Search Engine Marketing

Introduction and Context

Search engine marketing (SEM) is the practice of using search engines to make an organisation visible when people search for information, products, services or solutions. It combines search engine optimisation (SEO)—earning unpaid visibility—with paid search advertising, in which an advertiser pays for eligible clicks or other outcomes.

Search is distinctive because it often captures explicit demand. A person searching “emergency plumber near me” has expressed a more immediate need than a person who merely encounters a plumbing advertisement on social media. However, search behaviour exists across the customer journey. Queries such as “how to repair a leaking tap” may indicate information seeking, while “buy replacement tap online” indicates stronger commercial intent.

Effective SEM is therefore not simply a traffic-generation activity. It involves understanding search behaviour, building technically accessible websites, producing useful content, creating relevant advertisements, managing auctions and measuring business outcomes. SEO and paid search should be coordinated, but they are not interchangeable: SEO depends primarily on relevance, accessibility, authority and user experience over time; paid search depends on campaign design, auction participation, bids, budget and ad quality.

Learning Objectives

By the end of this chapter, you should be able to:

1. Explain how search engines crawl, index and rank web content.
2. Classify keywords according to intent and customer journey stage.
3. Apply on-page, technical and off-page SEO principles.
4. Design a logically structured paid search account.
5. Explain bidding, ad rank, quality and auction economics.
6. Calculate and interpret core search-performance metrics.
7. Optimise SEM campaigns ethically using evidence and experimentation.

1. How Search Engines Work

A search engine performs three connected activities: crawling, indexing and serving or ranking.

Crawling

Crawlers, also called bots or spiders, discover URLs through links, sitemaps and other signals. A crawler requests a resource and may retrieve its HTML, images, structured data and rendered content. Robots.txt can communicate which paths a site operator does not want crawled, although it is not a security mechanism and does not guarantee that a URL will remain unknown or absent from search results.

Crawling can be limited by server performance, poor internal linking, duplicate URLs, infinite calendar or filter combinations, and access restrictions. A crawlable page is not necessarily indexed.

Indexing

During indexing, the search engine analyses a page, identifies its content and relationships, and stores a representation in a large index. It may canonicalise similar URLs, select a representative version, process structured data and assess whether the page provides sufficient value for inclusion. A page can be technically accessible but excluded, duplicated, or considered unsuitable for a particular query.

Ranking and results serving

For a query, the search engine retrieves potentially relevant documents and orders them using numerous signals. These may include textual relevance, content quality, links and reputation, usability, freshness where appropriate, location, device, language and the meaning of the query. Ranking is query-dependent: a page can rank highly for one search and poorly for another.

Search results may include organic listings, text advertisements, shopping results, maps, images, videos, news and answer-oriented features. Visibility should therefore be evaluated across the whole search results page, not only the traditional ten blue links.

A useful conceptual model is:

User query → Query interpretation → Candidate retrieval → Relevance and quality evaluation
           → Auction and/or organic ordering → Search results → User interaction → Outcome data

This is a simplified model rather than a complete description of any particular search engine’s proprietary systems. Search engines continuously update their systems, and no ethical practitioner can guarantee a particular organic position.

2. Search Intent and Keyword Strategy

A keyword is a word or phrase used to represent a search query or a target topic. In practice, marketers should distinguish the keyword selected in an advertising platform from the user’s actual search term. Search platforms may match one selected keyword to several related queries, subject to match settings and platform rules.

Types of intent

Intent is not always explicit. “Running shoes” could be research, browsing or purchase-oriented. Search results themselves provide evidence: if results are dominated by product pages, shopping listings and retailers, commercial intent is likely; if they are dominated by tutorials, informational intent is more likely.

The customer journey

Keyword planning can map queries to awareness, consideration, conversion and retention. Early-stage content should answer genuine questions without forcing an immediate sale. Mid-stage content can compare options and explain suitability. Conversion pages should make the desired action clear and reduce uncertainty through price, specifications, reviews, delivery information, guarantees or contact options.

Keyword selection

A professional keyword process includes:

1. Define business objectives, audiences, products and geographic scope.
2. Generate seed topics from customer language, sales teams, support logs and competitor research.
3. Expand and group terms using search data and qualitative judgement.
4. Assess intent, relevance, competition, seasonality and likely value.
5. Map each group to one appropriate page or paid-search ad group.
6. Identify exclusions and negative keywords for paid campaigns.
7. Review actual search terms after launch.

Search volume is not demand quality. A lower-volume query with clear commercial intent may produce more value than a high-volume generic query. Keyword difficulty tools are estimates, not guarantees. The central test is whether the organisation can provide the best useful response to the searcher’s need.

3. Organic SEO Principles

On-page SEO

On-page SEO concerns signals and content within a page. Important practices include:

• Write a clear, descriptive title and compelling meta description. A meta description may influence click behaviour but is not a guaranteed ranking control.
• Use a meaningful main heading and a logical heading hierarchy.
• Address the query comprehensively without repetition or unnatural keyword insertion.
• Use descriptive URLs, relevant internal links and informative anchor text.
• Optimise images with appropriate dimensions, efficient formats, descriptive alternative text where the image conveys meaning, and empty alternative text when it is decorative.
• Provide visible authorship, sources, qualifications and update information where these help users assess reliability.
• Make the conversion path obvious and consistent with the search intent.

There is no required keyword density. Repeating a phrase to reach an arbitrary percentage can reduce readability and does not establish expertise. Search engines increasingly interpret meaning, entities and relationships rather than relying only on exact phrase matching.

Technical SEO

Technical SEO improves discoverability, interpretability, performance and usability. Key areas include:

• Crawl control: sound internal linking, appropriate XML sitemaps, correct status codes and careful handling of parameters.
• Indexation control: canonical links, noindex directives where appropriate, and avoidance of accidental blocking.
• Mobile experience: responsive layouts, readable text, usable controls and parity between mobile and desktop content.
• Performance: fast server response, efficient assets, stable layout and responsive interaction. Performance improvements should be evaluated with both laboratory and real-user data.
• Security: HTTPS, secure forms, sensible access controls and protection of customer information.
• Structured data: machine-readable information that may help search engines understand entities such as products, organisations, events or reviews. Structured data does not guarantee enhanced results and must accurately represent visible content.
• Accessibility: semantic HTML, keyboard access, sufficient contrast, captions where needed and meaningful alternatives for non-text content. Accessibility is a user responsibility in its own right, not merely an SEO technique.

Common technical failures include staging sites accidentally indexed, redirected chains, broken links, duplicate product pages, thin faceted-navigation URLs, slow third-party scripts and JavaScript-dependent content that is not reliably rendered.

Off-page SEO

Off-page SEO concerns signals beyond the organisation’s own pages. The most important long-term principle is earning references because the content, research, tool or service is genuinely useful. Relevant editorial links can support discovery and reputation, but links purchased to manipulate ranking, automated link schemes and deceptive exchanges create quality and governance risks.

Off-page activity can include digital public relations, original research, expert commentary, partnerships, community participation and accurate local citations. The objective is not to collect the largest possible number of links. Relevance, trust and editorial context matter more than a crude link count.

Local SEO

For organisations serving a location, local visibility depends on accurate business information, a well-managed business profile, appropriate categories, service-area clarity, locally relevant pages and authentic reviews. Reviews should not be fabricated, selectively manipulated or obtained through coercion. Consistency is useful, but minor formatting differences across directories are not equivalent to evidence of quality.

4. Paid Search Campaigns

Paid search places an eligible advertisement in an auction for a query or audience context. The advertiser normally pays when a user clicks a cost-per-click (CPC) advertisement, although other models and automated bidding goals exist.

Account structure

A logical structure commonly contains:

• Account: billing, access, security and shared settings.
• Campaign: objective, budget, location, language, network and bidding controls.
• Ad group: a tightly related theme of keywords and advertisements.
• Keyword or targeting criterion: the condition that makes an ad eligible.
• Advertisement and assets: headline, description, URL and supplementary information.
• Landing page: the destination designed to satisfy the query and business goal.

For example, an online bicycle retailer might separate campaigns for brand, commuter bicycles, mountain bicycles and accessories. Ad groups within a commuter campaign could distinguish “city bike” from “folding commuter bike”. Excessively granular structures are not automatically better; fragmentation can reduce data and make management inefficient.

Match settings and negative keywords

Paid platforms provide controls over how closely a keyword relates to a search query. Exact, phrase and broad approaches differ by platform and change over time, so practitioners must consult current platform documentation. Regardless of match setting, search-term reports are essential.

Negative keywords prevent ads from showing for unsuitable searches. A premium consultancy might exclude “free”, “course”, “jobs” and “template” if those searches do not match its offer. Negative keywords must be reviewed carefully because an exclusion that is too broad may remove valuable demand.

Ad relevance and landing pages

An effective advertisement reflects the user’s need, communicates a credible benefit and presents a clear next step. It should not promise something absent from the landing page. Landing pages should load reliably, work on mobile devices, display the relevant offer above the fold where practical, provide evidence and make conversion straightforward.

5. Auctions, Bidding and Quality

Search advertising is not simply a contest in which the highest bid always wins. A simplified conceptual representation is:

Actual platforms use more complex, context-sensitive systems, and the notation above is not a contractual platform formula. Quality-related factors may include expected click-through rate, ad relevance, landing-page experience, search context and competition. Eligibility, assets, thresholds and auction-specific rules also matter.

Bidding strategies may prioritise clicks, conversions, conversion value, impression share or a target return. Automated bidding can process signals that a human cannot manage manually, but it depends on reliable conversion tracking, sufficient data, appropriate constraints and responsible monitoring. Automation does not remove the need for strategy.

Useful equations include:

Here, impressions and clicks are counts, cost is expressed in the account currency, conversions are defined completed actions, CTR and conversion rate are percentages, CPC is currency per click, and CPA is currency per conversion. These formulas assume that the relevant events are consistently defined and attributed within the same reporting scope. A conversion rate based on clicks is not interchangeable with one based on sessions or users.

For value-based campaigns:

ROAS is a ratio, often expressed as “3.0” or “300%”. It is meaningful only when attributed revenue is measured consistently and does not by itself represent profit. If gross margin, fulfilment, staff and overhead costs are substantial, a campaign can have a high ROAS and still be unprofitable.

Worked example

An illustrative campaign records 50,000 impressions, 1,500 clicks, 60 conversions and £1,800 advertising cost. Attributed revenue is £7,200.

CTR = 1,500 ÷ 50,000 × 100% = 3%.

Conversion rate = 60 ÷ 1,500 × 100% = 4%.

CPC = £1,800 ÷ 1,500 = £1.20.

CPA = £1,800 ÷ 60 = £30.

ROAS = £7,200 ÷ £1,800 = 4.0, or 400%.

The campaign generated four pounds of attributed revenue per pound of advertising cost. Whether it should be expanded depends on margin, customer lifetime value, capacity and the reliability of attribution—not ROAS alone.

6. Measurement and Optimisation

Measurement begins with business outcomes. A lead-generation organisation might value qualified leads, while an online retailer might value profit-adjusted revenue. Define conversion events, primary and secondary actions, ownership, counting rules and data-retention practices before launch.

A practical optimisation cycle is:

1. Diagnose: identify changes in impressions, clicks, costs, conversion rate and outcomes.
2. Segment: compare query, device, location, audience, landing page, hour and campaign where data volume supports comparison.
3. Form a hypothesis: for example, “mobile users abandon because the form is difficult to complete.”
4. Change one meaningful variable or run a controlled experiment.
5. Allow adequate time and volume, while accounting for seasonality and lag.
6. Evaluate incremental business impact, not only platform metrics.
7. Document the result and decide whether to scale, reverse or investigate further.

Do not judge SEO solely by rankings. Assess qualified organic traffic, visibility for strategically important topics, engagement appropriate to the page, leads, sales and assisted outcomes. Rankings fluctuate with location, personalisation, competition and result features.

Attribution models are interpretations, not perfect observations of causality. Last-click attribution is simple but can undervalue discovery; multi-touch models distribute credit according to assumptions; incrementality tests ask what would have happened without the marketing activity. Where possible, use experiments, geographic tests or holdout designs alongside platform reporting.

7. Illustrative Case: Northstar Home Energy

Background

Northstar Home Energy, a fictional regional installer, sells heat-pump assessments and installations. Its previous marketing used broad paid keywords and directed all traffic to a generic home page.

Problem

The company received many clicks but few qualified enquiries. Search-term review showed that advertisements appeared for free government information, employment searches and do-it-yourself queries. The landing page did not explain eligibility, installation time or financing. Organic pages used technical jargon and had weak internal linking.

Analysis

The team grouped queries into informational (“are heat pumps suitable for old homes”), commercial investigation (“heat-pump running costs”) and transactional/local (“heat-pump installer Bristol”). It reviewed analytics, call records, form quality and page-speed data. It found that mobile users represented most traffic but encountered a long, difficult form.

Approach and solution

Northstar created separate paid campaigns for brand, local installation and high-intent service terms. It added negative keywords for employment, free information and DIY searches, while checking that exclusions did not remove relevant terms. New ad groups used closely related themes and landing pages matched the promise in each advertisement.

The SEO programme produced a plain-language suitability guide, a running-cost explainer with stated assumptions, locally relevant service pages and an internal-link structure connecting educational content to assessment pages. Technical work corrected canonical errors, improved mobile form usability and compressed oversized images. Conversion tracking distinguished brochure downloads, assessment requests and qualified telephone enquiries.

Outcome

After a measurement period, the company observed a lower proportion of irrelevant clicks, improved enquiry quality and a higher mobile completion rate. The exact business outcome would require verified sales and margin data; therefore, the team did not claim that every improvement was caused solely by one change. It continued testing page messaging and monitored whether paid demand was incremental or merely replacing organic demand.

Learning points

The case demonstrates that query relevance, account structure, landing-page experience, technical quality and measurement operate as a system. Cutting CPC alone would not solve a qualification problem, and increasing traffic without controlling intent could worsen efficiency.

8. Ethics, Governance and Contemporary Practice

Search marketing should respect privacy, consumer protection, accessibility and platform rules. Organisations must obtain and process personal data lawfully, provide appropriate consent and notice where required, secure customer information and avoid retaining unnecessary data. Tracking and remarketing should be governed rather than implemented indiscriminately.

Advertisements must be truthful, identifiable and supportable. Hidden charges, fabricated reviews, misleading urgency, deceptive redirects and discrimination in targeting can harm users and expose an organisation to legal and reputational consequences. SEO practitioners should reject cloaking, malicious links, doorway pages and scaled low-value content designed primarily to manipulate rankings.

Generative AI can assist with keyword clustering, content outlines, code suggestions, query analysis and reporting. It can also produce inaccurate claims, repetitive content, privacy breaches, biased language and unverifiable sources. Human subject-matter review, fact checking, originality, editorial accountability and secure data handling remain essential. AI should augment research and production rather than replace responsibility.

Contemporary search is increasingly multimodal, conversational and answer-oriented. Users may receive summaries, maps, product information or voice responses without visiting a website. Organisations should therefore publish clear, structured, trustworthy information and measure visibility and business outcomes beyond clicks. At the same time, a search feature’s appearance does not guarantee that a cited source receives traffic; strategic value must be assessed empirically.

Best-Practice Checklist

• Start with customer and business intent, not a list of high-volume keywords.
• Make important content crawlable, indexable, useful and accessible.
• Match each advertisement and landing page to a coherent query theme.
• Use negative keywords and search-term reports to control waste.
• Track qualified outcomes, revenue and margin where feasible.
• Treat automated bidding as a managed system requiring reliable data.
• Test changes with explicit hypotheses and account for attribution limits.
• Earn authority through useful work rather than manipulative links.
• Protect privacy and make advertising claims honest and verifiable.
• Document decisions, assumptions and changes for organisational learning.

Chapter Summary

Search engines discover pages through crawling, interpret and store them through indexing, and order eligible results according to query relevance, quality, context and other signals. Search engine marketing combines SEO and paid search, which share the need for relevance but use different mechanisms. Keyword strategy begins with intent and customer needs, not volume alone. On-page SEO improves clarity and usefulness; technical SEO supports crawling, indexation, performance, security and accessibility; off-page SEO develops reputation through genuinely valuable work.

Paid search requires coherent account and ad-group structure, appropriate match controls, negative keywords, relevant advertisements and effective landing pages. Auctions consider more than bids, and automated bidding is only as reliable as its goals and conversion data. CTR, CPC, conversion rate, CPA and ROAS describe performance but require consistent definitions and should be interpreted with margin, quality and incrementality. Ethical governance requires truthful advertising, responsible data practices, accessibility and resistance to manipulative SEO. Professional SEM is a continuous cycle of research, implementation, measurement, experimentation and learning.

Key Terms

Ad rank: A platform-specific assessment influencing an advertisement’s eligibility and position in an auction.

CPC: Cost per click; advertising cost divided by clicks.

Crawl: The process by which a search-engine bot discovers and requests web resources.

CTR: Click-through rate; clicks divided by impressions.

Conversion: A defined completed action that represents a valuable outcome.

Conversion rate: Conversions divided by the selected denominator, commonly clicks.

Crawling: Search-engine discovery and retrieval of URLs.

Indexing: Processing and storing a search engine’s representation of content.

Keyword: A word or phrase used to represent a search topic or advertising target.

Landing page: The destination page reached after an advertisement or search result is selected.

Negative keyword: A paid-search exclusion that prevents eligibility for specified searches.

Off-page SEO: Activities and signals outside a website that influence discovery or reputation.

On-page SEO: Content and page-level practices that improve relevance and usability.

Organic result: An unpaid search result.

Quality factors: Relevance, experience and context signals that can affect paid-search eligibility or performance.

ROAS: Return on advertising spend; attributed advertising revenue divided by advertising cost.

Search intent: The underlying purpose of a user’s query.

SEM: Search engine marketing, including SEO and paid search activity.

SEO: Search engine optimisation; practices that improve unpaid search visibility and usefulness.

Structured data: Machine-readable information that helps describe page entities and content.

Review Questions

Short Answer Questions

1. What are the three principal activities performed by a search engine before results are served?
2. Distinguish informational intent from transactional intent.
3. State two examples of technical SEO work.
4. What is the purpose of a negative keyword?
5. Why is ROAS not the same as profit?

Long Answer Questions

1. Explain how crawling, indexing and ranking are related, and describe why a crawlable page may not appear in search results.
2. Discuss how on-page, technical and off-page SEO principles work together to support organic visibility.
3. Design a paid-search account structure for a business with three distinct product categories and justify your decisions.
4. Explain the role of bidding, relevance, landing-page experience and conversion tracking in paid-search optimisation.
5. Evaluate the ethical and governance responsibilities of an organisation using SEO, paid search and AI-assisted content production.

Analytical / Application Questions

1. A campaign receives 80,000 impressions, 2,400 clicks, 96 conversions and costs £2,880. Calculate CTR, conversion rate, CPC and CPA, showing your workings.
2. A retailer’s traffic increases after broad match keywords are introduced, but qualified leads fall. Analyse possible causes and recommend a measurement-led optimisation plan.
3. A local service firm ranks for informational searches but receives few enquiries. Use intent mapping, content design, landing-page principles and measurement to propose an integrated SEO and paid-search response.

Multiple Choice Questions with Answers

Question 1

Which sequence best represents the basic search-engine process?

A. Indexing → crawling → payment

B. Crawling → indexing → serving or ranking results

C. Bidding → crawling → indexing

D. Ranking → publishing → crawling

Correct Answer: B. Crawling → indexing → serving or ranking results

Question 2

Which query most clearly expresses transactional intent?

A. How do heat pumps work?

B. History of heat pumps

C. Buy heat-pump installation package

D. Are heat pumps suitable for old homes?

Correct Answer: C. Buy heat-pump installation package

Question 3

What is the primary purpose of a negative keyword in paid search?

A. To increase the maximum bid automatically

B. To prevent ads showing for unsuitable searches

C. To create additional organic backlinks

D. To guarantee a top advertisement position

Correct Answer: B. To prevent ads showing for unsuitable searches

Question 4

Which practice is most directly associated with technical SEO?

A. Writing a comparison article

B. Earning an editorial link

C. Correcting canonical and indexation errors

D. Negotiating a sponsorship

Correct Answer: C. Correcting canonical and indexation errors

Question 5

A campaign has 10,000 impressions and 400 clicks. What is its CTR?

A. 0.4%

B. 4%

C. 25%

D. 40%

Correct Answer: B. 4%

Question 6

A campaign costs £900 and produces 30 conversions. What is its CPA?

A. £3

B. £27

C. £30

D. £930

Correct Answer: C. £30

Question 7

Why should automated bidding be monitored carefully?

A. It cannot use conversion data

B. It always ignores budget constraints

C. Its performance depends on reliable goals and data

D. It eliminates the need for landing pages

Correct Answer: C. Its performance depends on reliable goals and data

Question 8

Which statement about structured data is most accurate?

A. It guarantees a rich result for every page

B. It accurately describes content in machine-readable form but does not guarantee enhanced results

C. It replaces accessible visible content

D. It is only used for paid advertisements

Correct Answer: B. It accurately describes content in machine-readable form but does not guarantee enhanced results

Question 9

Why can a high ROAS still be insufficient for managerial decision-making?

A. ROAS measures impressions rather than revenue

B. ROAS ignores whether attributed revenue exceeds total relevant costs

C. ROAS is always calculated from organic traffic

D. ROAS cannot be compared across campaigns

Correct Answer: B. ROAS ignores whether attributed revenue exceeds total relevant costs

Question 10

Which approach best reflects ethical SEO practice?

A. Creating doorway pages for every minor keyword variation

B. Buying large numbers of irrelevant links

C. Publishing useful, accurate content and earning relevant references

D. Hiding different content from users and crawlers

Correct Answer: C. Publishing useful, accurate content and earning relevant references