for that site's name in Google to see the directories and outside sites that mention it and add context to its reputation.
5. Record the gaps in your table and turn them into content tasks. Put each buying question on its own row in a competitor analysis table, and note beside it whether you have a page that matches it or whether it is a question gap that needs a new page. The idea behind no page answers the question? Start with a brief helps you organise what the new page must cover before you write it, while for your existing pages you note what they lack in direct answers, facts and sources so you can add them.

Competitor analysis changed in 2026 because the buyer moved from typing scattered words and comparing blue links alone to asking one full question in Google AI Overviews, Google AI Mode, ChatGPT, Gemini and Perplexity, and reading a single combined answer that cites only a few sites. This shift means that watching the ranking of blue links alone no longer protects your share of the market, because a competing site can win the engine's recommendation and citation at the top of the page or inside the conversation, and take the customer before they ever reach the traditional list of links.

To understand how big this change is for your business, look at the interfaces the buyer uses today when searching. There are Google AI Overviews and Google AI Mode, the summaries and answers that artificial intelligence generates directly at the top of Google search, or in the conversational mode built into search, with links to supporting sources. Google shows these summaries today to buyers in the markets of Saudi Arabia, the UAE and Egypt for many business, comparison and purchase questions. Alongside Google search there are the independent AI answer engines, led by ChatGPT, Gemini and Perplexity, which the buyer opens to ask in their own natural language for a comparison between two options, or to recommend the best service for their need in their city. The engine writes a single answer and cites a small number of sites.

In this new environment, the deciding factor is no longer merely being on the first page. It is that your page earns a citation in AI answers, meaning that the AI engine names your site, or places the link of your page as a source on which it builds its answer and which the reader can click. And as buying questions multiply in your sector, a new competitive measure has emerged: share of voice in AI answers. It means your site's share of the AI answers that cite it, compared with the competing sites on the same buying questions. The higher your share of citations on your market's questions, the stronger your brand's presence in front of the buyer at the moment of comparison and decision.

To grasp the practical danger of ignoring this side, picture a home appliances store in Cairo that watches its traditional Google ranking on short words and feels safe. When a buyer in Cairo asks ChatGPT, or reads a Google AI Overview, for a comparison between two washing machines with after-sales service and warranty, the engine builds its answer and cites the page of a competing store, because that page set out a clear comparison table and spelled out the warranty and delivery terms. The first store owner does not see this loss in their usual visit reports, because the customer made their decision and moved to the competing site directly from inside the AI answer.

You find out which site AI engines cite instead of you by putting your customers' buying questions to the different engines from your target country, listing the sites that keep appearing among the sources of the answers, and then examining the page that won the citation. You look at what it offered in direct answers, facts and sources that your page lacked. Generative engines pick their sources according to technical eligibility in the index and the clarity of documented information in the text, so examining the winning page shows you straight away why it was chosen.

Some site owners think that entering AI answers in Google needs secret codes or special technical files that their competitors have. In fact the official rule is clear and public. Google Search Central's documentation, in its page on AI features and your website published in December 2025, states that for a page to be shown as a supporting link in Google AI Overviews or Google AI Mode, it must be indexed and eligible to be shown in ordinary Google search with a snippet, and that there are no extra requirements and no special files. This means that once your page is indexed and eligible to be shown with a snippet, the contest between you and your competing site moves to how useful the content is, how reliable and how fresh it is, and the context of the links that point to it. This rule is consistent with what the GEO research paper presented at the KDD conference in 2024 found in its experiments: adding quotations, statistics and documented sources raised a page's presence in generative engines' answers by about 30% to 40%.

You can start this examination by hand, putting your customers' questions to the engines and reading the sources, as explained in the guides on how to know whether your site leads ChatGPT and Gemini answers and how to know which competitor ChatGPT recommends instead of you, and why. But when you want to follow many answers across several engines over consecutive weeks without wasting time on manual copying, Tasadr is an Arabic platform for competitive analysis in AI search, which accompanies you on a complete journey from analysis to execution until you lead its answers.

On the page Who leads instead of you, and why?, for every question your field's buyers ask and by the chosen country, the platform shows you a full competitive picture that includes the following.

• A ranking of the sites that the engines cite instead of you, with your own site in its real place among them and the Tasadr score out of 100 for your site, whether you had already added those sites as your competitors or the analysis found them in the market's answers.
• Your share of citations against each competing site, week by week, so you can follow whether you are closing in on your competitors after editing your pages or the gap is widening in their favour.
• The most-cited page of each leading competitor, with a clear statement of what that page has that your own lacks, and a practical step suggested for you to carry out on your site.
• A complete profile of each competing site, and the question gap. It shows where that competitor leads, with which page, and beside which other sites, and which questions your competitors lead that you do not yet follow. It also classifies the kinds of sites the engines build their answers from in your sector.

You choose the right competitor analysis tool for your site by four clear criteria: whose data it shows for your site, how far it covers AI answers by country and by engine, how honest it is about not mixing estimates with real numbers, and whether it ends the analysis in a practical step you can carry out. You do not have to subscribe to any paid tool if you are at the start of the road and checking a few questions by hand. But you do need a specialised tool when you want to follow your presence and your competitors regularly and save your team's time.

Many business owners rightly ask whether they need paid tools for competitor analysis, or whether free sources are enough. The rational answer depends on the size of your site and the stage of your business. If you have a small site and want to check a limited number of buying questions at first, your free Google Search Console account, together with direct searching in Google and notes in a comparison table, is entirely enough to begin and to fix the obvious faults in your pages. But when your business grows and you need to watch how answers and citations change every week across different AI engines and in more than one Arab country, manual checking becomes slow and prone to error. At that point a specialised tool saves your team's effort and reveals the gaps early.

When you compare the competitive analysis tools available in the market, apply these four criteria before you decide. You will find a fuller explanation of them in the guide on the best tool for analysing your site's leadership in AI answers.

• It shows your own site's real data from its official source. Make sure it connects the Google Search Console and Google Analytics accounts of your own site, so that it reads your real clicks, impressions and visits accurately, instead of relying on outside guesses about your site's performance.
• It covers AI answers in Arabic, by country and by engine. AI answers and their sources differ between Saudi Arabia, the UAE and Egypt, and between one engine and another, so choose a tool that shows you, for buyers' questions in each country separately, the sites and pages each engine cites.
• It is completely clear about estimates for competing sites. Be wary of building your decisions on tools that show visit or click numbers for competing sites as if they were confirmed facts. Remember always that nobody can see the Search Console or Google Analytics of a competing site, and that a trustworthy tool, when it analyses competitors, concentrates on visible facts that can be shown, such as leading pages, citations, share of voice and page content.
• It ends in a clear practical step. Many tables and charts do not change your place in the answers if they leave you puzzled over the most important question, which is what to do now on your page. So choose a tool that compares your page with the leading site's page, names what it lacks, and suggests a direct step for editing it.

After you finish analysing your competitors, turn the results you have gathered into a practical action plan ordered by priority. The plan begins by making sure your pages are indexed and technically eligible, then updating the existing pages that competitors beat with a direct answer, facts and sources, then building new pages to close the question gap, strengthening the mentions of your business outside your site, and following your share of citations regularly. The analysis does not change your place in the market unless it becomes published text that the engine and the buyer read, so start with the pages closest to your business's buying questions instead of trying to edit the whole site at once.

To arrange your efforts in a systematic way, in line with the idea of what to do to lead: tasks ordered by priority, carry out these five steps in order.

1. Check that your main pages are indexed and eligible to be shown with a snippet. Open the page indexing report in Google Search Console and make sure your service, product and guide pages are indexed and eligible to be shown in ordinary Google search with a clear snippet, and deal with any page that is technically excluded. Google Search Central's documentation confirms that indexing and eligibility to be shown with a snippet are the condition for your page to be among the supporting links in Google AI Overviews and Google AI Mode.
2. Fix the existing pages where a competing site leads before you. Begin with the pages on your site where the analysis showed that a competing page comes before it or wins the citation in its place. At the opening of the page and of each main section, add a paragraph that answers the customer's question in its first sentence, and strengthen the text with the statistics, quotations, documented sources and practical details your page lacked. This rests on what the research experiments in the GEO paper showed, that quotations, statistics and sources raised presence in generative engines' answers by about 30% to 40%. Keep the language natural and free of keyword stuffing, which goes against Google's policies.
3. Create new pages that cover the buying question gap. Take the list of questions your competitors lead and for which you have no matching page on your site, order them by how close they are to a purchase decision, and then write for each question a complete, useful page that answers the main question and its related sub-questions clearly, and is kept up to date, as Google Search Central's guide recommends.
4. Build a presence and mentions for your site on trusted sources outside your own domain. Review the sites, specialised directories and articles that mention your competitors and that the engines rely on when they build their answers. Work to get your business introduced there and to earn natural links and mentions that add context to your site and help search engines and AI engines discover your pages and trust them.
5. Follow the change in your site and your share of citations week by week. After you publish the edits and the new pages, give the engines time to re-read and re-index the pages, then check your buying questions again to see how your site has changed against the competitors. You can do this follow-up by hand in the same table, or follow your share of citations against each competitor week by week through Tasadr, to see which pages have moved up and what step is suggested for each page.