Someone with a real question may type a few words into a search engine and choose among the results. Your page has to be reachable, understandable and genuinely useful when they arrive. Search engine optimisation (SEO) is the work of helping search engines understand a page and helping people decide whether it answers their question. It is a continuing part of making a useful website, not a purchase of a guaranteed position. Google describes it in essentially those two parts and explicitly says that no secret can automatically rank a page first (Google Search Central, 2025c).
This chapter shows you how to make one real page clearer. You will identify the person's question, check the page's answer and destination, write a useful title and description, link it from a relevant page, and record what you can verify after publication. The worksheet is blank: no fictional company, invented search volume or promised traffic. The ten diagrams use the same visual language as previous chapters, with a text explanation after each one.
Figure 1
Figure 1 in words: Situation records a customer question and the existing page. Objectives name the useful outcome. Strategy chooses which question to answer for whom. Tactics and Action include the page, its title, links and technical access. Control checks what the search engine could see and what people actually did. In a RACE view, discovery can help Reach, but the page still needs to support Act and later steps. No framework supplies a ranking guarantee.
Search, organic results and paid ads
A search query is the words or question a person enters. A search results page displays possible answers, including ordinary web results and sometimes other formats. An organic result is a listing selected by a search engine rather than a paid placement for that particular click. Paid search uses an advertising system and budget to show an ad under its own rules; Chapter 17 will cover its planning. A search engine may show ads and organic results together, but paying for an ad does not buy a higher organic ranking. Google says it does not accept payment to crawl more often or to rank a page higher in its organic results (Google Search Central, 2025b).
SEO is sometimes described as “free traffic”. That phrase hides the cost of researching, writing, maintaining, testing and improving pages. A result may also appear without anyone clicking it. A better first objective is help the right person reach a genuinely useful page. This connects to Chapter 14's budget: the hours and technical work need an owner. It connects to Chapter 15's channel choice: search may fit a question someone actively asks even when a social platform is not the best first route.
The word rank describes where or how prominently a result appears for a given search situation. There is no single permanent rank for a page. The query, location, language, device, search features and changes in the results can all matter; Google says relevance and quality are determined programmatically and the returned result can vary with those contexts (Google Search Central, 2025b). Do not use a promise of “number one” to set a budget or judge a supplier.
First understand the three stages
A page does not leap straight from publication to a visible result. Google explains three stages: crawling, in which an automated program discovers and fetches pages; indexing, in which it analyses content and may store information about the page; and serving, in which it selects results for a particular query. Not every discovered page is crawled; not every crawled page is indexed; and an indexed page is not guaranteed to appear for every query (Google Search Central, 2025b). Other engines have their own systems and guidance, so this is a beginner model drawn from one search provider.
Discovery can happen through a link from another known page or through a sitemap, a list of page addresses supplied by the site. A crawler is the automated visitor that fetches what a page returns. A page blocked to crawlers, unavailable to the public, or returning an error may be difficult or impossible for that engine to use. Do not confuse the existence of a URL in your editor with the page being public and reachable. A new article with a future publish time is a draft, even if you can preview it while signed in.
Figure 2
Figure 2 in words: First a search system learns a public URL from links or other discovery routes. It may crawl the reachable page, decide whether and how to index it, then consider it for a particular search. A failure or exclusion at one stage can prevent a later one. A passed stage is not a promise that the page will rank or receive clicks.
There is also rendering: processing the page so content that depends on scripts can be seen. Google says it renders pages during crawling, but this does not excuse hiding the main explanation behind an unreliable interface (Google Search Central, 2025b). For a beginner, the practical question is: can a visitor and a crawler reach the meaningful text and links on a stable public URL? If the site has a technical problem, give the developer the exact URL and what you observed rather than guessing which tag will fix it.
Indexing can involve choosing a canonical page when very similar content appears at several URLs. Canonical means the version a search engine treats as the main representative for that cluster; a site can provide a preferred canonical URL, but the engine's selection is its own process (Google Search Central, 2025b). On Dhruvi Infinity, an article's canonical public address is under /articles/..., while /knowledge is a discovery surface. When you eventually publish this series, inspect the actual generated article URL and its canonical setting. Do not link the public reader to an imagined slug or assume a draft's SEO preview proves indexing.
Begin with the question, then inspect the page
Choose one real question that the intended person asks at a particular moment. Search intent is the task behind the words: perhaps learning a meaning, comparing approaches, locating an official source, or completing a next step. The same term can carry several intents. Look at language from genuine customer conversations, support questions or site search if available. Public search suggestions and existing results can help you discover phrasing, but they do not reveal the full mind of your customer or guarantee demand.
Do not create a page just because a tool displays a large monthly number. A search volume estimate depends on a provider's data, location, period and grouping rules; a number without those labels is weak evidence. The most useful topic may be a clear question from actual people even when a commercial tool shows little volume. Record the question's source and date. If the wording is assumed, say so and plan how to check it. Avoid copying a competitor's article structure simply because it currently appears near the top.
Figure 3
Figure 3 in words: Record the actual or proposed wording, the person's underlying task, where that wording came from, and what a complete answer would need. Then identify the current page that can provide it or the gap to fill. A keyword phrase is a clue to human language, not a command to repeat the same words unnaturally.
Read the candidate page as someone new to the subject. Does the opening answer the question? Are unfamiliar words defined? Is the scope clear? Are important claims accurate and sourced? Does the page explain limits, current terms and the next action? Is it readable on a phone? Does it serve a person who arrived directly without knowing this series? Google recommends content that would still be useful to the intended audience if they arrived directly, and warns against writing mainly to attract search visits without satisfying the question (Google Search Central, 2026).
If an existing page already answers the question well, improve it instead of making a near duplicate. Add a missing definition, update a stale fact, clarify an instruction or connect it to the next useful page. A new page should have a distinct purpose. Thin duplication means several pages saying nearly the same thing without adding a useful distinction. Neither long length nor repeated keywords turns weak content into an answer. Google explicitly says there is no preferred word count for ranking (Google Search Central, 2026).
Make the result understandable before the click
The page title is the main label for the document and can contribute to the title link seen in search. A good title names the topic and helps the intended reader decide whether the page fits. The visible headline should deliver what it promises. A meta description is a short description associated with the page; a search engine may use it to form a snippet, but it can choose other text from the page instead. Google says it may draw title links from the page's title element and headings, and it may generate snippets from page content rather than displaying the supplied description unchanged (Google Search Central, 2025c).
Your Knowledge editor has a Title, Summary, optional SEO title and optional SEO description. The Title is the public headline and default SEO title; a tighter SEO title may help when the headline is longer. The Summary helps the Knowledge card and related content. The SEO description should explain the page's value clearly in a concise sentence. These are editorial inputs, not guaranteed search-result text. Check the Live SEO preview to see the values being supplied, then inspect the real public page after publication.
Figure 4
Figure 4 in words: The visible title tells a reader what the page covers. The summary helps discovery cards on the site. An SEO title can give search wording a concise form. The meta description can invite an appropriate click, though an engine may rewrite the displayed snippet. All four should describe the same real content, not promise different outcomes.
Write naturally. A title like “What is [real topic]? A beginner guide” may be suitable if the body explains the concept; a title promising a step-by-step tutorial should contain usable steps. Do not repeat a phrase just to fill space. A meta description is not an advertisement for a claim the page cannot substantiate. If the page explains limits or prerequisites, it can say so. A person clicking through and finding a misleading page has not been helped, even if the result drew attention.
The URL is the page address. Readable words can help a person recognise a destination, but do not keep changing a published URL merely to insert a keyword. If a move is necessary, technical redirects and canonical signals need care. In this first pass, record the actual URL generated by the site and keep internal links pointed to it. A public series index can link to each chapter by its verified canonical address; the series key by itself does not create ordered Previous and Next navigation.
Help people and crawlers follow useful links
An internal link joins one page on your site to another. It helps a reader continue to a definition, prerequisite, detailed explanation or next action. It can also help crawlers discover a page. Google says links are an important discovery route and advises descriptive link text that tells readers what they will find (Google Search Central, 2025b; 2025c). The link should work and lead to the relevant destination, not to a guessed path or a private draft.
For this series, the “Start here” guide can be the ordered route once its public URL is confirmed. A newly published Chapter 16 could link back to the relevant earlier article if that article is actually live. Until then, keep the chapter label as a navigation cue without inventing a click target. Link out to an accessible explanation or primary documentation when a reader needs it. An external link is not a magic ranking vote; its reason is that it helps verify or deepen understanding.
Figure 5
Figure 5 in words: A verified public index can lead to the current article. The article can link to a necessary earlier explanation and a useful next step. Every destination must be live and genuinely relevant. When a chapter is still a draft, the diagram's intended arrow is a planning note rather than a public link.
Use link words that make sense out of context. “Read the beginner explanation of customer segments” conveys more than “click here”. Do not create a wall of identical links or force every article to link to every other article. Revisit links when you change titles, merge pages or retire content. Broken links and a confusing route are problems for people first; correcting them also makes a site easier to explore.
Check the page people actually receive
SEO is partly technical access, but the first checks need no specialised tool: open the public URL in a signed-out browser; see whether it loads without a password; check that the main explanation and links appear on a phone; test the next action. If a visitor sees a server error, an empty page, a misleading title or a broken form, rewrite advice alone will not solve the experience. Record the URL, date, device and observation for whoever can fix it.
Some checks need an authorised site owner. A robots.txt rule can instruct crawlers not to fetch certain routes. A noindex directive asks compatible search engines not to include a page in their index. The two are different; blocking a crawl can stop an engine from seeing an on-page instruction. A canonical signal expresses a preferred main version when similar URLs exist. These are technical controls, not switches to change blindly in an article editor. Google identifies crawl blocks, indexing rules and duplicate canonical selection as distinct issues (Google Search Central, 2025b).
Figure 6
Figure 6 in words: A visitor check asks whether the public page works. Discovery asks whether relevant live pages link to it. Crawl asks whether the engine can fetch it. Index asks whether the engine selected it for storage and which canonical it chose. Result asks how it appears for queries. Each observation needs a source, date and appropriate owner; no row guarantees the next.
A sitemap may help an engine discover URLs, but submitting one does not force indexing. Google says pages can be discovered without one and that submitting it in Search Console may help discovery (Google Search Central, 2025a). Bing also says submission does not guarantee inclusion (Bing Webmaster Tools, n.d.). For a single new page, start by ensuring that it is public, useful and linked from a relevant live place. Ask the site owner whether a sitemap and webmaster tools are already configured rather than creating an account just for this worksheet.
Image alt text describes meaningful images for people who cannot see them and can help a search system understand a visual. Every figure in this chapter has an alt label and a fuller “Figure in words” paragraph because a diagram should remain intelligible without pixels. Avoid cramming keywords into alt text. The important content should remain readable text; an image table is used here because the editor cannot paste a table reliably, with the adjacent paragraph providing the essential relationship in words.
Measure without confusing a click with a customer
If you have authorised access to Google Search Console, it can show which pages were seen in Google Search, queries, impressions (appearances under its counting rule) and clicks from search results. It also offers page indexing and URL inspection information (Google Search Central, 2025a). You do not need it to learn the concepts. If access is absent, record that the data is unavailable and check the public page; do not pretend a manual search from your own device measures everyone else's results.
Click-through rate (CTR) is clicks divided by impressions for the same defined report and period. A change in that rate can have many causes: different queries, result features, title wording, audience, position or season. A click says a person selected a result, not that the page answered their question or produced a sale. Average position is a report measure across many impressions, not the fixed rank of a page. Check query, page, country, device and date filters before comparing periods. Search Console's numbers and a website analytics tool's sessions can differ because they record different events (Google Search Central, n.d.).
Figure 7
Figure 7 in words: A search impression indicates a result appeared under a report rule. A click indicates selection. A useful page action, such as reading a relevant explanation or following a clear next step, requires different evidence. A customer outcome requires its own definition. None of the lower steps automatically causes the higher one.
Look for quality of fit, not only traffic totals. A page drawing many unrelated queries might need a clearer title or scope. A page with fewer visits may still answer a specialist question well. Check whether visitors can complete the task and whether the team receives fewer repeated questions or more qualified enquiries, if those are appropriate goals. Do not invent a conversion rate or infer a cause from one week of data. Google says changes can take hours or months to be reflected and suggests allowing several weeks before judging many SEO changes (Google Search Central, 2025c).
At review, keep a note of what changed, when, and what else may have changed. If an index report says the page is excluded, inspect the stated reason and the canonical choice. If a page is indexed but receives no relevant impressions, revisit whether the question, answer and site context are a good fit. If impressions rise but clicks fall, examine the queries and preview, not just the description field. If clicks rise but the page fails the reader, fix the page or destination. The result is an investigation path, not an automated diagnosis.
Try it: improve one real page
Allow about 75 minutes for a first review. Choose a page you control, or use a draft and mark public checks as “not yet available”. No paid SEO tool or search account is required to complete the editorial part. If you cannot edit the live page, write a specific proposal for its owner.
Question and audience (10 minutes). Record one real customer question, its source and date, the intended reader and what a complete answer must include. Mark assumptions.
Page and purpose (8 minutes). Find the existing page or state that none is verified. Write its exact current or proposed address, status and the job it should do. Check for a near duplicate.
Answer audit (15 minutes). Read the page as a beginner. Mark where the answer begins, which terms need defining, what evidence supports it, and what next step or limit is missing.
Fields (10 minutes). Draft an honest headline, concise SEO title if needed, card summary and meta description. Each must describe the same page. Do not promise a result the body cannot deliver.
Connections (10 minutes). Identify one relevant live page that could link here and one genuinely useful destination from this page. Verify each public URL; mark unpublished chapter links pending.
Access (10 minutes). Check public loading, phone readability and the next action. If authorised data exists, note crawl/index/canonical status; otherwise assign that check to the site owner.
Review (12 minutes). Record the change, owner, date, question and next review. Decide which observation will show that the page helped a person, not just appeared in a report.
Figure 8
Figure 8 in words: The worksheet has empty spaces for real page URL, status, intended person and question; answer and evidence; visible title, optional SEO title, summary and meta description; verified incoming and outgoing links; public access and indexing status; owner and review. It contains no sample keyword numbers or fabricated ranking.
Your completed decision can be said aloud: “This page helps [real person] answer [real question]. We checked [evidence and date]. The public page is [actual URL or draft status], and its title promises [real answer]. We will fix [specific gap], verify [access or link], and review [relevant signal] on [date].” An honest unknown is more useful than a made-up traffic forecast.
How this fits the wider channel plan
Search is often valuable when someone expresses a need in words, but not every need is typed into an engine. Social platforms may introduce an unfamiliar idea, email may help someone who has chosen to hear from you, and direct support may resolve a personal problem. Chapter 15 asked which place fits the customer's task; this chapter asks whether a page can be found and answer it. A social post can point to a good page, and a useful page can be shared by a person. Coordinate the message and the next step rather than asking each channel to do everything.
Figure 9
Figure 9 in words: A relevant search can lead to a page with a complete answer. A social post may also lead to it if discovery there suits the task; an invited email may help someone continue; a support route may handle a personal issue. The page must keep its claim and next action consistent across routes. No route is assumed necessary for every offer.
If you use a public Knowledge guide as the chapter index, it should have an ordered list and live links. Article pages remain the canonical reading destinations; the Knowledge hub helps people discover them. The series key and title can group cards but do not by themselves create previous/next buttons or force a search engine to index the series. Check the actual public implementation before telling readers a navigation feature exists.
Figure 10
Figure 10 in words: If the public page fails, fix access and the destination first. If it works but an authorised index report shows exclusion, inspect the stated crawl, index or canonical reason. If it is indexed but mismatched to relevant queries, revisit question and answer. If useful people arrive but cannot complete the task, fix content or the next step. Record the observation and owner before changing tactics.
Common mistakes to catch
Repeating a phrase instead of answering a question.Publishing many near-identical pages to target slight query variations.Writing a title or snippet that overpromises.Assuming a preview equals a public indexed page.Treating a sitemap submission as guaranteed indexing.Changing live URLs without a plan for links and redirects.Counting clicks as sales.Judging one day's movement as proof of SEO success or failure. Each mistake can be corrected by returning to the real reader task, the actual public URL and the evidence from the appropriate stage.
The written explanation and worksheet are complete without videos. Public YouTube titles, dates, view counts and descriptions were inspected on 14 November 2026; the selected videos were not played or transcript-reviewed. Before publishing, check the relevant content, spoken language, captions, UK availability and embed permission. View counts are a dated shortlist comparison, not a measure of teaching quality.
Watch a simple introduction:Intro to Search for Beginners, Google Search Central (23 October 2019). Its description presents an animated beginner series about online presence and tools, and says subtitles are available in various languages. It is older but the basic question remains useful; it does not replace current technical documentation.
Go deeper on the process:Introducing How Search Works, Google Search Central, presented by Gary Illyes (15 February 2024). The description marks the three stages at 01:39 and a closing marker at 03:28; the exact runtime was not independently verified. It comes from the search provider itself and explains that provider's system, rather than independent academic research. The Open University course above supplies a separate educational perspective.
What to remember
SEO helps people find and judge a useful page; it cannot guarantee a ranking.
Keep discovery, crawling, indexing, serving, clicking and customer outcome separate.
Start with a real question, answer it in readable text, and give the page an honest title and working next step.
Link from relevant public pages and verify the actual canonical address. A draft or preview is not a published, indexed article.
Review dated evidence and reader usefulness. Search impressions and clicks are intermediate signals, not proof of a sale.
Further watching: Google Search Central (2019) ‘Intro to Search for Beginners’, YouTube, 23 October. Available at: https://www.youtube.com/watch?v=-0X1AVLIzt0 (Accessed: 14 November 2026). Google Search Central (2024) ‘Introducing How Search Works’, YouTube, 15 February. Available at: https://www.youtube.com/watch?v=5MIAugQ17ks (Accessed: 14 November 2026).
A beginner-friendly route through 26 short lessons on customers, offers, channels, content and measurement. Start anywhere, with no account or paid tools.
Learn the difference between features and benefits, write a promise customers understand, and check which kinds of online value you can honestly deliver.