Google Ads Starter FREE Course · Day 6 of 7 · Article 2 of 3
Ads and landing pages
Seven days, in the order a first account actually gets built. Every lesson is a full reference article with step-by-step blocks you can follow with Google Ads open in another tab.
This day's reading:
- Headlines and Descriptions
- Google Ads Assets and Extensions: Sitelinks, Callouts and More (you are here)
- Google Ads Landing Pages: How To Build One That Converts
All seven days
- Is Google Ads worth it for you? The maths
- How the auction works, and setting up without the traps 2 articles
- Measurement before money 3 articles
- Keywords and intent 3 articles
- Structure, budgets and bidding at launch 3 articles
- Ads and landing pages (you are here) 3 articles
- Launch, the first 30 days, and when to stop DIY-ing 2 articles
Prefer it in your inbox? Get the 7-day course by email — one lesson a day, no pitch until day seven.
Sitelinks, callouts, snippets, images, calls and prices make your ad bigger and more relevant without raising your bid. Which assets to add, how many, at which level, and how to tell the ones doing the work from the passengers.
Last updated: October 22, 2026
In This Guide
- Executive Summary
- Benchmarks & Numbers at a Glance
- How Assets Affect Ad Rank
- Sitelinks
- Callouts & Structured Snippets
- Image Assets
- Call, Lead Form & Location Assets
- Price & Promotion Assets
- Automated vs Manual Assets
- Asset Level Strategy
- Reporting & Optimization
- Common Mistakes to Avoid
- What Changed Recently
- References
1. Executive Summary
Google Ads assets (formerly called extensions) are supplemental pieces of information attached to a Search, Performance Max, or Demand Gen ad. They expand the ad’s footprint, add interactive paths, and contribute to Ad Rank through the expected impact of ad formats. Every senior marketer responsible for paid search in 2026 should internalize five principles.
- Assets are a system, not a checklist. Google selects asset combinations dynamically based on query relevance, device, Ad Rank, and predicted performance. Build a broad, high-quality library across sitelinks, callouts, structured snippets, images, calls, prices, and promotions so the system always has strong candidates to choose from.[1]
- Quantity without quality degrades performance. Google’s guidance is explicit: add assets only when they are accurate, relevant, and genuinely useful. Near-duplicate callouts, irrelevant sitelinks, and expired promotions waste the system’s selection budget and can mislead users.[2]
- Assets affect Ad Rank. The expected impact of assets is a named component of Google’s Ad Rank formula. Relevant, high-quality assets improve eligibility and can allow your ad to appear in positions or formats that a competitor with a higher bid but weaker assets cannot access.[1]
- Evaluate assets on business outcomes, not CTR alone. CTR is visible and tempting to optimize, but a high-CTR sitelink that attracts unqualified clicks is a cost center. Report on conversions, cost per qualified lead, revenue, and—where measurable—offline outcomes alongside CTR.[1]
- Governance is non-negotiable in 2026. Automated asset creation, generative AI asset generation, and the December 4, 2026 default opt-in for Performance Max image extraction mean that ads can now include content the advertiser did not manually write or approve. Build a weekly review cadence and know how to opt out of automated behaviors that conflict with brand or legal requirements.[3]
2. Benchmarks and Numbers at a Glance
| Metric | Typical range or threshold | Applies when | Source |
|---|---|---|---|
| Sitelink CTR uplift — non-branded searches | 10–20% relative increase | Vendor claim; no published sample size or date range stated | [4] |
| Sitelink CTR uplift — branded searches | 20–50% relative increase | Vendor claim; uplift is larger on branded queries where navigational intent is strong | [4] |
| Sitelink CTR uplift — average across query types | ~20% relative increase | Vendor claim published June 29, 2026; no sample size stated | [4] |
| Callout CTR uplift | ~10% relative increase | Vendor claim published June 29, 2026; no sample size stated | [4] |
| Structured-snippet CTR uplift | ~9% relative increase | Vendor claim published June 29, 2026; no sample size stated | [4] |
| Combined sitelink + callout + structured-snippet CTR uplift | >20% aggregate relative increase | Vendor claim published June 1, 2026; when all three asset types are active simultaneously | [5] |
| Image asset CTR uplift — mobile traffic | ~5% relative increase | Study/case study; mobile Search placements specifically | [6] |
| Image asset CTR uplift — single-advertiser case | 44% relative increase; final CTR 8% | One advertiser, three-month window; published August 2024, updated October 16, 2026; not generalizable | [7] |
| Image asset conversion-rate improvement — single-advertiser case | 128% relative improvement; final conversion rate 16% | Same one-advertiser, three-month case study as above; treat as directional, not a universal benchmark | [7] |
| Full asset implementation CTR effect (60-day period) | 31% relative increase after 60 days | Vendor claim/case study published June 29, 2026; one advertiser; not independently verified | [4] |
| Full asset implementation CPA effect (60-day period) | 18% relative CPA decrease after 60 days | Same one-advertiser case study as above; use as directional planning input only | [4] |
| Real-staff or local-landmark images vs. generic stock images — CTR | 20% relative CTR increase | Vendor claim/case study; one regional health clinic; published May 2026, updated July 2026 | [8] |
| Real-staff or local-landmark images vs. generic stock images — conversion rate | 15% relative conversion-rate increase | Same single health-clinic case study; treat as a directional hypothesis to test, not a guaranteed lift | [8] |
| Minimum sitelinks needed for display (desktop and mobile) | 2 active sitelinks | Google serving requirement; display is still subject to Ad Rank and available space | [9] |
| Maximum sitelinks associated per entity | 20 per account, campaign, or ad group | Google platform limit; Google recommends at least 6 with descriptions for important campaigns | [9] |
| Minimum callouts needed for display | 2 active callouts | Google serving requirement; up to 4 may show at once | [2] |
| Performance Max asset experiment — recommended minimum wait before replacing assets | 2–3 weeks | Google’s own guidance for PMax asset groups before swapping low-performing assets | [10] |
3. How Assets Affect Ad Rank and Performance
Google’s Ad Rank formula has five named inputs: bid, ad quality and expected click-through rate, landing-page experience, Ad Rank thresholds, and the expected impact of assets and other ad formats.[1] Assets are not cosmetic additions bolted onto a finished ad—they are evaluated before every auction to determine whether including them will make the ad more useful to the user searching at that moment.
The mechanism works as follows. Before each auction, Google’s systems estimate how much each available asset combination is likely to improve the ad’s usefulness for the specific query, device, location, and user context. That expected impact is factored into Ad Rank alongside the bid and quality signals. An advertiser with a lower maximum CPC bid but a richer, more relevant asset library can therefore achieve a higher Ad Rank—and win a better position or additional formats—than a competitor bidding more but carrying sparse or irrelevant assets.[1]
Two important corollaries follow. First, adding assets does not guarantee they will appear. If Ad Rank is insufficient, or if Google does not predict a benefit for a given query, assets will not serve regardless of how many are attached. Second, the system evaluates asset combinations, not individual assets in isolation. A sitelink pointing to a relevant service page contributes more expected impact when paired with accurate callouts and a high-quality landing page than when the rest of the ad is weak.[1]
Assets also affect performance through the physical space they occupy. Ads with sitelinks, images, or prices occupy more vertical or horizontal real estate on the results page, which can increase organic visual attention and reduce the probability that a user scrolls past the ad without noticing it.[5]
Worked example (illustrative)
Ad Rank Gap Closed by Asset Library Depth
- Setup: A Chicago personal-injury law firm spending $22,000 per month on Search. Competitor A bids $85 max CPC. This firm bids $70 max CPC but carries 8 sitelinks with descriptions, 8 callouts, 2 structured-snippet headers with 5 values each, a call asset with dayparting (8 a.m.–6 p.m. CT), and 4 image assets.
- Numbers: Vendor claim data indicates sitelinks alone contribute a ~20% relative CTR uplift on non-branded queries[4], and combined sitelink + callout + structured-snippet deployment correlates with >20% aggregate CTR uplift[5]. If the firm’s baseline Quality Score equivalent supports a quality-adjusted bid of $70 × 1.20 = $84 effective rank-equivalent before assets, the additional expected impact from a complete, relevant asset library can close or exceed the $15 gap against the $85 competitor bid.
- Decision: Maintain the full asset library at campaign level and keep minimum active sitelinks at 8; do not reduce to the 2-sitelink serving minimum to “simplify” the account.
- Why: Google’s Ad Rank formula explicitly credits expected asset impact, meaning a richer relevant asset set can functionally raise rank above a higher-bidding competitor with fewer assets.[1]
4. Sitelinks
Sitelinks are the most widely used and highest-impact asset type in Search advertising. Each sitelink consists of link text (up to 25 characters in most languages; 12 characters in double-width languages such as Chinese, Japanese, and Korean), an optional but strongly recommended pair of description lines (up to 35 characters each), and a destination URL.[9]
Google can display up to 6 sitelinks on desktop—either in rows or side by side—and up to 8 in a carousel format on mobile.[9] A minimum of 2 active sitelinks is required for display, but display is not guaranteed even when 2 are present; Google also requires sufficient Ad Rank and must predict that sitelinks will improve the user experience for the specific query.[9]
Quantity and Structure Recommendations
For any campaign that drives meaningful revenue, attach at least 6 sitelinks with complete descriptions.[9] This provides Google with enough candidates to populate the full desktop row format and to select the most relevant subset for each auction. Up to 20 sitelinks can be associated with a single account, campaign, or ad group.[9]
Build the sitelink hierarchy in three layers. At the account level, add universal destinations that apply to every campaign: Contact Us, About Us, Careers, and any site-wide trust page.[9] At the campaign level, add product-line, service, or geography-specific destinations. At the ad-group level, use sitelinks only when a tighter keyword theme genuinely justifies a distinct landing page that is not appropriate for the broader campaign.[2]
Copy and Landing-Page Alignment
Each sitelink must lead to a distinct, useful page that differs meaningfully from the ad’s main landing page and from every other sitelink destination. Labels like “Pricing,” “Book a Demo,” “Running Shoes,” and “Store Locations” work because they communicate a clear next action and set an accurate expectation of what the user will find.[9] Avoid sitelinks that point to the homepage, duplicate other sitelink destinations, or carry vague labels like “Learn More.”
Dynamic sitelinks—automatically generated by Google—should remain enabled unless there is a documented brand or legal reason to suppress them. Google can serve them alongside or instead of manually created sitelinks, which expands coverage in auctions where manually created options are less relevant.[9]
Worked example (illustrative)
Sitelink Setup for a Roofing Campaign Targeting Branded and Non-Branded Queries
- Setup: A Nashville roofing contractor spending $8,500 per month on Search with two campaigns: one branded (targeting the company name) and one non-branded (targeting “roof replacement Nashville” and related terms).
- Numbers: Vendor claim data shows sitelinks produce a ~50% relative CTR uplift on branded queries and ~10–20% on non-branded queries.[4] The branded campaign generates 1,200 impressions per month at a 12% baseline CTR (144 clicks). A 50% uplift would add ~72 clicks at a $4.20 average CPC, worth approximately $302 in incremental traffic value per month. The non-branded campaign generates 4,800 impressions at 6% baseline CTR (288 clicks); a 15% midpoint uplift adds ~43 clicks at $9.80 CPC = ~$421 additional monthly value.
- Decision: Apply 8 campaign-level sitelinks to the non-branded campaign (Book a Free Estimate / 25 char; Roof Replacement / 16 char; Storm Damage Repair / 19 char; Financing Options / 18 char; Our Warranties / 15 char; See Our Work / 12 char; About Our Crew / 14 char; Service Areas / 13 char), each with two 35-character description lines, distinct landing pages, and no destination duplicating the main ad URL.
- Why: Six sitelinks with descriptions is Google’s recommended minimum for high-volume campaigns, and the branded/non-branded CTR uplift differential justifies maintaining separate sitelink sets at campaign level rather than relying solely on account-level assets.[9]
Worked example (illustrative)
Fixing a Mobile Sitelink Truncation Problem
- Setup: A Phoenix plumbing company running Search ads with sitelinks averaging 28 characters of link text. The account manager notices that mobile impression share is 62% of total impressions but mobile CTR is 1.4 percentage points below desktop CTR.
- Numbers: Google’s character limit for sitelink text is 25 characters in most languages.[9] Of 10 active sitelinks, 4 exceed 25 characters: “Emergency Plumbing Services” (27 chars), “Water Heater Installation” (24 chars — compliant), “Drain Cleaning Specialists” (25 chars — compliant), “Schedule a Service Visit Today” (30 chars — violates limit). Google may truncate or suppress non-compliant sitelinks, reducing carousel completeness on mobile where up to 8 sitelinks can show.[9]
- Decision: Rewrite the 4 non-compliant sitelinks to 25 characters or fewer: “Emergency Plumbing” (18), “Drain Specialists” (17), “Book a Service Visit” (20), “Water Heater Repair” (19). Add description lines of 35 characters or fewer to each.
- Why: Sitelink text exceeding 25 characters risks suppression or truncation, directly reducing the number of sitelinks eligible to appear in the mobile carousel format that shows up to 8.[9]
Managed for you
Everything on this page, handled every month
You are reading how to do it properly. Omologist does it, inside a written mandate, for 15% of your budget.
✓ Conversion tracking built and tested before any ad runs
✓ Search campaigns built to what your budget can actually feed
✓ Changes made inside a written mandate, not whenever
✓ A report on the 1st that says what was done and why
15% of budget plus GST · minimum A$330 incl GST · accounts under about A$10,000 a month
5. Callouts and Structured Snippets
Callouts and structured snippets are both non-clickable, informational asset types that expand the ad’s content without adding a navigation destination. They serve complementary roles: callouts communicate differentiating benefits and selling points, while structured snippets enumerate specific offerings organized under a predefined header category.[2]
Callout Assets
Each callout is a short phrase of up to 25 characters.[2] At least 2 must be active for serving; up to 4 may appear in a single ad impression depending on available space.[2] The published vendor claim benchmark for callouts is a ~10% relative CTR uplift.[4]
Effective callouts name specific, factual advantages: “Free 2-Day Shipping,” “No Setup Fees,” “24/7 Expert Support,” “A+ BBB Rating,” “Licensed & Insured,” “Price-Match Guarantee.” Each should communicate something that is true, unique to the business or campaign context, and not already stated in the headline, description, sitelink, or structured snippet. Build a library of 8–12 distinct callouts so Google has alternatives across auctions; do not inflate the library with near-duplicates.[2]
Callouts can be set at the account, campaign, or ad-group level. Place genuinely universal claims (Free Shipping, Licensed & Insured) at the account level; place campaign- or product-specific claims (0% Financing Through December 2026, Same-Day Installation Available) at campaign or ad-group level.[2]
Structured Snippet Assets
Structured snippets require a predefined header (such as Services, Types, Brands, Styles, or Destinations) and a list of values beneath it.[11] Google recommends at least 4 values per header.[2] Each value should be kept to 25 characters or fewer, and for mobile-focused campaigns, approximately 12 characters or fewer improves display reliability.[2] The published vendor claim benchmark for structured snippets is a ~9% relative CTR uplift.[4]
Values must be concrete and categorical—product names, service types, brand names, destinations, model numbers—not promotional language. “Services: Installation, Maintenance, Repairs, Consulting” is correct. “Services: Best Prices, Free Quotes, Trusted Experts” is incorrect; those belong in callouts.[11]
Add multiple header/value sets when the business genuinely spans several categories. Google selects the most relevant set for each auction.[11]
Worked example (illustrative)
Callout and Structured-Snippet Build for an HVAC Company
- Setup: A Denver HVAC company spending $6,000 per month on a Search campaign targeting heating, cooling, and indoor-air-quality keywords. The account currently has 3 callouts and no structured snippets.
- Numbers: With only 3 callouts, the account meets the 2-callout serving minimum but does not provide Google with meaningful alternatives.[2] The vendor claim benchmark shows callouts contribute ~10% relative CTR uplift and structured snippets ~9%.[4] At a current CTR of 5.2% on 9,000 impressions per month (468 clicks at $7.50 CPC = $3,510 in click spend), a combined 19% relative uplift from adding both asset types would project to approximately 5.2% × 1.19 = 6.19% CTR, yielding ~557 clicks — 89 additional clicks at $7.50 = ~$668 in additional traffic value per month. Treat as directional; actual results depend on ad quality and auction context.
- Decision: Add 9 callouts (Licensed & Insured / 20 chars; 24/7 Emergency Service / 21 chars; Free In-Home Estimates / 22 chars; NATE-Certified Techs / 20 chars; No Overtime Charges / 19 chars; Financing Available / 20 chars; Satisfaction Guarantee / 22 chars; Same-Day Service / 16 chars; Family-Owned Since 1998 / 23 chars) and 2 structured-snippet headers: “Services: AC Repair, Furnace Install, Duct Cleaning, Heat Pumps, Air Scrubbers” and “Brands: Carrier, Trane, Lennox, Daikin, Rheem” — each value under 25 characters.
- Why: Google recommends at least 4 values per structured-snippet header and a library of more than 4 callouts so the system has selection flexibility across auctions; combined deployment of both asset types is associated with >20% aggregate relative CTR uplift in vendor data.[5]
Worked example (illustrative)
Separating Promotional Language from Informational Snippets
- Setup: A Seattle software-as-a-service company has structured snippets set to header “Services” with values: “Best ROI,” “Trusted by Thousands,” “Limited-Time Deal,” “Award-Winning.” All 4 values are promotional rather than categorical.
- Numbers: Google’s guidance requires structured-snippet values to be concrete offerings under the chosen header, not promotional claims.[11] Promotional values create a policy-compliance risk and reduce relevance, potentially suppressing the asset. The 4 values also exactly meet the 4-value minimum; if any is disapproved, the asset drops below the serving threshold.
- Decision: Replace all 4 values with categorical service names under 25 characters: “CRM Integration,” “Workflow Automation,” “Analytics Dashboard,” “API Access,” “SSO & Permissions.” Move “Trusted by Thousands,” “Award-Winning,” and any time-bound incentive to callouts. Increase the snippet library to 6 values for resilience against individual disapprovals.
- Why: Structured snippets must list factual product or service types; promotional language is a callout function, and misclassifying it risks asset disapproval and reduces serving eligibility below the 4-value recommended minimum.[2]
6. Image Assets
Image assets allow eligible Search and Performance Max ads to include a visual alongside text, increasing the ad’s physical footprint on the results page and communicating product, environment, team, or outcome information that words alone cannot convey efficiently.[12]
Specifications
| Format | Aspect ratio | Recommended resolution | Minimum resolution | Max per format (Search) |
|---|---|---|---|---|
| Square | 1:1 | 1200 × 1200 px | 300 × 300 px | 20 |
| Landscape | 1.91:1 | 1200 × 628 px | 600 × 314 px | 20 |
For Performance Max, Google’s current guidance recommends at least 7 image assets, including at least one 1200 × 1200 square image.[10] For Search campaigns, the minimum effective setup is 4 unique images including both square and landscape formats.[12]
Keep the primary subject centered so it survives cropping across different placements and devices. Avoid images dominated by text, heavy filters, blank space, or distorted subjects. Use high-resolution, sharp imagery—not scaled-up low-resolution assets. Prioritize images that are specific to the product, service, team, facility, or outcome being advertised rather than generic stock imagery.[12]
What to Test
Create genuinely different visual concepts rather than minor crops of the same image. Productive test categories include: product-in-use, close-up product detail, before-and-after outcome, environment or facility interior, real staff or team members, and branded location exterior. A vendor claim from a regional health clinic study found that real-staff and local-landmark images produced a 20% relative CTR increase and a 15% relative conversion-rate increase versus generic stock photos.[8] A published study of one personal-injury-law advertiser showed CTR increasing from 9.89% and 12.07% (across two campaigns) to 16.16% after adding image assets.[6] Treat single-case results as hypotheses to test, not guaranteed outcomes.
Dynamic image assets—where Google automatically selects images from landing pages—should be evaluated against manually uploaded images. They are useful for coverage but require governance: verify that landing-page images pulled automatically are brand-appropriate, current, and policy-compliant.[12]
Performance Assessment
Use CTR, conversion rate, cost per qualified lead, and (where available) offline outcome as evaluation criteria. Do not suppress an image asset solely because it has fewer impressions than another; Google serves images selectively, and impression count reflects serving eligibility as much as quality.[12]
Worked example (illustrative)
Image Asset Testing Plan for a Home Services Company
- Setup: An Atlanta home-remodeling company spending $14,000 per month on Search. The account has 1 landscape image (a generic kitchen stock photo at 1200 × 628 px) and no square images. Image CTR is 3.1% versus a text-only ad CTR of 3.0% — essentially flat, suggesting the current image is adding no incremental value.
- Numbers: Google recommends at least 4 unique images including both formats.[12] A vendor claim from one case study shows a 44% relative CTR increase when relevant images are used (final CTR of 8% versus a lower baseline).[7] Even at the more conservative 5% relative mobile CTR uplift benchmark[6], moving from 3.1% to 3.255% across 18,000 monthly impressions adds ~279 clicks. At a $6.50 average CPC, that is ~$1,814 in additional traffic value per month if conversion rate is maintained. The 128% conversion-rate improvement in the single-case study[7] is not used in planning here because the single-case provenance makes it unsuitable for projection.
- Decision: Upload 6 new images by November 1, 2026: 2 square (1200 × 1200 px) showing real completed kitchens with staff visible; 2 landscape (1200 × 628 px) showing before-and-after bathroom remodels; 1 square showing the company crew in branded uniforms; 1 landscape showing the showroom interior. Retire the generic stock image. Run for a minimum of 21 days before comparing performance labels.
- Why: Real-staff and authentic-project imagery is associated with a 20% relative CTR uplift versus generic stock images in vendor data[8], and Google requires at least 4 unique images including both aspect ratios to give the system adequate creative choice.[12]
7. Call, Lead Form and Location Assets
These three asset types share a common purpose: they add a direct action path that does not require the user to navigate a full website before taking the next step. For many US local businesses, calls and directions are the primary conversion, making this asset group the most economically significant in the account.
Call Assets
A call asset displays a clickable phone number directly within the ad on mobile devices, and a visible number on desktop. Google recommends combining a call asset with a location asset for businesses where users may visit a physical location but should reach one central number.[2]
Implementation requirements for a high-performing call asset setup:
- Use a valid business number capable of handling the advertised service type.
- Enable Google call reporting and set a minimum call duration (commonly 60–90 seconds) as a conversion action so that connected, substantive calls are counted—not misdials.
- Apply dayparting so the call asset serves only during staffed hours. A call asset serving at 11 p.m. for a business that answers at 8 a.m. wastes budget on missed calls.
- Use unique tracking numbers at the campaign or service level when attribution across multiple locations or service lines is needed.
- Do not use call-only ads for new campaigns; Google deprecated the creation of new call-only ads, and existing call-only ads are scheduled to stop serving in February 2027. Build Responsive Search Ads with call assets instead.[13]
Lead Form Assets
Lead form assets capture prospect information directly within the Google Ads interface, bypassing the website landing page. They are effective for quote requests, demo bookings, consultation inquiries, and B2B contact forms where the prospect can submit enough information without visiting the site.[14]
Design principles for high-quality lead form performance:
- Use a benefit-driven headline that states what the prospect receives: “Get a Free Roof Inspection Quote,” not “Contact Us.”
- Request only the minimum fields needed for qualification and follow-up. Each additional required field reduces completion rate.
- Add 1–2 qualifying questions when lead quality matters more than raw volume.
- Include the privacy-policy URL and ensure consent language meets applicable US state requirements (CCPA, state-level equivalents).
- Connect leads to a CRM immediately; assign a response-time SLA. Lead value decays rapidly with delayed follow-up.
- Pass closed-lead outcome data back to Google Ads as offline conversions where possible, so Smart Bidding can optimize toward qualified leads rather than all submissions.
Location Assets
Location assets link a Google Business Profile to the ad, surfacing the address, map pin, distance, and phone number for users nearby.[12] They are essential for any business where a physical visit is a desired outcome. Location assets support store-visit conversion measurement and can appear on Google Maps placements.[12]
Worked example (illustrative)
Call Asset Dayparting and Conversion Threshold Setup
- Setup: A Houston urgent-care clinic spending $9,200 per month on Search. The clinic is staffed 7 a.m.–9 p.m. CT, Monday through Sunday. The account has a call asset active 24 hours per day with no call-duration conversion threshold set. The call report shows 340 calls per month, but the clinic’s front desk estimates fewer than 180 are answered or represent genuine appointment intent.
- Numbers: 340 reported calls − 180 qualified = 160 wasted call interactions. At an average CPC equivalent for call clicks of approximately $11.40, wasted calls represent approximately $1,824 per month in budget with no return. Restricting serving to 7 a.m.–9 p.m. removes 14 hours of daily overnight exposure. Setting a 45-second minimum call duration filters misdials and very short disconnects.
- Decision: Set call asset schedule to 7:00 a.m.–9:00 p.m. CT, 7 days per week. Create a call conversion action with a minimum duration of 45 seconds. Enable Google call reporting for all call asset interactions. Review the first 30 days of post-change data (by October 31, 2026) against the prior 30 days for qualified call volume and cost per qualified call.
- Why: Call assets serving outside staffed hours generate unanswered calls that consume budget without producing conversions; Google’s guidance is to daypart call assets to periods when calls can be answered.[2]
Worked example (illustrative)
Lead Form Asset Qualification Setup for a B2B SaaS Account
- Setup: A Boston B2B SaaS company spending $18,000 per month on Search targeting “enterprise project management software” and related terms. The current lead form has 3 fields (name, email, company) and no qualifying questions. The sales team reports that 55% of leads submitted through the form do not meet the minimum employee-count threshold of 50 employees.
- Numbers: Monthly form submissions: 210. Qualified leads (≥50 employees): 210 × 0.45 = 94.5, rounded to 95. Cost per qualified lead: $18,000 ÷ 95 = $189.47. Adding one qualifying question (“How many employees does your company have? / Under 25 / 25–49 / 50–199 / 200+”) and showing the form only to users selecting 50+ is projected to reduce total submissions while increasing the qualified proportion. If total submissions fall to 130 but 80% are qualified: 130 × 0.80 = 104 qualified leads. Cost per qualified lead: $18,000 ÷ 104 = $173.08 — a $16.39 reduction per qualified lead despite lower raw volume.
- Decision: Add the single employee-count qualifying question to the lead form asset, set to required. Update the thank-you message to confirm a 1-business-day response SLA. Pass qualified-lead status to Google Ads as an offline conversion within 7 days of CRM entry.
- Why: Adding one qualifying question filters under-threshold prospects at the form level, reducing cost per qualified lead from $189.47 to a projected $173.08 — a 8.7% improvement — while maintaining sufficient qualified volume for the sales team’s pipeline targets.[14]
Not ready to change anything today?
We post the sharpest bits of these guides — what a change is worth, how long it takes to show up, and when to leave it alone. Short enough to read on your phone.
Follow on LinkedIn Follow on Instagram8. Price and Promotion Assets
Price and promotion assets both serve a pre-qualification function: they communicate cost or offer information before the click, so users who see the ad already know the approximate price or discount when they arrive at the landing page.[12] The practical consequence is that these asset types may reduce raw CTR while improving conversion rate and cost per qualified sale—which is the correct trade-off when the objective is revenue efficiency rather than traffic volume.
Price Assets
Price assets display multiple products, services, or categories with associated prices in a horizontal, card-style format beneath the ad.[12] They are well-suited to recurring services with defined packages, product categories with distinct price tiers, memberships, classes and courses, and subscription products.
Implementation rules that directly affect serving and policy compliance:
- Display prices that are accurate at the time of serving. Price assets that mismatch the landing page violate Google’s policy and damage user trust.
- Use “starting at” or “from” language only where the actual price may vary; do not display a starting price that no product in the category actually reaches.
- Make currency (USD), billing frequency (per month, per year, per session), and any required minimum spend or package explicit within the asset.
- Update prices immediately when offers change. A stale price asset is worse than no asset because it creates a post-click expectation gap.
- Add multiple categories rather than a single price point; Google needs at least 3 price items to serve the asset.[15]
Outside the US: price assets must comply with currency display rules, consumer-protection regulations, and distance-selling requirements applicable in each country where the campaign serves. In the EU and UK, “from” pricing claims are subject to stricter proof-of-availability requirements than in the US.
Promotion Assets
Promotion assets highlight a specific, time-limited incentive with a prominent visual treatment in the ad.[16] They work best when the offer is materially better than the standard proposition—a meaningful percentage discount, a free item, a promotional code, or a new-customer incentive—and when the offer is active throughout the asset’s scheduled serving window.
Key implementation requirements:
- State the actual offer value: “25% off all HVAC tune-ups” rather than “Save big.”
- Use a start date and end date; schedule the asset to stop serving on or before the offer expiration date.
- List the applicable product, minimum spend, and any exclusions required to avoid a misleading-offer policy violation.
- Match the promoted offer exactly on the destination landing page and at checkout. Discrepancies between the asset and the page are a common disapproval cause.
- Use occasion labels (Black Friday, Holiday, Back to School, Summer Sale) only when the promotion genuinely aligns with that event.
- Do not use promotion assets as permanent fixtures for a standard price; they are intended for genuinely temporary incentives.[16]
Worked example (illustrative)
Promotion Asset Scheduling for a Black Friday HVAC Campaign
- Setup: A Minneapolis HVAC company running a Search campaign with a 20%-off furnace-tune-up promotion from November 24 through November 30, 2026 (Black Friday week). The offer applies to standard tune-ups priced at $149; the discounted price is $119.20. The company has a promotion asset with no end date set, and last year the offer continued serving through December 14 without the discount being honored at the door.
- Numbers: In 2025, the expired promotion asset served for 14 additional days beyond the offer end. Estimated impressions during the overage period: 6,800. Estimated CTR with promotion asset active: 7.4%. Estimated clicks after expiry: 6,800 × 0.074 = ~503 clicks at $8.10 CPC = $4,074 in spend. Calls from users expecting $119.20 but charged $149 generated 23 complaint calls documented by the front desk. In 2026, the promotion asset will have an explicit end date of November 30, 2026, set within the asset settings.
- Decision: Create promotion asset with: occasion label “Black Friday,” discount type “Percent off,” discount value “20%,” item “Furnace Tune-Up,” promotion code field left blank (no code required), start date November 24, 2026, end date November 30, 2026. Verify landing page shows $119.20 price from November 24 through November 30 only. Set a calendar reminder for December 1, 2026 to confirm asset has stopped serving and revert landing-page pricing to $149.
- Why: Google requires that promotion assets accurately reflect the active offer at the time of serving; an expired promotion asset that still serves constitutes a misleading-ad policy violation and generates qualified-traffic waste.[16]
9. Automated vs. Manual Assets
As of October 2026, Google’s asset ecosystem operates on a controlled-automation model: manual assets establish the strategic foundation, and automated assets extend coverage in auctions where manually created options are less relevant or where Google’s systems predict a benefit from algorithmically generated content.[17]
Manual Assets: Role and Governance
Manual assets are the advertiser’s primary tool for ensuring that critical messaging—legal disclaimers, priority offers, strategic landing pages, brand-consistent copy—appears consistently and accurately. Every significant campaign should carry a complete set of manually created assets covering sitelinks, callouts, structured snippets, and the asset types appropriate to the campaign’s conversion actions.[1]
Practical controls:
- Audit manual asset text for accuracy, policy compliance, and landing-page alignment at least monthly.
- Schedule time-sensitive manual assets (promotion, price, seasonal callout) with explicit start and end dates.[2]
- For Performance Max, cover text, image, and video formats: up to 15 headlines, 5 descriptions, 20 images, and 5 videos per asset group.[10]
- Allow at least 2–3 weeks before replacing assets judged to be underperforming in a PMax asset group, giving Google’s system time to collect meaningful data.[10]
Automated Assets: Role and Governance
Automated assets are generated or selected by Google when its systems predict a performance benefit. Types include automatically created sitelinks, dynamically generated callouts, dynamic image assets, and—via the API—assets generated by the new AssetGenerationService introduced in API v22.[18] The December 4, 2026 default opt-in for Performance Max image extraction from landing pages is the most consequential recent automation change for most advertisers.[3]
Governance requirements for automated assets:
- Review automated assets weekly for brand safety, factual accuracy, tone, and destination quality. Google’s API supports retrieving statistics and pausing or removing individual automated assets.[17]
- Pause or remove automated assets that are outdated, off-brand, legally sensitive, or inconsistent with the campaign strategy.
- For Performance Max: before December 4, 2026, decide explicitly whether image extraction from landing pages is acceptable. If not, set
GENERATE_IMAGE_EXTRACTIONtoOPTED_OUTon each new PMax campaign. Note that this new campaign-level default can override the account-level dynamic-image setting.[3] - Treat API-level capabilities (text disclaimer assets, LINE/SMS business-message providers, generative asset creation) as requiring separate eligibility verification before production use; API availability does not equal universal advertiser UI availability.[18]
| Dimension | Manual assets | Automated assets |
|---|---|---|
| Advertiser control | Full: copy, destination, schedule, device | Partial: can pause or remove; cannot control generation triggers |
| Brand safety | High: reviewed before upload | Requires active governance; automated assets can surface outdated or off-brand content |
| Coverage | Limited to what the advertiser builds | Can fill gaps in auctions where manual assets are less relevant |
| Effort | Higher initial setup; ongoing maintenance | Lower initial effort; higher ongoing review burden |
| Legal compliance | Advertiser responsibility; fully visible pre-serve | Advertiser is still responsible; requires post-generation review |
| Performance Max image extraction default (from Dec 4, 2026) | N/A | Opted in by default for newly created PMax campaigns; must explicitly opt out if unwanted[3] |
Worked example (illustrative)
Auditing Automated Assets Before the December 4, 2026 Image-Extraction Default
- Setup: A Portland retail brand managing 4 Performance Max campaigns across apparel and footwear. The brand’s style guide prohibits imagery showing products without models. The December 4, 2026 PMax image-extraction default will automatically pull images from landing pages for any new campaign created on or after that date.
- Numbers: The brand plans to launch 2 new PMax campaigns on December 10, 2026 — a winter outerwear campaign and a gift-card campaign. Without action, both campaigns will default to
GENERATE_IMAGE_EXTRACTION: OPTED_INand may pull 1200 × 1200 and 1200 × 628 images directly from landing pages, including flat-lay product shots that violate the style guide. The 2 existing campaigns (created before December 4) retain their current configuration and are unaffected.[3] - Decision: For both December 10 campaigns, set
GENERATE_IMAGE_EXTRACTIONtoOPTED_OUTat campaign creation via the API or the campaign settings UI. Upload 7 manually approved model images per campaign (at least one 1200 × 1200 square) before launch. Add a note in the account change history on December 4, 2026 confirming opt-out status was reviewed for all future PMax campaigns. - Why: The December 4, 2026 default overrides the account-level dynamic-image setting, so account-level opt-outs are insufficient protection for new PMax campaigns created on or after that date.[3]
“So what do I actually have to do?”
Tell us the business and give us the account. We build the conversion tracking and prove it works with a test conversion before any ad runs, then launch the campaigns your budget can feed.
After that, changes happen inside a written mandate, so you know in advance what gets changed without asking and what never gets touched. A report on the 1st, a budget email around the 23rd, and nothing else unless it needs your yes or no.
How it works10. Asset Level Strategy: Account, Campaign, and Ad Group
Every asset in Google Ads is associated with a specific level of the account hierarchy: account, campaign, or ad group. Assets at a higher level are inherited downward—an account-level sitelink is eligible to serve with any ad in the account unless overridden or excluded at a lower level. Getting this structure right is the single most operationally important asset decision because misapplied levels are the root cause of irrelevant assets serving on unrelated campaigns.
Level-by-Level Framework
| Level | Best use | Asset types commonly applied | Main risk if misused |
|---|---|---|---|
| Account | Universal facts true across every campaign: company overview, contact page, trust signals, free shipping | Sitelinks (Contact Us, About Us, Careers), callouts (Licensed & Insured, BBB Accredited), location assets | An account-level sitelink pointing to a product page that does not exist in every campaign’s scope misleads users on unrelated campaigns |
| Campaign | Product line, service category, geography, or seasonal assets specific to the campaign’s scope | Sitelinks to category pages, structured snippets listing campaign-specific services, call assets with campaign-specific tracking numbers, promotion assets | Overlapping campaign-level assets can conflict; audit for duplication with account-level assets |
| Ad group | Keyword-theme-specific sitelinks or callouts where the landing page or benefit is meaningfully different from the campaign-level asset | Sitelinks to very specific landing pages, structured snippets for a narrow product set | Overly narrow asset sets reduce serving frequency; use ad-group level only when relevance gain is clear |
Google’s recommended starting point is to create sitelinks at the account level first to maximize coverage speed, then layer campaign-level assets for category or geographic differences, and use ad-group assets only when a specific keyword cluster justifies a unique destination or message.[9] The same principle applies to callouts and structured snippets.[2]
After building the initial structure, run an inheritance audit quarterly: confirm that every account-level asset remains accurate and relevant to every campaign in the account. A company that acquires a new product line or exits a service category must update account-level assets immediately, because outdated account-level assets serve across every eligible campaign by default.
Worked example (illustrative)
Three-Level Asset Architecture for a Multi-Location Medical Spa
- Setup: A medical spa brand with locations in Austin, TX and Miami, FL running Search campaigns: one brand campaign, one Austin-specific campaign, and one Miami-specific campaign. Monthly total spend: $31,000. Current setup: all sitelinks applied at account level, including “Miami Location” and “Austin Location” sitelinks that appear cross-campaign, causing irrelevant destination exposure.
- Numbers: The account-level sitelink “Miami Location” serves on the Austin campaign, which targets ZIP codes 78701–78753 (Austin metro). In a given month, the Austin campaign receives 5,200 impressions; Google estimates that irrelevant sitelinks reduce expected asset impact, potentially suppressing serving. Moving location-specific sitelinks to campaign level reduces the account-level sitelink count from 8 to 4 (Contact Us, Book a Consultation, Our Treatments, Financing Options) — all 4 remain within the 2-sitelink serving minimum at account level.[9]
- Decision: Move “Miami Location,” “Miami Specials,” and 2 Miami-specific structured snippets to the Miami campaign. Move “Austin Location,” “Austin Specials,” and 2 Austin-specific structured snippets to the Austin campaign. Retain 4 universal sitelinks and 6 universal callouts at account level. Set city-specific call assets with local tracking numbers at each city campaign level.
- Why: Assets must be accurate and relevant to every campaign where they may apply; account-level assets inherit downward and can serve in unintended contexts, reducing relevance and expected asset impact within Ad Rank.[9]
11. Reporting and Optimization
Asset reporting in Google Ads operates across three distinct layers: the account-wide asset library, campaign and ad-group associations, and individual asset performance labels. Effective optimization requires working across all three layers simultaneously rather than reacting to a single metric in isolation.[1]
To access asset-level data, navigate to the Assets section in the Google Ads interface and switch to the Associations view. This view shows which assets are attached to which campaigns or ad groups, their approval status, impression volume, clicks, CTR, conversions, and the relative performance label assigned by Google.[1] For Performance Max, use the asset_group_asset resource via the Google Ads API to retrieve asset-level clicks, impressions, conversions, and the performance_label field for assets of the same type.[19]
Performance Labels and What They Mean
Google assigns one of four relative performance labels to assets of the same type within the same asset group or ad group: Learning, Low, Good, and Best.[1] These labels are relative comparisons within an asset type, not absolute quality scores. A sitelink labeled “Low” is underperforming relative to other sitelinks in the same group — it does not mean the sitelink is broken or should be immediately deleted. Context, volume, and conversion data must inform the decision.
Do not compare performance labels across asset types. A callout and a sitelink have different functions, different click mechanics, and different display triggers; ranking them against each other produces misleading conclusions.[1]
Recommended Reporting Cadence
| Cadence | Actions | Primary signals |
|---|---|---|
| Weekly | Check disapprovals, broken destination URLs, expired promotion assets, unexpected automated assets, major impression drops | Policy status, serving status, asset source |
| Monthly | Compare CTR, conversion rate, cost per conversion, and impression volume by asset type and campaign; identify consistent “Low” performers with sufficient volume | CTR, conversions, cost per conversion, impression share |
| After a test | Allow at least 2–3 weeks and meaningful impression volume before replacing assets; avoid decisions based on fewer than 300–500 impressions per asset variant[10] | Statistical signal, conversion volume, cost per conversion |
| Quarterly | Full asset library audit: remove duplicates, retire obsolete messaging, refresh creative, verify account-level assets remain universally relevant | Full account asset inventory, Ad Strength diagnostics, landing-page alignment |
Metrics to Prioritize
CTR is a useful directional signal but must not be the sole optimization criterion. An asset that generates high CTR to a poorly converting landing page inflates cost without producing business outcomes. Prioritize the following hierarchy:[1]
- Primary: Conversions, conversion value, cost per conversion, return on ad spend
- Secondary: CTR, total clicks, interaction rate
- Diagnostic: Impressions, serving rate, performance label, Ad Strength
- Supplemental: Qualified lead rate, opportunity rate, revenue (from CRM integration or offline conversion import)
For accounts where call volume and store visits matter, import offline conversions and connect Google Ads to a CRM so that asset-level investment can be traced to downstream revenue, not just form submissions or call connections.[1]
API Reporting for Larger Accounts
For accounts with more than five active campaigns or more than 50 active assets, manual interface reporting becomes inefficient. Automate recurring asset reporting using the Google Ads API resources: asset_field_type_view for aggregated statistics across automated and manually added assets; campaign_asset, customer_asset, and ad_group_asset for association-level data; and asset_group_asset for Performance Max asset groups.[19]
Worked example (illustrative)
Monthly Asset Audit Identifying a High-CTR but Low-Conversion Sitelink
- Setup: A Chicago-based B2B software account spending $22,000 per month on Search campaigns. The account has 7 sitelinks attached at campaign level to its primary “Enterprise CRM” campaign. Monthly reporting reveals that the “Free Trial” sitelink has a 6.8% CTR — the highest among all 7 sitelinks — but a conversion rate of 0.4% (2 conversions from 500 clicks at $8.80 average CPC, totaling $4,400 in spend for 2 conversions, a cost per conversion of $2,200).
- Numbers: The campaign’s average cost per conversion across all sitelinks is $310. The “Free Trial” sitelink’s cost per conversion of $2,200 is 610% above the campaign average ($2,200 ÷ $310 = 7.1×). The sitelink’s landing page is a self-serve sign-up page requiring a credit card; the campaign targets director-level and above decision-makers who research enterprise solutions. The other 6 sitelinks average a combined 2.1% CTR but deliver conversions at $280–$340 each.[4]
- Decision: Pause the “Free Trial” sitelink and replace it with a “Request a Demo” sitelink pointing to a consultation-booking page. Set a 30-day review window starting November 1, 2026, with a minimum threshold of 400 impressions on the replacement asset before comparison.
- Why: A high-CTR asset that delivers a cost per conversion 7× above the campaign average is destroying profitability; conversion quality, not CTR, is the correct primary optimization criterion.[1]
Worked example (illustrative)
Using API Reporting to Catch an Expired Promotion Asset Serving Across 9 Campaigns
- Setup: A national e-commerce retailer spending $180,000 per month across 14 Search campaigns. A “Summer Sale — 30% Off” promotion asset was created in June 2026 with an end date of July 31, 2026. An automated weekly API export using
campaign_assetpulls all active asset statuses every Monday at 6:00 AM CT. - Numbers: The August 4, 2026 export flags the promotion asset as “Eligible — Serving” on 9 of 14 campaigns despite its stated end date having passed. Manually checking the interface confirms the asset was not given a hard end date within the platform — only a note in a spreadsheet. In the week of July 28–August 3, the expired promotion asset served 41,200 impressions and generated 620 clicks at $1.47 average CPC, totaling $911.40 in clicks leading to a landing page that no longer shows the 30% discount. Estimated wasted spend over 5 post-expiration days: $911.40, plus potential trust damage from mismatched ad-to-page experience.
- Decision: Pause the expired promotion asset immediately across all 9 campaigns on August 4, 2026. Update the standard operating procedure to require hard start and end dates on every promotion asset at creation, enforced by a weekly
campaign_assetAPI audit that flags serving promotion assets within 3 days of or beyond their end date.[19] - Why: Promotion assets without platform-enforced end dates continue serving after expiration, creating a misleading ad-to-page experience and wasting budget on traffic that cannot convert on the advertised offer.[1]
Segmenting Asset Performance by Device
Assets serve differently on desktop and mobile. Sitelinks can show up to 6 on desktop and up to 8 in carousel format on mobile.[9] Image assets are primarily a mobile and visual-placement phenomenon. Segment asset performance by device in the interface or via the API to identify whether a “Low”-labeled asset is underperforming universally or only on one device type. A callout performing well on desktop but poorly on mobile may reflect character-count issues rather than messaging weakness — revisit the copy length against the recommended 12-character mobile guideline for structured-snippet values.[2]
Ad Strength as a Diagnostic, Not a KPI
Ad Strength and asset coverage scores are useful for identifying gaps in creative variety but are not reliable proxies for business performance. An ad group with “Excellent” Ad Strength can still deliver a high cost per lead if the assets attract unqualified traffic. Use Ad Strength to diagnose missing asset types or insufficient variety, then validate improvements using conversion data.[20]
12. Common Mistakes to Avoid
The following errors appear consistently in mature Google Ads accounts and degrade both asset quality and campaign economics. Each represents a specific, correctable misconfiguration or process failure.
Mistake 1: Optimizing Assets on CTR Alone
CTR measures interest, not intent. Assets that generate high CTR to the wrong landing page, at the wrong price point, or for the wrong audience increase spend without increasing revenue. Always pair CTR with conversion rate and cost per conversion before making any asset decision.[1]
Mistake 2: Applying Fewer Than 4 Sitelinks
Google requires a minimum of 2 sitelinks for serving, but recommends at least 4, and best practice for important campaigns is 6–8 with descriptions.[9] Accounts running only 2–3 sitelinks reduce the system’s ability to select the most relevant combination for a given query, limiting coverage and Ad Rank contribution from assets.
Mistake 3: Using Duplicate or Near-Duplicate Callouts
Adding callouts such as “Fast Shipping,” “Quick Delivery,” and “Ships Fast” as three separate assets does not increase variety — it wastes the 4-callout display limit on redundant claims and reduces the information value of the asset group. Each callout must communicate a distinct benefit that does not appear in the headline, description, sitelink, or structured snippet.[2]
Mistake 4: Ignoring Asset-Level Policy Disapprovals
A disapproved asset silently reduces coverage. In accounts with large asset libraries, individual disapprovals can go unnoticed for weeks, particularly at the ad-group level where assets are less visible in the default interface view. A weekly policy-status check via the Associations view or API export catches disapprovals before they compound into sustained performance loss.[1]
Mistake 5: Leaving Expired Promotion Assets Active
Promotion assets without hard end dates continue serving after the offer expires, delivering an ad-to-page mismatch that harms conversion rate and user trust. Always set a platform-enforced end date on every promotion asset, and audit active promotion assets weekly.[1]
Mistake 6: Creating New Call-Only Ads Instead of RSAs with Call Assets
Google no longer supports the creation of new call-only ads, and existing call-only ads are scheduled to stop serving in February 2027.[13] Any account still running call-only ads should migrate to Responsive Search Ads with call assets immediately. Delaying this migration risks abrupt performance disruption when the deprecation deadline passes.
Mistake 7: Using Generic Stock Images as Image Assets
Image assets featuring generic stock photography — particularly images of diverse smiling groups in office settings — have been associated with lower performance relative to authentic product, facility, staff, or outcome imagery. One published comparison found that real-staff and local-landmark images produced a 20% higher CTR and a 15% higher conversion rate than generic stock images for the same advertiser.[8] Use authentic, specific visuals that match the ad copy and landing page.
Mistake 8: Applying Account-Level Assets That Are Only Relevant to a Subset of Campaigns
Account-level assets inherit downward to all campaigns and ad groups where they are eligible. A location-specific sitelink, a seasonal promotion, or a product-specific structured snippet applied at account level will serve in campaigns where it is irrelevant, reducing expected asset impact within Ad Rank and potentially misleading users.[9] Reserve account-level assets for universally accurate claims only.
Mistake 9: Replacing Assets Before They Have Sufficient Data
Google recommends allowing approximately 2–3 weeks before replacing low-performing assets in Performance Max, and the same principle applies to Search.[10] Replacing an asset after 50 impressions produces no meaningful signal; the label “Learning” or “Low” at low volume reflects data scarcity, not creative failure. Set a minimum impression threshold — typically 300–500 impressions — before treating a performance label as actionable.
Mistake 10: Failing to Opt Out of Landing-Page Image Extraction When It Is Brand-Inappropriate
From December 4, 2026, newly created Performance Max campaigns will default to GENERATE_IMAGE_EXTRACTION = OPTED_IN, meaning Google AI will pull images directly from landing pages to use as creative assets.[3] For brands with strict visual identity guidelines, regulated imagery, or landing pages containing images that are contextually inappropriate for ad creative (such as medical procedure photography or legal document screenshots), this default must be explicitly overridden. Failing to set GENERATE_IMAGE_EXTRACTION = OPTED_OUT at campaign creation — not just at the account level — will allow extraction to proceed even when the account-level dynamic image setting is disabled.[3]
Worked example (illustrative)
Correcting a Callout Duplication Problem That Wastes the 4-Asset Display Limit
- Setup: A Seattle-based moving company spending $9,500 per month on Search. The account has 9 active callouts at the campaign level: “Fast Moving,” “Quick Movers,” “We Move Fast,” “Licensed & Insured,” “Free Quotes,” “Local Seattle Movers,” “No Hidden Fees,” “Family Owned,” and “Affordable Rates.” Up to 4 callouts can display per auction.[2]
- Numbers: Of the 9 callouts, 3 (“Fast Moving,” “Quick Movers,” “We Move Fast”) communicate the same speed claim using synonymous phrasing. In any single auction where 2 of these 3 are selected, 50% of the available callout display slots (2 of 4) are consumed by the same message repeated twice. The account is paying no extra cost for callouts themselves (assets are free to add[1]), but the display opportunity cost means users see speed mentioned twice and miss “No Hidden Fees” or “Licensed & Insured” — both of which address higher-consideration objections for a $1,200–$2,400 average move value.
- Decision: Consolidate “Fast Moving,” “Quick Movers,” and “We Move Fast” into a single callout: “Same-Day Moves Available.” Add two replacement callouts communicating distinct benefits: “Flat-Rate Pricing” and “AMSA-Certified Crew.” Final callout count: 9, with each of the 9 communicating a unique benefit point, giving the system 126 possible 4-callout combinations (9 choose 4) versus the previous set’s heavily redundant combinations.
- Why: Each callout must add incremental information; near-duplicates consume limited display slots without adding persuasive variety, reducing the expected information value of the asset combination.[2]
Worked example (illustrative)
Avoiding the December 4, 2026 Image-Extraction Default in a Healthcare Performance Max Campaign
- Setup: A multi-location dermatology group in Phoenix, AZ, launching a new Performance Max campaign on December 10, 2026, targeting cosmetic procedure inquiries. Monthly budget: $14,000. Landing pages include before-and-after procedure photography (close-up skin conditions and post-procedure results) that is medically accurate but visually inappropriate for automated ad creative serving across Display, YouTube, and Gmail placements.
- Numbers: The December 4, 2026 default sets
GENERATE_IMAGE_EXTRACTION = OPTED_INfor all new Performance Max campaigns, regardless of whether the account-level dynamic image setting is disabled.[3] If the campaign is created without explicitly settingGENERATE_IMAGE_EXTRACTION = OPTED_OUT, Google AI may extract procedure photography from landing pages and use it as image assets across all Performance Max placements — an estimated 6 placement networks (Search, YouTube, Display, Discover, Gmail, Maps) — with no pre-approval step for the extracted imagery. - Decision: When creating the campaign via the Google Ads API on December 10, 2026, explicitly set
GENERATE_IMAGE_EXTRACTION = OPTED_OUTat the campaign level. Upload 7 manually approved image assets (3 square 1200×1200, 4 landscape 1200×628) featuring treatment environment and staff photography that has passed internal brand review. Document the opt-out decision in the account change log with a timestamp. - Why: The new default overrides the account-level dynamic image setting, so campaign-level opt-out is the only reliable mechanism to prevent automated extraction of clinically sensitive imagery from landing pages.[3]
13. What Changed Recently (Last 30 Days)
The following changes were announced or became active in October 2026. Each has a direct operational implication for US advertisers managing Search, Performance Max, or Demand Gen campaigns.
1. Performance Max Image Extraction Will Default to Opted-In on December 4, 2026
This is the most consequential near-term change for advertisers running or planning Performance Max campaigns. Starting December 4, 2026, Google will set GENERATE_IMAGE_EXTRACTION = OPTED_IN as the default for all newly created Performance Max campaigns, across the Google Ads interface, the API, and Google Ads Editor.[3]
When enabled, Google AI will extract high-resolution images from campaign landing pages and use them as creative assets across Search, YouTube, Display, Discover, Gmail, and Maps placements.[3] The critical operational detail is that this new campaign-level default overrides the account-level dynamic image extensions setting. Advertisers who have disabled dynamic image extensions at the account level cannot rely on that setting to block extraction in newly created Performance Max campaigns after December 4.[3]
Accounts that should act before December 4, 2026:
- Healthcare, legal, financial services, and other regulated verticals where landing-page imagery may be clinically or legally sensitive
- Brands with strict visual identity guidelines that require pre-approval of all ad creative
- Advertisers whose landing pages include imagery that is accurate in context but unsuitable for cross-network automated ad serving
- Developers and agencies building campaign creation workflows via the Google Ads API, who must update creation scripts to include an explicit opt-out field
Existing Performance Max campaigns are not affected — they retain their current configuration.[3] Only campaigns created on or after December 4, 2026 are subject to the new default.
2. Text Disclaimer Assets Added to the Google Ads API (v22)
Google Ads API v22 introduces TEXT_DISCLAIMER as a new asset type, along with the corresponding ServedAssetFieldType.TEXT_DISCLAIMER enum.[18] This indicates infrastructure support for disclaimer text as a distinct, reportable asset type separate from ad copy. As of October 2026, Google’s public help center documentation does not confirm broad interface availability or detail serving rules for this asset type. Treat TEXT_DISCLAIMER as an API-level capability and verify account eligibility and interface support before deploying in production campaigns.[18]
This asset type is likely to be relevant for financial services, healthcare, legal, and other regulated categories where disclaimer language must appear in the ad unit. Monitor Google’s advertiser-facing documentation for serving guidance as it becomes available.
3. LINE and SMS Business Message Providers Added
BusinessMessageAsset now supports LINE and SMS through new provider values and fields in the API.[18] This expands message-initiated engagement beyond previously supported providers. For US-focused advertisers, SMS-initiated business messaging may become a viable asset type for industries where text-based lead capture is preferred, such as real estate, automotive, and home services. Interface availability and US eligibility should be confirmed before testing in active campaigns.[18]
4. Generative Asset Creation via AssetGenerationService (API v22)
API v22 introduces AssetGenerationService, which can generate text and image assets using generative AI.[18] This is an API-level service, not a universally available front-end feature. Do not treat this as confirmation that all advertiser accounts have immediate access to a new creative generation workflow. Access requirements, policy constraints, and output review obligations should be verified with Google before using generative asset creation in production.[18]
From a governance standpoint, any assets generated via AssetGenerationService should be reviewed for brand accuracy, factual correctness, policy compliance, and landing-page alignment before being set to eligible. Automated generation does not eliminate the advertiser’s responsibility for asset content.[1]
5. Asset-Group URL Controls for Performance Max
Tracking templates, custom parameters, and final URL suffixes can now be configured directly on AssetGroup for Performance Max campaigns.[21] Previously, these tracking settings were managed at the campaign or account level, making asset-group-specific attribution more difficult for advertisers running multiple asset groups within a single Performance Max campaign. This change allows more granular tracking without requiring separate campaigns per asset group.
Before implementing asset-group-level URL controls, audit existing campaign-level and account-level tracking templates to identify conflicts. A tracking parameter set at all three levels simultaneously (account, campaign, and asset group) will follow Google’s hierarchy, but inconsistencies can produce broken tracking or duplicate parameter values.[21]
6. Asset Experiments Coming to Performance Max
Google has announced that asset experiments are being introduced to Performance Max, enabling controlled testing of changes to headlines, descriptions, images, and videos within asset groups.[22] As of October 2026, this feature appears to be rolling out rather than universally available. Use the feature only where it is visible in the account interface, and apply standard experiment discipline: change one asset variable at a time, preserve a control asset group, avoid editing the tested asset group during the experiment window, and allow sufficient time (minimum 2–3 weeks) for meaningful signal accumulation.[10]
7. Call-Only Ad Deprecation Timeline Remains Active
This is not a new October 2026 announcement, but it is the most urgent migration deadline for advertisers still running call-only ads. Google has confirmed that existing call-only ads will stop serving in February 2027.[13] New call-only ad creation is already blocked. Any account still running call-only ads must complete migration to Responsive Search Ads with call assets before February 2027. This includes verifying that call conversion actions, call reporting, and minimum-call-duration settings are correctly configured on the replacement campaigns.[13]
Worked example (illustrative)
Auditing and Migrating Call-Only Ads Before the February 2027 Deprecation Deadline
- Setup: A Houston plumbing company spending $8,200 per month across 3 Search campaigns. Two campaigns use Responsive Search Ads with call assets. One campaign — “Emergency Plumbing” — still runs 4 active call-only ads that collectively generated 310 calls in September 2026 at an average cost per call of $18.40, totaling $5,704 in call-driven spend. The February 2027 stop-serving deadline gives the account approximately 4 months to migrate.
- Numbers: The “Emergency Plumbing” campaign has a $5,800 monthly budget. Call-only ads account for 4 of the 4 ads in this campaign (100% share). If call-only ads stop serving on February 1, 2027 and no replacement is in place, the campaign loses 100% of its ad delivery — an estimated $5,800 in monthly spend with zero impressions. Migration to RSAs with call assets requires: 3 RSAs with 15 headlines and 4 descriptions each, 1 call asset with the existing tracking number, 1 call conversion action set at 60-second minimum duration (matching current quality threshold), and dayparting set to Mon–Sun 6:00 AM–10:00 PM CT (hours the call center is staffed).[13]
- Decision: Build and launch 3 RSAs with call assets in the “Emergency Plumbing” campaign by November 1, 2026, run them alongside the call-only ads through January 15, 2027 to validate call volume parity, then pause the call-only ads on January 16, 2027 — 15 days before the February 2027 stop-serving deadline.
- Why: Call-only ads stop serving in February 2027; completing migration by January 16, 2027 provides a 15-day buffer to identify and correct any call conversion tracking gaps before the forced deprecation removes the option to compare performance against the legacy format.[13]
Worked example (illustrative)
Building a Pre-December 4 Audit Checklist for Performance Max Image Extraction
- Setup: A digital marketing manager at a US insurance brokerage oversees 6 active Performance Max campaigns spending a combined $47,000 per month. The brokerage’s landing pages include customer testimonial photos, auto-accident scene imagery used for claims context, and attorney headshots (for a legal referral product). The account-level dynamic image extensions setting is currently disabled.
- Numbers: Of the 6 campaigns, all 6 are Performance Max. Under the December 4, 2026 default, any new Performance Max campaign created on or after that date will default to
GENERATE_IMAGE_EXTRACTION = OPTED_IN, regardless of the account-level dynamic image setting.[3] The 6 existing campaigns are not affected. However, the team plans to create 2 new Performance Max campaigns in Q1 2027 — a January 15, 2027 “Auto Insurance” campaign ($12,000/month budget) and a February 3, 2027 “Home Insurance” campaign ($9,000/month budget). Both will be created after the December 4 default change. - Decision: By November 21, 2026 (2 weeks before the December 4 deadline): (1) Document which landing page images are brand-approved for ad creative on a per-campaign basis. (2) Update the API campaign-creation script to include
GENERATE_IMAGE_EXTRACTION = OPTED_OUTfor all insurance campaigns. (3) Upload 7 manually approved images per planned campaign (3 square 1200×1200, 4 landscape 1200×628) so approved creative is ready at launch. (4) Add a creation-checklist item requiring explicit opt-out confirmation for every new Performance Max campaign created after December 4, 2026.[3] - Why: The new default overrides the account-level setting, so campaign-level opt-out at creation time is the only reliable control; without it, auto-accident and claims imagery on landing pages would be eligible for extraction and automated serving across Display, YouTube, Gmail, and Discover placements.[3]
Related reading
- Building a Google Ads Strategy
- How to Improve your Google Ads Campaigns
- Google Ads Looker Studio Reporting Best Practices
- Maximise Conversions, Maximise Conversion Value Best Practice.
Or skip the work and get a Search campaign set up by a Google Ads specialist.
References
- [1] https://support.google.com/google-ads/answer/7331111?hl=en support.google.com
- [2] https://support.google.com/google-ads/answer/12073962?hl=en support.google.com
- [3] http://ads-developers.googleblog.com/2026/10/upcoming-changes-to-image-extraction-in.html ads-developers.googleblog.com
- [4] https://adbrains.ai/resources/google-ads-assets-extensions-which-to-use-and-why adbrains.ai
- [5] https://www.mbadv.agency/google-ads/ad-assets-and-formats www.mbadv.agency
- [6] https://support.google.com/google-ads/answer/6167122?sjid=17379746141931119139-EU support.google.com
- [7] https://www.prontomarketing.com/case-studies/iconic-roofing-google-ads/ www.prontomarketing.com
- [8] https://creativeadslab.com/google-ads-asset-library-boost-ctrs-in-2026/ creativeadslab.com
- [9] https://support.google.com/google-ads/answer/6280012?hl=en support.google.com
- [10] https://support.google.com/google-ads/answer/14528220?hl=en support.google.com
- [11] https://www.relevantaudience.com/google-ads-en/google-ads-sitelink-assets/ www.relevantaudience.com
- [12] https://support.google.com/google-ads/answer/2375416?hl=en support.google.com
- [13] https://developers.google.com/google-ads/api/docs/deprecations developers.google.com
- [14] https://support.google.com/google-ads/answer/17092074?hl=en support.google.com
- [15] https://support.google.com/google-ads/answer/6238826?hl=en support.google.com
- [16] https://support.google.com/google-ads/answer/13676244?hl=en support.google.com
- [17] https://developers.google.com/google-ads/api/docs/assets/automated-assets developers.google.com
- [18] https://developers.google.com/google-ads/api/docs/release-notes developers.google.com
- [19] https://developers.google.com/google-ads/api/performance-max/asset-reporting developers.google.com
- [20] https://support.google.com/google-ads/answer/10724748?hl=en support.google.com
- [21] https://ads-developers.googleblog.com/2026/ ads-developers.googleblog.com
- [22] https://business.google.com/us/accelerate/announcements/ business.google.com

