Week 7 Lecture: Paid Search and PPC

Discussion

1 Introduction & Recall

⏱ 10 min

Aims; poll on match types, negatives, and Quality Score levers

Acquisition

2 PPC Foundations

⏱ 30 min

Auction and Quality Score, match types, negatives, RSAs, landing pages, efficiency metrics

Practice

3 Guided Whole-Class Discussion

⏱ 15 min

Mini case: pick a match type and one negative, propose a Quality Score fix and a safeguard

Assessment

4 Attendance & Exit Check

⏱ 5 min

Attendance, one-minute exit check

Module Arc

Weeks 1–2

Foundations and Planning

→

Weeks 3–5

Content and Media

→

Weeks 6–8

Acquisition Channels

→

Weeks 9–12

Relationships and Measurement

→

Weeks 13–15

Governance and AI

We are in the middle of the Acquisition Channels block. Your SEO mini-audit from Week 6 gave you a keyword set with an intent classification. This week adds match type logic, bid structure, and ad copy to turn that research into a PPC skeleton.

Today’s Agenda

Lecture: eight core PPC concepts

  1. How the Google Ads auction works: Ad Rank formula
  2. Quality Score: the multiplier that changes everything
  3. Keyword match types: Broad, Phrase, Exact
  4. Negative keywords: budget protection
  5. Responsive search ads: headlines, descriptions, extensions
  6. PPC efficiency metrics: CTR, CPC, CVR, CPA, ROAS
  7. CPA diagnostic: finding the root cause of high cost
  8. Worked example: PPC skeleton for Tidewood Guesthouse

Tutorial preview

You will produce a PPC skeleton for your campaign:

  • Name your campaign and ad group; assign match types from your Week 6 keyword themes
  • Write a 10-term negative keyword list
  • Draft 6 to 8 headline variants and 4 description variants for a starter RSA
  • Pick a conversion metric, set a daily cap, and choose two safeguards

Learning Objectives (1 of 2)

After studying this chapter, you should be able to:

  1. Explain how the Google Ads auction determines ad position using bid, Quality Score, and expected impact of extensions.
  2. Apply keyword match type logic to build a keyword list that targets the right audience intent without wasting budget on irrelevant queries.
  3. Write a set of negative keywords that protect the campaign from irrelevant traffic.
  4. Structure a responsive search ad with 6 to 8 headline variants and 4 description variants aligned to the target keyword’s intent.

Learning Objectives (2 of 2)

After studying this chapter, you should be able to:

  1. Audit a landing page against the four tests that decide whether it converts the Conviction-stage audience paid search delivers: claim match, mobile speed, a single call to action, and message match.
  2. Calculate and interpret the primary PPC efficiency metrics: click-through rate, conversion rate, cost per click, and cost per acquisition.
  3. Apply the evidence framework to PPC decisions: classify keyword choices, bid strategy, and landing page decisions as evidence-based, inferred, or assumed.

All seven objectives appear in your Week 7 PPC skeleton submission. The Campaign ROAS Calculator in the web book chapter supports Objective 6 directly.

The PPC Auction

The auction determines which ads appear, in what order, and at what cost. A higher bid only sometimes buys a higher position. A lower bid on a high-quality ad regularly outperforms a higher bid on a poor-quality one.

Ad Rank = Maximum Bid × Quality Score × Expected Impact of Extensions

The Ad Rank Formula in Practice

Worked comparison:

Advertiser Max bid Quality Score Ad Rank Position
Advertiser A MVR 50 4 200 Second
Advertiser B MVR 30 8 240 First

Advertiser B pays less per click AND achieves a higher position because Quality Score doubles the effect of a lower bid. Improving Quality Score is an efficiency strategy as much as a quality improvement.

Quality Score: Three Components

Quality Score is calculated from three components, each scored as above average, average, or below average.

Component What it measures How to improve it
Expected CTR Whether the ad is expected to be clicked more or less often than similar ads for the same keyword Write headline variants that match search intent; test using RSA pin settings; use the target keyword in Headline 1
Ad relevance How closely the ad copy matches the intent of the keyword Each ad group should target a narrow set of related keywords; ad copy must reflect the keyword’s specific intent
Landing page experience How relevant, useful, and fast the landing page is for users who click the ad Match the landing page headline to the ad claim; ensure the page loads in under 3 seconds on mobile; provide a single clear CTA

A Quality Score below 5 is a warning signal: investigate each component before making bid changes.

The Quality Score cycle:

Keyword Match Types

Match types determine how closely a user’s search query must match your keyword before your ad is triggered. Tighter match types mean higher intent specificity and lower irrelevant traffic. Broader match types mean higher reach and higher budget risk.

The choice of match type is a budget protection decision as much as a reach decision.

The Match Type Spectrum

Match type Syntax Example keyword Also triggers for
Broad guesthouse Maldives guesthouse Maldives “budget hotel Maldives”, “backpacker hostel Maldives”, “Maldives resort deals”
Phrase "guesthouse Maldives" “guesthouse Maldives” “guesthouse Maldives Malé”, “affordable guesthouse Maldives”
Exact [guesthouse Maldives] [guesthouse Maldives] “guesthouse Maldives”, “guesthouses Maldives” only

Match Type Strategy for Funnel Stages

A well-structured PPC campaign uses a mix of match types mapped to funnel stages and budget priorities.

Funnel stage Hierarchy stage Recommended match type Rationale
Awareness Awareness Broad with strong negatives Discover new relevant queries; high reach needed
Consideration Knowledge to Liking Phrase match Specific intent preserved; some variation acceptable
Decision Preference to Conviction Exact match Maximum relevance; budget allocated to highest-intent queries
Direct response Conviction to Purchase Exact match, branded User is already searching by name; high conversion likelihood

A common mistake: using broad match for conversion campaigns and then wondering why the CPA is high. The CPA is high because the campaign is paying for irrelevant clicks from queries that have no purchase intent.

Negative Keywords: Budget Protection

Negative keywords prevent the ad from being triggered when specified terms appear in the search query. They are the single most important budget protection mechanism in PPC management.

A campaign without a negative keyword list is a campaign with open budget exposure.

Building the Negative Keyword List

Negative keywords come from two sources: anticipation (before launch) and search query data (after launch).

Before launch: anticipation list

For a guesthouse direct-booking campaign, the anticipation list should include:

  • “free” (users seeking no-cost stays)
  • “hostel”, “backpacker”, “dorm” (users seeking a different accommodation type)
  • “jobs”, “career” (users seeking employment, over a stay)
  • “Wikipedia” (users seeking reference content)
  • “Airbnb”, “Booking.com”, “Agoda” (users defaulting to an OTA rather than a direct booking)
  • “review”, “TripAdvisor” (users researching rather than ready to book)

These are obvious mismatches identifiable in advance from knowledge of the campaign category.

After launch: search query review

After two weeks of campaign data, review the Search Query Report. This shows every actual search term that triggered your ad.

Add to the negative list:

  • Any query producing clicks but no conversions after 20 or more clicks
  • Any query where the intent is clearly misaligned
  • Any query from a geographic market outside your service area

Review cadence:

  • Week 1 to 4: weekly review and update
  • Month 2 onward: fortnightly review
  • Never treat the negative list as complete

Budget protection stays continuous, well past the launch-time task of building the first list.

Responsive Search Ads

Responsive search ads allow up to 15 headline variants and 4 description variants. The platform’s machine learning tests combinations and optimises toward the combinations that earn the highest click-through rate for each query and user context.

The discipline: every headline and description must be independently useful and independently relevant.

RSA Structure and Writing Rules

The three key headline positions:

Position Role Example
Headline 1 Target keyword or primary claim “Tidewood Guesthouse Maldives” (28)
Headline 2 Primary differentiator “House Reef, No Boat Needed” (26)
Headline 3 CTA, trust signal, or supporting claim “Book Direct, No Booking Fees” (28)

Additional headline slots (4-15): rotation variants the system will test. Each must stand alone.

Description lines (90 characters each):

  • Description 1: expand on the primary claim with a specific, verifiable differentiator
  • Description 2: provide evidence or a call to action that reinforces Description 1

The independence rule: write each headline so it makes sense on its own, even if paired with a completely different headline. A headline reading “With House Reef Access” only makes sense paired with a specific other headline, and the system decides its own pairings.

Example RSA for Tidewood Guesthouse:

Headlines (character count in brackets): 1. Tidewood Guesthouse Maldives (28) 2. House Reef, No Boat Needed (26) 3. Priced Below Nearby Resorts (27) 4. Book Direct, No Booking Fees (28) 5. International Guests Welcome (28) 6. Check September Availability (28) 7. Private House Reef Access (25)

Descriptions (90 max): 1. Snorkel our own house reef steps from your room, at a fraction of resort prices. (80) 2. Book direct and skip OTA fees. September rooms available now. Enquire online today. (83)

Character counts: each headline must be 30 characters or fewer. Each description must be 90 characters or fewer. Check before submitting.

Pinning: Certainty at the Cost of Learning

By default the system places any headline in any position, and may leave an asset out of a given combination entirely.

Pinning overrides that. Pin a headline to position 1 and it appears in position 1 every time the ad serves, or the ad sits out that auction entirely.

Worth pinning

  • A legal disclaimer that must appear
  • A regulated price or claim wording
  • A brand name that has to lead

The cost of pinning

  • Each pin removes combinations the system could have tested
  • Pin most of the set and the machine learning has almost nothing left to optimise
  • Testing is the mechanism that lifts CTR over time

The working compromise: build two ad variants and leave at least one of them completely unpinned. Your Week 7 submission asks for exactly this.

Ad Extensions: Expanding the Ad’s Real Estate

Ad extensions (now called “assets” in Google Ads) appear below the headline and description and expand the ad’s footprint on the page without increasing cost.

Extension type What it shows Use case
Sitelink Additional links to specific pages on your site “Room Types”, “Guest Reviews”, “House Reef Photos”, “Book Now”
Callout Short phrases highlighting key attributes “House Reef Access”, “No Booking Fees”, “Airport Transfer”, “Free Cancellation”
Call Phone number click-to-call Direct enquiry for the reservations team
Location Address and map link Physical location of the guesthouse for travellers planning a route
Structured snippet List of amenities or features “Amenities: Snorkelling Gear, Airport Transfer, Free Wi-Fi, Breakfast Included”

Extensions are an input to Ad Rank through “expected impact of extensions.” Adding relevant extensions increases Ad Rank without increasing the bid, effectively improving position efficiency.

PPC Efficiency Metrics

Four metrics determine whether a PPC campaign is performing efficiently. They must be read together: each one explains a different segment of the conversion pathway, and each one points to a different root cause when performance is poor.

CTR, CVR, CPC, CPA. In that sequence.

The Four Metrics and Their Diagnostic Chain

Metric Formula What a poor result indicates Primary improvement lever
CTR Clicks / Impressions Keyword-ad relevance mismatch or low Quality Score Improve headline relevance; test variants; refine match types
CVR Conversions / Clicks Landing page problem or audience intent mismatch Improve page relevance, speed, and CTA
CPC Total spend / Total clicks Low Quality Score, high competition, or overbidding Improve Quality Score; use more specific keywords
CPA Total spend / Total conversions Campaign generating conversions inefficiently Address root cause: CTR (ad) or CVR (page) first

CPA Diagnostic: Finding the Root Cause

Bidding Strategies

Google Ads offers several automated bidding strategies alongside manual CPC bidding.

Strategy How it works Best when
Manual CPC You set the maximum CPC for each keyword You have historical data and want full control; useful in the first 2-4 weeks
Maximise Clicks Algorithm spends the budget to get as many clicks as possible Awareness-stage campaigns where volume matters more than conversion quality
Target CPA Algorithm adjusts bids to achieve your CPA target across all conversions You have at least 30-50 conversions per month and a stable CPA target
Target ROAS Algorithm adjusts bids to achieve your revenue return target E-commerce campaigns where revenue per conversion is known and variable
Maximise Conversions Algorithm spends the budget to get as many conversions as possible You want to maximise output within a fixed budget without a specific CPA target

For new campaigns: start with Manual CPC for the first two to four weeks to gather clean data on which keywords convert. Switch to Target CPA only once you have at least 30 conversions in the conversion window. An automated bidding strategy with insufficient data will optimise towards noise.

Landing Page Design for PPC

A PPC click only starts the conversion process. The landing page determines whether that click-through rate ever translates into a conversion rate.

Four tests the landing page must pass before the campaign launches:

  1. Specificity: does the landing page match the ad’s specific claim? A landing page promoting the general property when the ad promised house reef access will produce a high bounce rate. The user clicked for reef access, so show it immediately.

  2. Speed: does the page load in under three seconds on mobile? Each additional second of load time increases bounce rate. A click that costs MVR 25 and bounces in 2 seconds produces no conversion data and no value.

  3. Single CTA above the fold: the action the ad promised sits fully visible before the user scrolls. One primary CTA, clearly visible without scrolling, on mobile.

  4. Message match: the headline on the landing page should echo the headline of the ad. This confirms to the user they have arrived in the right place. A mismatch (ad says “house reef access”; page says “Welcome to our guesthouse”) breaks the scent trail and signals irrelevance.

Campaign ROAS Calculator: Interactive Tool

The Campaign ROAS Calculator is available in the Week 7 chapter of the web book.

What it does:

Enter your campaign inputs (budget, target CPC, estimated conversion rate, and average order value in MVR) and the calculator projects:

  • Estimated clicks from the budget
  • Estimated conversions at the CVR assumption
  • Cost per acquisition (CPA)
  • Return on Ad Spend (ROAS)
  • A bar chart comparing budget spent against revenue generated

Tutorial task:

Use the calculator with your PPC skeleton’s CPC estimates from Ubersuggest. Label the conversion rate input as [ASSUMPTION]. At what budget does your campaign break even? What CVR is required for a ROAS of 2.0?

The ROAS formula:

\[ROAS = \frac{Revenue}{Ad\ Spend}\]

A ROAS of 1.0 means you recovered the ad spend in revenue. A ROAS above 2.0 is typically a profitable campaign (after accounting for product cost or service delivery cost). A ROAS below 1.0 means you spent more on ads than you recovered in revenue.

CPA and ROAS connection:

\[CPA = \frac{CPC}{CVR}\]

If CPC is MVR 20 and CVR is 4%, CPA is MVR 500. If the product value is MVR 2,000 and COGS is MVR 1,200, the margin is MVR 800. A CPA of MVR 500 stays profitable; a CPA of MVR 900 erases the margin.

Worked Example: PPC Skeleton for Tidewood Guesthouse

Scenario: the Week 6 SEO audit built its keyword cluster around “best guesthouse Maldives house reef access” as the primary term. That same cluster becomes the foundation of the PPC campaign.

Campaign structure:

  • Campaign: “Tidewood Direct Book 2026”
  • Ad group 1: House Reef Snorkelling Malé
  • Keywords:
    • [best guesthouse maldives house reef access] (Exact)
    • “affordable guesthouse maldives snorkelling” (Phrase)
    • “guesthouse vs resort maldives price” (Phrase)
    • Broad with negatives: guesthouse maldives, budget accommodation maldives

Negative keywords (anticipation list): “free”, “hostel”, “backpacker”, “dorm”, “jobs”, “Wikipedia”, “Airbnb”, “Booking.com”

RSA headlines: 1. Tidewood Guesthouse Maldives 2. House Reef Snorkelling, No Boat Needed 3. Priced Below Nearby Resorts 4. Book Direct, No Booking Fees 5. International Travellers Welcome 6. Check September Availability

Budget and efficiency assumptions:

Input Value Classification
Daily budget 50 clicks/week target Decision
Target CPC MVR 15-25 Assumption
Estimated CVR 5% Assumption
CPA target MVR 800 per enquiry Decision
Landing page load target Under 2 seconds mobile Decision
Testimonials 3, international travellers Evidence (existing)

CPA target rationale: the MVR 800 CPA target is set against the value of one booking enquiry and gets tested directly once the first two weeks of campaign data confirm or challenge the 5% conversion rate assumption.

Week 1-2 actions: run manual CPC; review the search query report weekly; revise the conversion rate assumption once real data replaces it.

Device, Location, and Time Bid Adjustments

Bid adjustments allow campaigns to increase or decrease bids for specific targeting dimensions without creating separate campaigns.

Adjustment type Example adjustment Rationale
Device Mobile: +20% If mobile converts at a higher rate than desktop, bid more aggressively for mobile clicks
Location Greater Malé: +30% If the target audience is concentrated in a specific location, increase bids there
Time of day Weekday evenings 6pm-10pm: +25% If conversion rate is highest during specific hours, allocate budget there
Audience Remarketing list: +50% Users who have already visited the site convert at higher rates; bid more for them

Evidence standard: base bid adjustments on data over general industry knowledge. Check whether your specific campaign’s mobile CVR runs higher or lower than desktop CVR before applying a mobile adjustment. Apply adjustments after at least two to four weeks of data.

The AI Protocol for Week 7

AI may help you draft ad headline variants and identify negative keyword categories. Live tools and real campaign data alone supply the numbers your PPC skeleton requires, and you verify every character count yourself.

What requires a live tool or real data:

  • Keyword volume and competition data: Ubersuggest or Google Keyword Planner (free in the Ads interface)
  • Competitor ad copy: Google Ads Transparency Center (free, no sign-up)
  • CPC estimates: from keyword tool research or live campaign data
  • Conversion rate: from historical campaign data or labelled [ASSUMPTION] if unavailable

Exit Ticket

Before you move into the tutorial, answer these three questions:

  1. What decision does the evidence support? Name one keyword from your campaign where the match type decision (Broad, Phrase, or Exact) is supported by the intent classification evidence from your Week 6 audit. Explain the match type choice in one sentence.

  2. What remains uncertain? Name one assumption in your PPC skeleton that will only be resolved by two to four weeks of live campaign data. What metric will you use to test it?

  3. What ethical risk did you identify? PPC campaigns can use remarketing audiences built from website visitor data. Name one ethical constraint on how that data may be used to target ads, and what disclosure is required to the audience whose data is used.

Write these answers now. They form the foundation of your PPC skeleton’s assumptions section. Students who label assumptions before launch make better optimisation decisions in week two.