Acquisition and Search Analytics

Introduction and Context

Acquisition analytics examines how potential customers are reached, attracted and converted through unpaid and paid digital channels. In digital marketing, acquisition is not simply a traffic-generation activity. It is a measurement and optimisation problem: marketers must determine which audiences, messages, search terms, placements and partners produce valuable outcomes at an acceptable cost.

This chapter focuses on four major acquisition areas:

• search engine optimisation (SEO);
• paid search, including search-engine advertising;
• display advertising; and
• affiliate marketing.

These channels use different mechanisms. SEO depends on relevance, technical accessibility and authority; paid search depends on auctions, bids, quality and relevance; display depends on audience selection, creative exposure and frequency; and affiliate marketing depends on partner-generated referrals and agreed commission arrangements. Despite these differences, all can be evaluated through a common analytical structure: reach, visibility, engagement, cost, conversion and incremental business value.

A sound evaluation therefore moves beyond surface-level measures such as clicks or impressions. It asks whether the traffic was relevant, whether conversions were profitable, whether reported outcomes were genuinely caused by marketing activity, and whether the measurement system is reliable.

Learning Objectives

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

1. distinguish SEO, paid search, display and affiliate acquisition;
2. interpret reach, rankings, impressions, click-through rate, cost per click, quality and conversion metrics;
3. calculate and explain CTR, CPC, conversion rate, CPA and ROAS;
4. evaluate search and display performance using relevant benchmarks and diagnostic methods;
5. explain attribution, incrementality and the limitations of platform-reported data;
6. recommend practical optimisation actions while considering privacy, transparency and governance.

1. The Acquisition Measurement Framework

A useful acquisition framework follows the customer journey from exposure to business outcome:

Audience → Exposure → Interaction → Visit → Conversion → Revenue → Incremental value

Each stage answers a different question.

The framework should not be treated as a rigid funnel. Customers may search several times, encounter multiple advertisements, visit through different devices and convert later. Consequently, analysts should combine funnel metrics with path analysis, cohort analysis and controlled experiments.

Essential Terminology

Reach is the number of distinct users or accounts exposed to a message during a specified period. It is different from impressions, which count total exposures. One person seeing an advertisement five times contributes one to reach and five to impressions.

Frequency is the average number of exposures per reached user. When reach and impressions refer to the same population and period:

Frequency is dimensionless. The relationship assumes consistent counting rules and deduplicated reach. It is not necessarily exact when platforms estimate reach or use different identity systems.

Impression means one recorded opportunity for an advertisement or search result to be displayed. An impression does not guarantee that the user noticed or fully viewed it.

Click-through rate (CTR) measures clicks relative to impressions:

CTR is expressed as a percentage. It is meaningful only when the numerator and denominator use the same campaign, period, placement and counting definition. A high CTR can indicate relevance, but it does not prove profitable performance.

Cost per click (CPC) is the average amount paid for each recorded click:

In auction platforms, actual CPC may differ from the maximum bid. CPC should be interpreted alongside conversion rate and customer value.

Conversion rate (CVR) is the proportion of defined visits or clicks that produce a conversion:

The denominator must be stated. A click-based conversion rate and a session-based conversion rate are not interchangeable.

Cost per acquisition (CPA) measures the cost of obtaining one recorded conversion:

CPA is valid only when conversions are defined consistently and are sufficiently attributable to the campaign. If conversions have different values, CPA alone may be misleading.

Return on advertising spend (ROAS) compares attributed revenue with advertising cost:

ROAS is a ratio, often stated as “3.0:1” or “300%.” It is not the same as profit. If gross margin is 40%, a ROAS of 2.0 may not cover product, fulfilment, labour and overhead costs.

Worked Example: From Clicks to ROAS

An online retailer records 80,000 impressions, 2,400 clicks, 120 purchases and £18,000 in attributed revenue. Advertising cost is £3,600.

CTR:

CPC:

Click-based conversion rate:

CPA:

ROAS:

The campaign generated efficient recorded outcomes, but a complete decision would also require margin, refunds, repeat purchases, customer lifetime value and incrementality evidence.

2. Search Engine Optimisation Analytics

SEO is the process of improving a website’s visibility in organic search results. Search engines generally attempt to rank results according to relevance, usefulness, quality, accessibility and user experience. Exact ranking systems are proprietary and continuously changing; therefore, marketers should avoid claiming that one isolated factor guarantees a ranking position.

SEO Measurement Dimensions

Visibility and reach. Analysts monitor organic impressions, organic clicks, indexed pages, query coverage and the proportion of the target audience reached. Search Console-type data can show how often pages appeared for queries, although impressions are often estimated and do not equal actual attention.

Rankings. Average position, ranking distribution and presence in high-value result features can indicate visibility. Average position is an aggregate statistic: it can conceal large differences between queries, locations, devices and pages. A position of 5 across two queries does not mean that every user saw the page in position 5.

CTR. Organic CTR is calculated from clicks and organic impressions. It is influenced by ranking, title and description quality, brand familiarity, search intent, competing features and device layout. Comparing CTR across unrelated query groups is unreliable.

Technical health. Important indicators include crawl accessibility, indexation, broken links, redirect quality, mobile usability, page performance, structured data validity and canonicalisation. Technical scores from software tools are diagnostic aids, not direct measures of search-engine success.

Content and authority. Content should satisfy user intent, be accurate and demonstrate appropriate expertise. Links, mentions and reputation can contribute to authority, but link acquisition must be editorially legitimate. Automated, manipulative or deceptive practices can create search penalties and reputational risk.

Business outcomes. Organic sessions, qualified leads, transactions, revenue, assisted conversions and customer quality are more decision-relevant than rankings alone. SEO programmes often have delayed effects, so cohort and trend analysis are important.

Search Intent and Query Classification

Search queries may be classified as informational, navigational, commercial investigation or transactional. The same keyword can have different intent in different contexts. For example, “running shoes” may indicate product research, while “buy waterproof running shoes size 9” is more explicitly transactional.

A professional analyst maps queries to landing pages and evaluates whether the page satisfies the underlying need. Poor performance may result from a mismatch between query intent and page content rather than from insufficient traffic.

SEO Diagnostic Logic

• High impressions, low CTR: review title, description, result features and intent alignment.
• High CTR, low engagement: inspect landing-page relevance, speed, mobile experience and content quality.
• Strong engagement, low conversion: examine offer, trust, form design, pricing and conversion friction.
• Falling impressions: investigate seasonality, technical changes, algorithm updates, competitors and demand changes.
• High ranking but low commercial value: reassess keyword targeting and business relevance.

3. Paid Search Analytics

Paid search places advertisements in response to user queries. It is particularly valuable when the query expresses a current need. Campaign structures commonly separate brand, non-brand, product, service, competitor and remarketing-related activity.

Auction, Relevance and Quality

A paid-search auction typically considers the advertiser’s bid, expected ad performance, landing-page experience and other relevance or policy signals. Platforms use proprietary formulas and terminology; therefore, simplified auction explanations should not be treated as universal equations.

Quality-related measures help diagnose whether an advertisement and landing page are relevant to the query. Quality should not be treated as a single objective measure of campaign success. A campaign can have high quality and low business value, or lower quality on a highly profitable niche term.

Analysts should examine:

• query-to-keyword relevance;
• ad-message alignment;
• landing-page usefulness;
• expected CTR;
• device and location performance;
• search-term quality;
• impression share and lost impression share;
• conversion volume and value; and
• brand versus non-brand economics.

Impression share is generally the proportion of eligible impressions received. Lost impression share may arise from budget or rank limitations. Because eligibility is estimated, impression share is a diagnostic measure, not a complete measure of market demand.

Paid Search Optimisation

Optimisation should begin with business objectives. For lead generation, a low CPA may still be poor if leads are unqualified. For ecommerce, revenue or profit-based bidding may be more suitable than maximising conversions. Useful actions include adding negative keywords, separating intent groups, improving ad assets, testing landing pages, adjusting budgets by marginal return and removing low-quality placements or queries.

Budget decisions should consider diminishing returns. The next unit of spending may produce fewer conversions than earlier spending. Thus, a campaign’s average CPA can conceal the marginal CPA of additional budget.

4. Display Advertising Analytics

Display advertising uses visual or rich-media formats across websites, applications, video environments and digital networks. It can support awareness, consideration, remarketing and direct response.

Display measurement requires caution because impressions may be served below the fold, viewed briefly, blocked, invalid or exposed repeatedly to the same users. Important measures include reach, frequency, viewability, CTR, post-click conversions, view-through conversions, CPA, assisted outcomes and brand lift.

Reach, Frequency and Viewability

A campaign with broad reach may build awareness but produce limited immediate clicks. A campaign with high frequency may reinforce memory or cause irritation. The appropriate level depends on audience, creative, purchase cycle and objective.

Viewability is a technical opportunity-to-see measure, not proof that a user consciously noticed an advertisement. Standards and platform definitions may differ. Analysts must document the measurement provider and threshold used.

Display Creative and Landing Pages

Creative performance should be evaluated by audience and placement, not only by aggregate CTR. Strong creative communicates a clear proposition, uses legible design, matches the audience’s stage and provides an appropriate call to action. Frequency analysis can reveal creative fatigue, while sequential testing can compare messages across the customer journey.

Display can create demand rather than merely capture existing demand. Consequently, last-click reporting may undervalue it. Brand-lift studies, geo experiments, holdout groups and time-series methods can provide stronger evidence of impact.

5. Affiliate Marketing Analytics

Affiliate marketing rewards external partners for referred traffic or conversions. Partners may include content publishers, comparison sites, loyalty programmes, influencers, coupon sites and technology platforms. Commercial arrangements can use cost per acquisition, revenue share, fixed fees or hybrid models.

Core measures include clicks, conversion rate, approved transactions, reversal rate, commission, effective CPA, revenue, new-customer rate and incremental contribution.

A simple affiliate commission calculation is:

This assumes the commission rate applies uniformly to eligible revenue and excludes returns, cancellations, taxes or excluded products. In practice, contract rules must define eligibility.

Affiliate reporting faces several challenges. Coupon or loyalty partners may receive credit for customers who were already likely to purchase. Some partners may use brand bidding, forced clicks, misleading claims or unauthorised voucher codes. Therefore, a high volume of affiliate conversions does not automatically represent incremental demand.

Governance should specify disclosure requirements, prohibited traffic sources, brand-bidding rules, data-use restrictions, fraud controls, approval processes and reversal procedures.

6. Attribution and Incrementality

Attribution assigns credit for a conversion to one or more marketing interactions. Common models include last-click, first-click, linear, position-based, time-decay and algorithmic attribution. Each model is a rule for allocating credit; none automatically proves causation.

Last-click attribution is simple and operationally useful, but it tends to favour channels close to conversion. First-click highlights discovery but may overvalue early interactions. Multi-touch models distribute credit, yet their results depend on incomplete journeys, identity resolution and modelling assumptions.

Incrementality

Incrementality asks: what additional outcomes occurred because of the marketing activity, compared with what would have happened without it?

A basic lift calculation is:

The outcome rates must be measured over comparable populations and periods. If treatment and control groups contain 10,000 and 10,000 users respectively, with conversion rates of 6% and 5%, then:

Estimated incremental conversions in the treatment group are:

Thus:

This interpretation assumes random or adequately balanced assignment, consistent measurement, no material contamination between groups and comparable exposure conditions. If the control group is not valid, the estimate may reflect selection bias rather than marketing impact.

An incremental CPA can be calculated as:

Similarly, incremental ROAS is:

These measures are generally more decision-relevant than platform-attributed CPA and ROAS, although experiments can be expensive and may have limited generalisability.

7. Integrated Channel Evaluation

Channels should be evaluated both independently and as a portfolio. SEO may capture existing demand, paid search may capture high-intent demand, display may stimulate awareness and affiliates may extend distribution. Removing one channel can alter the performance of others.

A practical dashboard should contain:

The analyst should segment results by brand status, query intent, device, location, audience, landing page, creative, partner and new versus returning customer. Aggregation can conceal important patterns, such as profitable non-brand search being subsidised by branded search.

Illustrative Case: Northstar Outdoor Equipment

Background

Northstar Outdoor Equipment, a fictional retailer, sells hiking equipment online. Its marketing team invests in SEO, paid search, display remarketing and affiliate partnerships. Reported revenue is increasing, but management is uncertain whether growth comes from genuinely incremental marketing or from customers who would have purchased anyway.

Problem

The team reports the following monthly results:

Management initially ranks channels by attributed ROAS. However, this comparison is incomplete because SEO cost includes internal labour, affiliate revenue may include coupon-assisted purchases, and display may influence later paid-search conversions.

Analysis

Paid-search ROAS is:

Display ROAS is:

Affiliate ROAS, treating commission as cost, is:

SEO’s attributed return on internal cost is:

These figures are not directly comparable. The team therefore conducts three further investigations:

1. It separates branded and non-branded paid-search terms.
2. It applies a holdout test to a sample of display audiences.
3. It reviews affiliate transactions for new-customer status, coupon use, reversals and partner type.

The display holdout group produces a 5% conversion rate, while the exposed group produces a 6%. The estimated lift is 1 percentage point. If 20,000 users were in the exposed group, incremental conversions are 200. The reported 180 conversions are therefore not necessarily inconsistent with the test, because definitions, windows and populations differ; the experiment should be checked carefully before drawing conclusions.

Affiliate review finds that coupon partners receive a large share of last-interaction credit, while content publishers introduce more new customers. Northstar changes commission rules, protects brand terms, and reports new-customer contribution separately.

Approach and Outcome

Northstar adopts a dashboard combining attributed and incremental measures. It reduces low-quality display frequency, retains prospecting display where lift is demonstrated, expands high-performing non-brand SEO content, and reallocates affiliate budget toward partners with verified new-customer and repeat-purchase value.

Learning Points

• Attributed ROAS is not causal ROAS.
• Cost definitions must be consistent across channels.
• Channel quality matters as much as conversion volume.
• Experiments are particularly important for channels that influence rather than immediately capture demand.
• Governance rules are necessary to prevent affiliate and advertising practices that create misleading credit.

8. Data Quality, Ethics and Governance

Acquisition analysis depends on reliable data. Common problems include duplicate conversions, missing consent, cross-device identity gaps, time-zone differences, delayed refunds, ad blockers, bot traffic, inconsistent UTM parameters and changes in platform definitions.

A measurement plan should specify event names, conversion windows, attribution rules, currency, tax treatment, cancellation handling, data owners and quality checks. Analysts should reconcile platform data with analytics systems and financial records rather than assuming that any one source is complete.

Privacy regulation requires lawful processing, transparency, purpose limitation, data minimisation and appropriate retention. Marketers should obtain valid consent where required, avoid collecting unnecessary sensitive data, provide clear disclosures and respect opt-out signals. Audience targeting should not exploit vulnerable groups or use sensitive characteristics in discriminatory ways.

AI-assisted bidding, creative generation and audience modelling can improve efficiency, but they can also reproduce bias, obscure decision logic and create inaccurate claims. Human review, testing, brand-safety controls and documented governance remain necessary. Automated optimisation should be constrained by business rules, not allowed to maximise a narrow metric without regard to profitability or customer welfare.

9. Best-Practice Operating Process

1. Define the business outcome: revenue, qualified leads, margin, retention or awareness.
2. Establish a measurement plan and consistent conversion definitions.
3. Segment audiences, intent, device, geography, brand status and customer type.
4. Diagnose the journey from exposure to conversion rather than relying on one metric.
5. Compare platform attribution with analytics and financial data.
6. Use controlled tests when causal impact is important.
7. Optimise marginal performance, not only historical averages.
8. Review quality, fraud, privacy, accessibility and brand safety.
9. Document assumptions, changes and limitations.
10. Reassess channel interaction and long-term customer value.

Chapter Summary

Acquisition and search analytics evaluates how SEO, paid search, display and affiliate activity creates measurable and valuable customer outcomes. Reach and impressions describe scale; rankings and impression share describe visibility; CTR describes response to exposure; CPC describes click cost; conversion rate describes action; CPA describes recorded acquisition efficiency; and ROAS compares attributed revenue with advertising cost.

These measures must be interpreted in context. High traffic or low CPA does not guarantee profitable customers, while display and SEO may create value that last-click attribution fails to recognise. Incrementality, measured through valid control groups or other robust methods, provides stronger evidence of causal impact. Reliable decisions also require consistent definitions, segmentation, data-quality controls, privacy governance and attention to customer quality and long-term value.

Key Terms

Reach: Number of distinct users exposed to a message.

Impression: One recorded opportunity for an advertisement or result to be displayed.

Frequency: Average number of exposures per reached user.

Ranking: The position of an organic result for a search query.

CTR: Clicks divided by impressions, expressed as a percentage.

CPC: Cost divided by recorded clicks.

Conversion rate: Conversions divided by eligible visits or clicks.

CPA: Cost divided by recorded conversions.

ROAS: Attributed revenue divided by advertising cost.

SEO: Practice of improving organic search visibility and usefulness.

Paid search: Advertising delivered in response to search queries.

Quality: Relevance and experience signals used to diagnose paid-search performance.

Display advertising: Visual or rich-media advertising across digital environments.

Affiliate marketing: Partner-based promotion compensated for agreed outcomes.

Attribution: Allocation of conversion credit among marketing interactions.

Incrementality: Additional outcome caused by marketing compared with a valid counterfactual.

Control group: Comparable group not receiving the tested marketing treatment.

Viewability: A technical measure of whether an advertisement met a defined opportunity-to-see standard.

Landing page: Web page designed to receive traffic from a campaign or search result.

Marginal performance: Outcome generated by the next unit of spending or activity.

Review Questions

5 Short Answer Questions

1. Distinguish reach from impressions.
2. What does CTR measure, and why does a high CTR not guarantee profitability?
3. State the formulas for CPC and CPA.
4. Why can average organic ranking be misleading?
5. What is the purpose of a control group in an incrementality test?

5 Long Answer Questions

1. Explain how SEO performance should be evaluated from visibility through to business outcomes.
2. Discuss the role of relevance and quality in paid-search performance.
3. Compare the measurement challenges of display advertising and affiliate marketing.
4. Explain why attributed ROAS should not automatically be interpreted as profit or causal return.
5. Design an integrated acquisition dashboard for an online retailer and justify its measures.

3 Analytical / Application Questions

1. A campaign produces 50,000 impressions, 1,500 clicks, 75 conversions and £9,000 in cost. Calculate CTR, CPC, conversion rate and CPA. Interpret the results and identify one limitation of each measure.
2. A display campaign exposes 30,000 users and records a 7% conversion rate. A comparable control group of 30,000 users records a 5.5% conversion rate. Estimate incremental conversions and explain the assumptions required for a credible conclusion.
3. An affiliate programme has strong attributed ROAS but most conversions use a coupon code immediately before purchase. Propose an investigation and recommend at least three governance or measurement changes.

Multiple Choice Questions with Answers

Question 1

A campaign receives 4,000 clicks from 100,000 impressions. What is its CTR?

A. 0.4%

B. 4%

C. 25%

D. 40%

Correct Answer: B. 4%

Question 2

An advertiser spends £2,400 and records 120 conversions. What is the CPA?

A. £2

B. £20

C. £200

D. £2,880

Correct Answer: B. £20

Question 3

Which measure compares attributed revenue with advertising cost?

A. CTR

B. CPC

C. ROAS

D. Frequency

Correct Answer: C. ROAS

Question 4

Which SEO result most directly suggests that search visibility is not generating attention?

A. High impressions and low organic CTR

B. Low impressions and high conversion value

C. High repeat purchase rate

D. Low internal labour cost

Correct Answer: A. High impressions and low organic CTR

Question 5

Why can last-click attribution undervalue display advertising?

A. It ignores advertising cost completely

B. It gives all credit to the first interaction

C. It may fail to credit earlier demand-influencing exposures

D. It measures only organic rankings

Correct Answer: C. It may fail to credit earlier demand-influencing exposures

Question 6

A treatment group converts at 8% and a control group at 6%. What is the estimated incremental lift?

A. 2 percentage points

B. 14 percentage points

C. 25 percentage points

D. 48 percentage points

Correct Answer: A. 2 percentage points

Question 7

Which affiliate practice creates a major incrementality concern?

A. Publishing useful product comparisons

B. Receiving credit for customers who were already about to purchase

C. Reporting approved transactions

D. Disclosing commercial relationships

Correct Answer: B. Receiving credit for customers who were already about to purchase

Question 8

What does frequency generally describe?

A. The proportion of clicks that convert

B. The average number of exposures per reached user

C. The cost of one conversion

D. The percentage of eligible impressions won

Correct Answer: B. The average number of exposures per reached user

Question 9

Which statement about ROAS is correct?

A. It is always equal to profit

B. It measures only organic ranking quality

C. It compares attributed revenue with advertising cost

D. It proves that advertising caused every conversion

Correct Answer: C. It compares attributed revenue with advertising cost

Question 10

A campaign has 12,000 impressions and reaches 3,000 distinct users. What is its average frequency?

A. 0.25

B. 2

C. 4

D. 9,000

Correct Answer: C. 4