Responsive Search Ad Examples: Complete Asset Sets for Five Businesses

Five complete fictional responsive search ad asset sets—each with 10 headlines, four descriptions, two paths, character counts, minimum-combination checks, and a corrected failure.

By
Hookin Team, Performance Editorial
Published
September 10, 2026
Reading time
18 min read
Views
32 views
On this page
  1. Start with the RSA contract, not a screenshot
  2. Example 1: Local plumbing — sell a scheduled visit, not an emergency response
  3. Example 2: B2B scheduling SaaS — keep every CTA on the demo path
  4. Example 3: Insulated-bottle ecommerce — keep the offer and its limits together
  5. Example 4: Online language course — separate access, payment, and outcomes
  6. Example 5: Office cleaning — qualify the scope before promising a schedule
  7. Stress-test the smallest ad before admiring the largest one
  8. Your serving pool may be wider than your RSA table
  9. Measure a pool change, not a mythical winning headline
  10. The reusable lesson behind all five examples
  11. Sources

A responsive search ad is not one piece of copy. It is a pool of copy that can be reordered, shortened, omitted, pinned, and—under some configurations—combined with assets or URLs that are not visible in the ad-level editing table. That is why a polished screenshot is a weak RSA example. The more useful question is: does every eligible version still describe the same offer and the same next step?

The five responsive search ad examples below answer that question with complete fictional teaching sets. Each has 10 headlines, four descriptions, two display-path fields, counted characters, explicit business facts, a minimum one-headline/one-description check, and a failure corrected without inventing a new benefit. They have not been uploaded, approved, served, or measured in Google Ads.

A downloadable CSV of all 80 assets and a self-contained RSA example explorer accompany the article.

Start with the RSA contract, not a screenshot

For a manually entered RSA, Google currently requires at least three headlines and two descriptions and allows up to 15 headlines and four descriptions. Headlines can contain up to 30 characters, descriptions up to 90, and each optional display-path field up to 15. But the serving floor is smaller: Google says an ad can show as little as one headline and one description, and assets may appear in different orders. Its editor preview is not exhaustive, and some formats can shorten text. See Google Ads’ RSA documentation and its definition of display paths.

Question Working rule for these examples
What must be entered? 3–15 headlines and 2–4 descriptions
What can be shown at minimum? 1 headline and 1 description
What are the English-field limits? 30 characters per headline; 90 per description; 15 per path
Can order change? Yes, unless eligible assets are pinned to positions
Which pinned positions can carry must-show text? Use Headline 1 or Description 1 for an every-format requirement; Google lists Headline 2 among pinnable protected positions but also warns it may be omitted in some formats
Are paths guaranteed? No. They are optional and should describe what the landing page contains
Does the preview prove every version works? No. It shows potential ads, not every combination, and shortening can occur

An inventory label such as H07 is only a worksheet ID. It does not mean “Headline position 7”—there are no seven visible headline positions. Likewise, filling all 15 headline fields is not the definition of completeness. A set is complete when the copy pool, its facts, its pins, its wider campaign context, and its destination can be audited together.

Five-step RSA safety workflow: verify facts, inspect the manual asset pool, test the smallest ad, audit wider campaign assets and URLs, then check the destination and measurement.
A complete set begins with verified business facts and ends with the actual destination. The ad-level table is only the middle of the system.

For consistency, the five examples use one controlled teaching configuration: text customization off; Final URL expansion off; URL inclusions off; no campaign-level headlines or descriptions; and no other enabled RSA donating assets. Those are not universal recommendations. They simply make the manual pool examinable. Later, we will add the wider account checks back in. All example.com destinations below are nonfunctional placeholders.

Example 1: Local plumbing — sell a scheduled visit, not an emergency response

Search intent: A Madison homeowner looking for a planned faucet, toilet, or indoor sink-leak repair visit.

Brief constraints

  • Residential work in Madison, Wisconsin; no sewer, gas, or electrical work.
  • Weekday appointments only; no 24/7, emergency, same-day, or guaranteed-arrival claim.
  • The form requests a visit. A team member confirms availability afterward.
  • The customer receives an estimate before work, but the estimate is not described as free or fixed.

Hypothetical final URL: https://example.com/plumbing/madison. The real page would need the Madison service area, covered jobs, hours, estimate process, and a request form whose confirmation language matches the ad.

Slot Asset text Count Pin
H01 Madison Plumbing Visits 23 H1
H02 Scheduled Madison Plumber 25 H1
H03 Faucet Repair Appointments 26
H04 Toilet Repair Visits 20
H05 Help With Sink Leaks 20
H06 Request a Plumbing Visit 24
H07 Estimates Before Repair Work 28
H08 Plan a Weekday Visit 20
H09 Residential Plumbing Help 25
H10 Example Plumbing Co. 20
D01 Schedule plumbing repairs in Madison. Weekday visits; no emergency callouts. 76
D02 Request a faucet or toilet repair visit in Madison. We confirm availability first. 82
D03 For Madison sink leaks and home plumbing repairs, ask for an estimate before work. 82
D04 Tell us about your Madison plumbing job online. Our team will arrange the next step. 84
P01 plumbing 8
P02 madison 7

Minimum coherent teaching render
Madison Plumbing Visits (23)
Request a faucet or toilet repair visit in Madison. We confirm availability first. (82)

Why it holds together: Two service-led alternatives are pinned to Headline 1. The other headlines can rotate as details or actions, while every description still makes sense without a second description or a third headline.

One combination to reject

Failure Why it breaks Correction
“Emergency Plumber Open 24/7” (27) + “Weekday visits by appointment.” (30) A small weekday qualifier cannot repair a headline that falsely promises an emergency service. Narrow the claim at the source. “Scheduled Madison Plumber” (25) + “Request a faucet or toilet repair visit in Madison. We confirm availability first.” (82)

Reader action: Confirm the actual service area, opening hours, covered jobs, estimate terms, and what the form truly does before adapting this set.

Example 2: B2B scheduling SaaS — keep every CTA on the demo path

Search intent: A team buyer comparing shared scheduling workflows, not a consumer looking for a free personal calendar app.

Brief constraints

  • Shared booking links and team availability are in scope; round-robin routing is not assumed.
  • Google Calendar and Outlook connections are supported; “works with every calendar” is not.
  • Meeting buffers and email reminders are available; SMS, no-show reduction, and time-saved outcomes are not claimed.
  • The conversion is a guided product-demo request. There is no assumed free trial, free plan, or cardless start.

Hypothetical final URL: https://example.com/team-booking/demo. The page would need to explain team use, named calendar connections, buffers, email reminders, pricing-on-request, and the demo form.

Slot Asset text Count Pin
H01 Team Scheduling Software 24
H02 Book Meetings Across Teams 26
H03 Shared Availability Links 25
H04 Google Calendar Connection 26
H05 Outlook Calendar Connection 27
H06 Set Meeting Buffer Times 24
H07 Send Email Reminders 20
H08 See the Team Workflow 21
H09 Request a Product Demo 22
H10 Example Team Scheduler 22
D01 Explore team scheduling with shared booking links. Request a guided product demo. 81
D02 Connect Google or Outlook calendars for team scheduling. See the setup in a demo. 81
D03 Set meeting buffers and email reminders in our scheduling app. Ask for a walkthrough. 85
D04 Compare team booking workflows in a product demo. Pricing is available on request. 82
P01 team-booking 12
P02 demo 4

Minimum coherent teaching render
Example Team Scheduler (22)
Connect Google or Outlook calendars for team scheduling. See the setup in a demo. (81)

Why it holds together: The pool covers category, workflow, named integrations, features, and one conversion stage. It avoids mixing a self-serve trial CTA into a sales-assisted demo journey.

One combination to reject

Failure Why it breaks Correction
“Start Your Free Trial” (21) + “Request a Product Demo” (22) A free-trial CTA creates a different conversion path from the demo-only brief. Every CTA should describe the same next step. “See the Team Workflow” (21) + “Request a Product Demo” (22)

Wider-pool check: another active RSA could donate Start a Free Trial as a link-based asset and send the click to its donor URL, https://example.com/individual/free-trial, instead of this set’s https://example.com/team-booking/demo. Audit donor text and donor destinations together. Google documents this cross-RSA behavior here.

Reader action: Audit other active RSAs in the ad group as well as this one. A borrowed link-based asset can carry its donor ad’s URL, so a personal free-trial donor can break an otherwise coherent team-demo set.

Example 3: Insulated-bottle ecommerce — keep the offer and its limits together

Search intent: A shopper evaluating one 24 oz insulated stainless-steel bottle.

Brief constraints

  • The product has a screw cap, carry loop, and three colors. No leakproof or cup-holder claim is assumed.
  • It is described for cold drinks only and should be hand washed; no hot-drink or dishwasher claim.
  • Free standard shipping applies at a $60+ merchandise subtotal, after discounts and before tax, to the contiguous United States.
  • No specific temperature-retention duration is claimed.

Hypothetical final URL: https://example.com/water-bottle/24oz. The page would need product dimensions, care instructions, colors, cold-use language, and the exact shipping threshold, geography, and method.

Slot Asset text Count Pin
H01 24 oz Insulated Bottle 22 H1
H02 Insulated Water Bottle 22 H1
H03 Screw Cap With Carry Loop 25
H04 Made for Cold Drinks 20
H05 Stainless Steel Bottle 22
H06 Refill for Your Daily Routine 29
H07 Choose Your Bottle Color 24
H08 View the Bottle Details 23
H09 Shop the 24 oz Bottle 21
H10 Example Bottle Co. 18
D01 Cold drinks only. Free standard shipping on $60+ orders to the contiguous US. 77 D1
D02 Shop a 24 oz insulated steel bottle with a screw cap and carry loop. View the details. 86
D03 Choose one of three colors for your 24 oz bottle. Hand wash the bottle and cap. 79
D04 See dimensions and care instructions for this insulated bottle before you order. 80
P01 water-bottle 12
P02 24oz 4

Minimum coherent teaching render
24 oz Insulated Bottle (22)
Cold drinks only. Free standard shipping on $60+ orders to the contiguous US. (77)

Why it holds together: Two product-identifying headlines are pinned to Headline 1, and the cold-use plus shipping qualification is pinned to Description 1. That pinning choice is editorial: it protects a condition the example treats as an every-render invariant.

One combination to reject

Failure Why it breaks Correction
“Free Shipping” (13) + “Shop a 24 oz insulated steel bottle with a screw cap and carry loop. View the details.” (86) “Free Shipping” can survive while its threshold and geography disappear. Keep the attractive term and its material conditions in the same protected Description 1. “Shop the 24 oz Bottle” (21) + “Cold drinks only. Free standard shipping on $60+ orders to the contiguous US.” (77)

Wider-pool check: a campaign-level D1 pinned to the same position can also be eligible. A candidate such as Browse Our Bottle Collection would drop the cold-use and shipping conditions. Remove it or rewrite every D1 candidate to preserve them. Google explains the shared-position eligibility here.

Reader action: Verify the SKU, care instructions, threshold calculation, eligible geography, shipping method, and campaign-level Description 1 candidates before reuse.

Example 4: Online language course — separate access, payment, and outcomes

Search intent: An adult beginner comparing a self-paced Spanish course before buying.

Brief constraints

  • The course contains 20 self-paced lessons with everyday dialogue, listen-and-repeat audio, and quizzes.
  • There are no live classes, tutors, certificates, or guaranteed fluency outcomes.
  • One purchase provides 12 months of access; there is no subscription, auto-renewal, lifetime access, or installment plan.
  • The hypothetical US list price is $79 plus applicable tax, with an outline and sample lesson available before purchase.

Hypothetical final URL: https://example.com/spanish/beginners. The page would need the learner level, lesson outline, sample lesson, price, tax treatment, access period, and explicit self-paced format.

Slot Asset text Count Pin
H01 Beginner Spanish Online 23
H02 Self-Paced Spanish Course 25
H03 20 Guided Spanish Lessons 25
H04 Practice Everyday Dialogues 27
H05 Listen and Repeat Exercises 27
H06 Preview a Sample Lesson 23
H07 12 Months of Course Access 26
H08 One-Time Course Purchase 24
H09 Learn at Your Own Pace 22
H10 Example Spanish Course 22
D01 Start beginner Spanish online with 20 self-paced lessons. Preview a sample lesson. 82
D02 $79 plus applicable tax buys 12 months of Spanish access. No subscription or renewal. 85
D03 Practice everyday Spanish with audio exercises and lesson quizzes. No live classes. 83
D04 Review the lesson outline and a sample before buying this beginner Spanish course. 82
P01 spanish 7
P02 beginners 9

Minimum coherent teaching render
Example Spanish Course (22)
$79 plus applicable tax buys 12 months of Spanish access. No subscription or renewal. (85)

Why it holds together: The headlines distinguish course format, contents, access term, purchase basis, and preview action. None needs a neighboring headline to finish its grammar or to reveal that the course is self-paced.

One combination to reject

Failure Why it breaks Correction
“Only $6.58 per Month” (20) + “12 Months of Course Access” (26) $79 ÷ 12 = $6.5833…, but that is only a monthly equivalent. It does not create a purchasable monthly plan; $6.58 × 12 also rounds to $78.96, not $79. “One-Time Course Purchase” (24) + “$79 plus applicable tax buys 12 months of Spanish access. No subscription or renewal.” (85)

Reader action: Confirm the actual price, tax display, access clock, renewal behavior, lesson format, and any outcome evidence before making a stronger promise.

Example 5: Office cleaning — qualify the scope before promising a schedule

Search intent: A Madison office manager seeking a recurring workplace-cleaning plan.

Brief constraints

  • The service is for offices, not homes.
  • Floors, restrooms, and bins can be included; exact scope follows a walkthrough and written agreement.
  • Frequency, access, and cleaning hours are discussed. Nightly, 24/7, instant, and same-day service are not assumed.
  • No sanitization, pathogen-removal, sterile, certification, or performance-guarantee claim is made.

Hypothetical final URL: https://example.com/office-clean/madison. The page would need the office-only service area, scope process, access discussion, walkthrough request, and quote process.

Slot Asset text Count Pin
H01 Office Cleaning in Madison 26
H02 Recurring Office Cleaning 25
H03 Cleaning Plans for Workplaces 29
H04 Restrooms, Floors and Bins 26
H05 Schedule a Site Walkthrough 27
H06 Discuss Your Access Hours 25
H07 Scope Agreed Before Service 27
H08 Get an Office Cleaning Quote 28
H09 For Offices, Not Homes 22
H10 Example Office Care 19
D01 Plan recurring office cleaning in Madison. Request a walkthrough and a scoped quote. 84
D02 Discuss floors, restrooms and bin service for your office. Frequency is agreed in writing. 90
D03 Arrange office access and cleaning hours before service. Start with a site walkthrough. 87
D04 Our Madison service is for offices, not homes. Share your space and cleaning needs. 83
P01 office-clean 12
P02 madison 7

Minimum coherent teaching render
Example Office Care (19)
Discuss floors, restrooms and bin service for your office. Frequency is agreed in writing. (90)

Why it holds together: The pool says what is cleaned and how the engagement starts without turning routine cleaning into an unsupported health-efficacy claim.

One combination to reject

Failure Why it breaks Correction
“A Germ-Free Office” (18) + “Kills Every Pathogen” (20) A cleaning scope is not evidence that a service creates a germ-free environment or kills every pathogen. Replace the outcome claim with the work actually included. “Restrooms, Floors and Bins” (26) + “Discuss floors, restrooms and bin service for your office. Frequency is agreed in writing.” (90)

Reader action: Confirm the served geography, facility types, inclusions, access requirements, quote process, and any substantiation behind hygiene language.

Stress-test the smallest ad before admiring the largest one

The fastest useful audit is not “Did we fill every slot?” It is “What survives when most slots disappear?” For an unpinned 10-headline, four-description set, there are 10 × 4 = 40 simple one-headline/one-description pairs. Three examples here are unpinned, contributing 120 pairs. The plumbing set has two eligible H1 pins and four descriptions, contributing 2 × 4 = 8. The bottle set has two eligible H1 pins and one eligible D1 pin, contributing 2 × 1 = 2. That makes 130 minimum pairs across the five teaching sets.

This is a local structural count, not a claim that Google will serve all 130, serve them equally, or consider them all eligible in a live account. It is still useful: a human can read 130 short pairs and catch unsupported promises, incomplete grammar, mismatched CTAs, or qualifiers that vanish. The same exercise becomes less tractable if you begin with every possible three-headline/two-description ordering.

Use this order:

  1. Lock the fact contract. Record service area, offer, price basis, restrictions, next action, and evidence for any outcome claim.
  2. Count the fields mechanically. These examples use ASCII text and count spaces and punctuation. Google notes that double-width-language characters are counted differently.
  3. Read every eligible minimum pair. Each pair should identify enough of the offer to remain honest and actionable.
  4. Check selected larger renders. Look for repeated CTAs, contradictory formats, duplicate ideas, and grammar that only works in one order.
  5. Open the wider asset pool. Inspect campaign-level assets, other active RSAs, associated assets, text customization, Final URL expansion, URL inclusions, pins, and schedules.
  6. Check the real destination. Google’s policies prohibit misleading or materially incomplete claims, and its destination rules require the ad to accurately reflect where a click leads. See Misrepresentation and Destination requirements.
  7. Use the account preview and policy review, but do not confuse them with performance evidence. The platform can reveal formatting or eligibility issues that a local worksheet cannot.

Your serving pool may be wider than your RSA table

A manual RSA can be internally coherent and still lose that coherence when account settings change.

First, Google lets advertisers associate up to three headlines and two descriptions at the campaign level, applying them to all enabled RSAs in that campaign. If an ad-level asset and a campaign-level asset are pinned to the same position, all same-position candidates can be eligible. A “must-show” Description 1 is therefore only as safe as the complete set of Description 1 candidates. Review the campaign-level asset rules.

Second, enhanced flexibility can use unused RSA headlines as link-based assets. Within one ad group, text borrowed from another active RSA points to the donor RSA’s final URL. That is useful flexibility, but it means “the final URL” is not always a single row in the current editor. The scheduling example above shows why a demo ad and a free-trial donor need a joint audit. Google describes the donor-URL behavior in its RSA help page.

Third, pinning depends on URL automation settings. Google says that when Final URL expansion selects another URL—or when URL inclusions are enabled—pinned RSA assets are not respected. To always use the pinned assets, its current instruction is to turn off both settings. Final URL expansion is a campaign-level option within AI Max and is enabled by default when an advertiser opts in to AI Max. See Google’s current Final URL expansion documentation.

The practical rule is simple: rerun the copy audit whenever the pool or destination logic changes, even when the 10 headlines and four descriptions remain untouched.

Measure a pool change, not a mythical winning headline

Ad Strength is creation feedback about the quantity, diversity, and relevance of available assets. Google explicitly says it does not determine serving eligibility and is not used to calculate Ad Rank, Quality Score, or auction wins. That makes it a useful diagnostic, not proof that one set will produce a business result. Google defines the distinction here.

Asset-level statistics need similar restraint. Google now exposes clicks, conversions, conversion value, cost, and impressions for responsive search ad assets, but calls the metrics non-summable and asset ratios directional. If one ad impression contains three assets, each asset records an impression: 100 ad impressions can therefore sum to 300 asset impressions. CTR, CPA, and ROAS at asset level are influenced by the other assets served alongside that asset, so they do not isolate one headline’s causal effect. See Google’s asset-level metrics guidance.

A better test changes one decision at a time. For the scheduling example, the completed test brief would be:

Test field Decision
Control Request a Product Demo — 22 characters
Variant Explore Our Scheduling Demo — 27 characters
Business question Does the CTA change improve cost per qualified demo request on the same demo journey?
Primary outcome Cost per qualified demo request
Secondary outcomes CTR and form completion rate
Held constant Other assets, pins, landing page, targeting, bidding, conversion definition, and automation context
Platform route Google Ads Ad Variations with a proposed 50/50 experiment split
Status Designed only; not created or run

Google Ads’ current Ad Variations workflow supports headline or description text changes and an experiment split; its documentation says cookies are used to help keep users in one version. That is a platform capability, not evidence that this unrun CTA variant will win. See the setup instructions.

The reusable lesson behind all five examples

The best responsive search ad examples are not collections of clever lines. They are decision systems. The copy starts with a narrow, explicit fact contract. Each asset contributes a standalone role. Critical information is not hidden in an optional third headline, second description, or path. Pins are used only when a condition truly must survive—and then the wider pool and URL settings are audited so the condition is not accidentally bypassed.

Copy the structure, not the fictional claims: define what is true, write one coherent conversion path, count the fields, read the smallest versions, inspect outside assets and destinations, then measure the whole change against a business outcome. That process produces fewer screenshots worth admiring—and far more ads worth trusting.

Sources

Back to blog

Keep reading

Turn the idea into a playable

Build and test an interactive ad in Hookin. No code required.

Start free