Google Search Ads: How Search Campaigns Work

  • Google Ads

Google Ads Starter FREE Course · Day 5 of 7 · Article 1 of 3

Structure, budgets and bidding at launch

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:

  1. Google Search Ads: How Search Campaigns Work (you are here)
  2. Budget Management
  3. Bidding Strategies
← Day 4: Keywords and intent Day 6: Ads and landing pages →
All seven days
  1. Is Google Ads worth it for you? The maths
  2. How the auction works, and setting up without the traps 2 articles
  3. Measurement before money 3 articles
  4. Keywords and intent 3 articles
  5. Structure, budgets and bidding at launch (you are here) 3 articles
  6. Ads and landing pages 3 articles
  7. 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.

Search campaigns are where high-intent demand is captured. How to structure them now that broad match and Smart Bidding do the matching, what to write into responsive search ads, and where negatives and budgets still need a human.

Last updated: November 2, 2026

In This Guide

  1. Executive Summary
  2. Benchmarks & Numbers at a Glance
  3. Campaign Structure & Account Architecture
  4. Keyword Strategy & Match Types
  5. Negative Keywords & Search Terms
  6. Responsive Search Ads & Ad Strength
  7. Assets & Extensions
  8. Smart Bidding (tCPA / tROAS)
  9. Budgets & Bid Management
  10. Audience Signals & Targeting
  11. Common Mistakes to Avoid
  12. What Changed Recently
  13. References

Google Ads Search Campaigns: Comprehensive Reference Guide — November 2026

1. Executive Summary

Google Ads Search in November 2026 is an AI-first platform. The five principles below reflect current Google guidance and the direction of every major product change in the past 12 months.

  • Principle 1 — Consolidate for signal density. Smart Bidding optimizes across auctions using conversion data. Fragmented campaigns starve the algorithm. Separate campaigns only when budget, geography, language, product economics, or business objectives genuinely require independent control.[1]
  • Principle 2 — Broad match is the default growth mechanism when measurement is reliable. Google’s 2026 guidance positions broad match as the primary discovery layer when conversion-based Smart Bidding and accurate conversion tracking are in place. Exact and phrase match serve as a control layer, not a complete strategy.[2][3]
  • Principle 3 — Conversion quality determines everything downstream. Smart Bidding, Target CPA, and Target ROAS all optimize toward whatever signal you label as a Primary conversion. If that signal is a low-value micro-conversion or a duplicate event, the algorithm optimizes for the wrong outcome. Fix measurement before changing bid strategy.[1][4]
  • Principle 4 — AI Max is the near-term structural transition. Dynamic Search Ads migrate automatically to AI Max in February 2027. Legacy campaign-level broad match and standalone Automatically Created Assets have been closed to new creation since August 3, 2026. Test AI Max in controlled experiments now rather than accepting forced migration without guardrails.[3]
  • Principle 5 — Assets are a competitive lever, not an afterthought. Responsive Search Ads with “Good” or “Excellent” Ad Strength, a full asset suite (sitelinks, callouts, structured snippets, images, call, location), and AI-assisted headline variety materially affect auction eligibility and Quality Score. Every ad group without at least two RSAs and four asset types is leaving performance on the table.[1][5]

2. Benchmarks and Numbers at a Glance

All figures below are vendor-reported benchmarks, not independently audited data. Use them as orientation points, not performance guarantees. Where two sources disagree, the more conservative figure is noted. US market figures throughout.

Metric Typical range or threshold Applies when Source
Search CTR, all industries 6.64%–6.79% Vendor claim; cross-industry average; use as a floor, not a target [6][7]
Search CTR range by industry 5.56% (lowest) to 12.80% (highest, Arts & Entertainment) Vendor claim; 22 industry categories; Arts & Entertainment highest, lowest category not named [7]
Search CPC, all industries $4.53–$5.42 per click Vendor claim; cross-industry average; Benchmarketing median $4.53, LocaliQ/WordStream $5.42 — use $4.53 as more conservative [6][7]
Search CPC range by industry $1.63 (Arts & Entertainment) to $9.87 (Attorneys & Legal Services) Vendor claim; 22 industry categories [7]
Search conversion rate, all industries 7.32%–8.18% Vendor claim; Benchmarketing median 7.32%, LocaliQ/WordStream 8.18%; full range 2.64%–16.22% across industries [6][7]
Search cost per lead, cross-industry median $60.98 Vendor claim; Benchmarketing 22-category median; Attorneys & Legal Services highest at $131.63, Arts & Entertainment lowest at $26.84 [7]
Search impression share, cross-industry average 33.9%–37% Vendor claim; Conner Crowe 31-account sample 33.9%; Benchmarketing cross-industry average 37% [8][9]
Search impression share, top performers 65% or higher Vendor claim; Benchmarketing threshold for top-performing accounts [9]
Search lost IS to budget (non-brand campaigns) 32% Vendor claim; Benchmarketing non-brand benchmark; brand campaigns average 15% lost to budget [9]
Search lost IS to rank (non-brand campaigns) 38% Vendor claim; Benchmarketing non-brand benchmark; brand campaigns average 20% lost to rank [9]
Accounts where budget loss exceeded rank loss 74.2% of sampled accounts (23 of 31) Vendor claim; Conner Crowe 31-account sample; implies budget constraint is more common than Quality Score or bid problems [8]
Search CPC, 33-account benchmark $8.53 per click Vendor claim; Conner Crowe 33-account sample; higher than cross-industry averages, likely reflecting industry mix in sample [8]
AI Max lift (non-retail) +14% more conversions or conversion value at similar CPA/ROAS Google internal 2025 data — vendor claim, not an independent guarantee; applies to non-retail advertisers activating AI Max [3]
Daily budget overspend allowance (tCPA campaigns) Up to 2× average daily budget on a single day Google policy; remains within monthly charging limit; applies to all Smart Bidding strategies [10]
Minimum tROAS historical data window 3 weeks or 1–2 conversion cycles, whichever is longer Google recommendation before activating value-based bidding; applies when switching from Maximize conversion value [4]

3. Campaign Structure and Account Architecture

Google’s 2026 account-structure guidance centers on one principle: give Smart Bidding enough conversion data to optimize effectively by eliminating structural fragmentation that serves legacy management habits rather than actual business control needs.[1]

When to Separate Campaigns

Separate campaigns only when a genuine operational difference exists. The following table maps the decision:

Keep separate when… Consolidate when…
Budgets must be independently capped (e.g., brand vs. non-brand investment rules) Campaigns share the same daily or monthly spend cap
Target CPA or ROAS differs materially between product lines The desired efficiency target is the same across products
Conversion values or margins differ significantly Products have comparable unit economics or reliable value tracking is in place
Different countries, currencies, languages, or legal requirements Targeting differences do not require separate operational control
Different primary conversion actions (leads vs. purchases) The same primary conversion goal applies throughout
Brand vs. non-brand needs distinct reporting or investment governance Brand/non-brand distinction does not affect budget or bidding decisions

Do not create separate campaigns to isolate exact, phrase, and broad match versions of the same theme. Google’s current guidance explicitly recommends against this.[1][11] Do not segment campaigns by device; Smart Bidding already evaluates device signals at auction time.[1]

Ad Group Architecture

Build themed ad groups organized around coherent user intent, not around match-type variations or minor keyword permutations. Google recommends moving away from single-keyword ad groups (SKAGs) toward intent-based themes.[11] Each ad group should have:

  • A tightly related set of keywords reflecting one intent theme
  • At least 2 RSAs with “Good” or “Excellent” Ad Strength
  • A closely aligned landing page with a clear conversion path
  • Relevant negative keywords to block clearly off-topic queries
  • Active search-term monitoring to catch irrelevant demand early

Shared Budgets and Portfolio Strategies

Use shared budgets among campaigns with compatible objectives and targets only. Do not pool brand and non-brand campaigns if brand would absorb spend needed for incremental acquisition.[1] Portfolio Target CPA is appropriate when multiple campaigns share the same objective and target, allowing the system to optimize collectively across the group.[10]

Worked example (illustrative)

Consolidating Over-Segmented HVAC Account

  • Setup: A Denver HVAC account spending $18,000/month across 9 campaigns — 3 campaigns per service line (AC repair, furnace repair, new installs), each segmented into exact, phrase, and broad match.
  • Numbers: Each of the 9 campaigns averages $2,000/month. At a cross-industry median conversion rate of 7.32%[7] and $5.42 CPC[6], each campaign receives approximately 369 clicks and 27 conversions per month. Google’s Smart Bidding minimum for stable optimization is generally cited as 30–50 conversions per 30-day window; 7 of 9 campaigns fall below 30 conversions. Consolidating to 3 theme-based campaigns ($6,000 each) yields approximately 1,107 clicks and 81 conversions per campaign per month, clearing the threshold by 62%.
  • Decision: Consolidate from 9 campaigns to 3 campaigns (AC Repair, Furnace Repair, New Installs), each using broad match with Target CPA set at the historical blended CPA of $74, and remove all match-type-segmented campaigns.
  • Why: Google’s guidance states that campaigns with insufficient conversion volume cannot optimize effectively, and segmenting by match type rather than business objective is explicitly discouraged.[1][11]

Done for you

Everything on this page, done for you before you launch

We build the campaign properly the first time — so you are not unpicking it three months from now.

✓  Keywords validated against real search volume, grouped by intent

✓  Responsive search ads written to the asset counts that actually perform

✓  A negative keyword list built before you spend, not after

✓  Delivered in 24 hours as a spreadsheet you keep

See what you get — $129

Human-reviewed · No subscription · We never touch your Google Ads account

4. Keyword Strategy and Match Types

Google’s 2026 match-type guidance is the most significant structural shift from prior-year best practices. Broad match is the recommended default growth mechanism when conversion-based Smart Bidding and reliable conversion measurement are confirmed.[2][3]

Match Type Roles in 2026

Match type Primary role Use when
Broad match Discovery and scale Conversion-based Smart Bidding is active, conversion tracking is reliable, and the goal is incremental qualified demand[2][3]
Phrase match Intermediate control A specific word order or phrase is required in the query while allowing some variation; less emphasis in 2026 than prior years[11]
Exact match Control layer and brand protection Protecting branded terms, managing high-value specific queries, or where impression share control is required[4]

Google explicitly states that broad match does not require duplicate versions of the same keyword in exact, phrase, and broad when Smart Bidding is active; broad match can reach searches eligible for narrower match types and additional relevant searches.[1][3] Do not create numerous close variants such as singular/plural, reordered words, or minor spelling variations — Google states these may match the same searches and add no performance value.[11]

Keyword Themes and Intent Organization

Organize ad groups around intent themes, not keyword permutations. Useful theme dimensions include product or service category, specific product, user problem or need, use case, commercial intent stage (pricing, quote, booking, demo, purchase), brand versus non-brand intent, and geography where the location changes the offer or landing page.[1]

Broad Match Prerequisites

Before activating broad match as the primary match type, confirm all of the following:[1][2]

  • At least one conversion-based Smart Bidding strategy is active (Maximize Conversions, Target CPA, Maximize Conversion Value, or Target ROAS)
  • Conversion tracking fires on qualified business outcomes, not page views or low-value micro-conversions
  • RSAs provide sufficient asset variety to address multiple query intents within the theme
  • The search terms report is reviewed at least weekly
  • Negative keywords are current and cover known irrelevant intent

Worked example (illustrative)

Transitioning a Legal Services Account to Broad Match

  • Setup: A Chicago personal injury law firm spending $12,000/month on Search, currently using only exact and phrase match, generating 91 conversions (form submissions) per month at an average CPA of $131.63[7] (consistent with the Attorneys & Legal Services cross-industry benchmark).
  • Numbers: Current state: 91 conversions × $131.63 = approximately $11,978/month spend. CPC at the legal services benchmark of $9.87[7] implies 1,214 clicks/month, with a conversion rate of 91 ÷ 1,214 = 7.49% (slightly above the 5.55% legal services benchmark, reflecting a well-managed account). A 14% incremental lift from broader matching (consistent with Google’s AI Max internal benchmark)[3] would add approximately 13 conversions (91 × 0.14 = 12.7), reaching 104 conversions at the same $12,000 budget, lowering CPA to $115.38.
  • Decision: Run a 6-week Google Ads experiment — 50% traffic to existing exact/phrase structure, 50% to broad match + Target CPA at $131 — with qualified consultation requests (not raw form fills) as the Primary conversion action. Do not apply account-wide until the experiment reaches statistical significance.
  • Why: Google recommends testing broad match changes separately so performance can be compared against the prior structure, and the legal vertical’s $131.63 CPL benchmark[7] makes uncontrolled testing financially material at this spend level.[1]

5. Negative Keywords and Search Terms Management

Negative keywords are a traffic-quality and intent-control system, not a substitute for Smart Bidding.[2] Their role becomes more important, not less, as broad match and AI Max expand the range of queries that can trigger ads.

Negative Match Type Selection

Negative match type Use when Caution
Negative exact Excluding one specific query precisely Narrowest exclusion; may not block variations of the same intent
Negative phrase Excluding a recurring phrase while preserving potentially relevant variations Review to ensure it does not suppress qualified searches containing that phrase
Negative broad Blocking a thematic topic across all query forms Can suppress a wider set than intended; audit after adding

Priority Negative Categories

Use the Search terms report to identify and exclude:[2][4]

  • Irrelevant topics unrelated to the offer
  • Jobs, careers, training, or salary searches (e.g., “how to become a…”, “salary for…”)
  • Free, used, DIY, or low-commercial-intent queries where the business does not serve that need
  • Products or services not offered
  • Locations outside the service area
  • Competitor brand terms requiring separate treatment or budget allocation
  • Informational research queries producing no conversion history

List Architecture

Maintain at least three distinct negative keyword lists:[2]

  • Account-wide shared list: Universal exclusions (jobs, careers, free, DIY, unserved locations) applied to every campaign
  • Brand exclusion list: Competitor names and terms that require separate budget treatment
  • Campaign-specific lists: Exceptions, local differences, and product-specific exclusions

Maintain a change log for major additions and removals. After adding a new negative, verify in the Search terms report within 14 days that qualified queries have not been suppressed.

Worked example (illustrative)

Building a Negative Keyword List for a Plumbing Account After Enabling Broad Match

  • Setup: A Phoenix plumbing account spending $5,000/month activates broad match + Maximize Conversions in January 2027. After 14 days, the search terms report shows 1,840 unique queries. Of those, 210 contain the word “job,” “career,” “apprentice,” or “training”; 95 contain “DIY” or “how to”; and 47 contain a city not served (Flagstaff, Prescott).
  • Numbers: At $4.53 CPC[7], 352 irrelevant clicks (210 + 95 + 47) = $1,594.56 in wasted spend over 14 days, or $3,189/month — 63.8% of the $5,000 monthly budget. The account’s industry CPL benchmarks are not available, but at the cross-industry median of $60.98[7], $3,189 of waste represents approximately 52 leads that could have been generated instead.
  • Decision: Add negative phrase “jobs,” negative phrase “careers,” negative phrase “apprentice,” negative phrase “training,” negative broad “DIY,” negative phrase “how to,” and negative exact [plumber Flagstaff] and [plumber Prescott] to the campaign-specific list immediately. Set a calendar reminder to re-review the search terms report on February 1, 2027.
  • Why: Google recommends using the Search terms report to add negatives for searches that should not generate ads, and negative phrase is the preferred match type for recurring unwanted themes.[2]

6. Responsive Search Ads and Ad Strength

Responsive Search Ads (RSAs) are the only standard text ad format available for Search campaigns in 2026. Every ad group should contain at least 2 RSAs, each with a unique final URL and “Good” or “Excellent” Ad Strength.[1] Google may also feature RSA headlines in sitelink positions, so every headline must be meaningful as a standalone navigation element.[3]

Asset Coverage Requirements

  • Supply up to 15 headlines and 4 descriptions; provide as many useful, unique options as possible[12]
  • Practical minimum: 8–10 distinct headlines, including at least 2 containing relevant keywords and 3 that communicate benefits, differentiators, or calls to action beyond the keyword itself[4]
  • Use all 4 description slots; ensure each is meaningfully different — not four minor variations of the same sentence
  • Make every asset independently understandable; Google may combine headlines and descriptions in any order[12]

Headline Framework

A balanced 15-headline set should include:[4][12]

  • 2–3 keyword or category headlines
  • 2–3 core-benefit headlines
  • 2–3 differentiators or proof points (e.g., licensed, insured, years in business, certifications)
  • 2–3 offer, price, availability, or guarantee headlines
  • 1–2 audience or use-case headlines
  • 1–2 calls to action

Pinning: When to Use It and When Not To

Pinning restricts the combinations Google can test and typically lowers Ad Strength.[1][12] Pin only when a message must appear in a specific position for legal, regulatory, compliance, or brand-identity reasons. If pinning is necessary, pin multiple distinct assets to the same position — not a single asset — to preserve variation.[12] Never pin every headline into fixed positions merely to reproduce one traditional ad.

Ad Strength as a Diagnostic

Treat Ad Strength as a diagnostic signal, not a performance guarantee. A “Good” or “Excellent” score indicates asset diversity and coverage; it does not guarantee conversion results. Evaluate RSAs by conversion rate, CPA, ROAS, and lead quality — not Ad Strength alone.[1][13] Do not add low-quality filler headlines just to move from “Poor” to “Good.”

AI-Generated Assets

Use AI-generated RSA headlines and descriptions as a drafting aid, not as unsupervised final copy. Before publishing any AI-generated asset:[12]

  • Verify every factual claim, price, guarantee, and availability statement
  • Remove generic, repetitive, exaggerated, or off-brand language
  • Confirm copy matches the exact landing page and keyword intent
  • Check that no testimonials, certifications, or performance statistics have been invented
  • Preserve required legal or regulatory disclosures

Worked example (illustrative)

Rebuilding an RSA from “Poor” to “Excellent” Ad Strength

  • Setup: A national SaaS accounting software account has one RSA per ad group with 5 headlines, 2 descriptions, and Ad Strength rated “Poor.” All 5 headlines contain the keyword “accounting software” with minor variations. Both descriptions are nearly identical.
  • Numbers: The ad group receives 620 impressions/month, 42 clicks (6.77% CTR, near the 6.79% cross-industry median[7]), and 3 conversions at a $5.42 CPC[6] = $226.44 in spend. Adding 10 additional distinct headlines (benefits: “Cut Close Time by 3 Days,” “Bank-Level Encryption,” “Integrates With QuickBooks”; differentiators: “SOC 2 Type II Certified,” “Dedicated Onboarding Team”; CTAs: “Start Your 30-Day Free Trial,” “Book a Live Demo Today”; offer: “No Setup Fees”; use-case: “Built for Multi-Entity Teams”; availability: “Support Available 7 Days a Week”) and 2 distinct descriptions brings total assets to 15 headlines and 4 descriptions. Ad Strength moves from “Poor” to “Excellent.”
  • Decision: Publish the expanded RSA alongside the existing one as a second RSA in the ad group. Set a 30-day review date (December 15, 2026) to compare conversion rate and CPA between the two RSAs using the Ad Variation report.
  • Why: Google requires at least 2 RSAs per ad group with “Good” or “Excellent” Ad Strength, and headline diversity is the primary lever for moving from “Poor” Ad Strength.[1][4]

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 Instagram

7. Assets and Extensions

Assets (formerly extensions) are free to add and serve when Google predicts they will improve performance.[3] They affect Quality Score‘s expected click-through rate component, ad eligibility, and the visual footprint of the ad on the SERP. Every campaign should have at least 4 asset types active.[3]

Asset Type Priority and Requirements

Asset type Minimum recommended Key requirement
Sitelinks 6–8 per campaign; distinct destinations Each sitelink must link to a unique, relevant destination; not near-duplicates of each other[3]
Callouts At least 4 (minimum 2 required to serve; up to 4 may show at once) Short, factual, specific differentiators — not vague claims; maximum impact in approximately 12–25 characters[14]
Structured snippets At least 1 relevant header with 4+ values Choose header that accurately describes values; keep values under 12 characters for mobile[14]
Images At least 4 unique images (square + landscape formats) Center important content to survive cropping; directly relevant to keyword and landing page[14]
Call assets 1 verified number; schedule availability accurately Use only a staffed number; set call scheduling to match actual office hours
Location assets Link to appropriate Business Profile locations Keep addresses and hours current; required for location extensions to serve
Promotion assets As relevant; include dates and eligibility Promotion start/end dates, discount amount, and redemption conditions must be accurate and current
Price assets As relevant; minimum 3 price items Keep prices, units, and destination pages current; stale prices create compliance risk

Sitelink Best Practices

Use sitelinks to send users directly to important sections: product categories, pricing, contact or booking pages, store locations and hours, reviews, promotions, shipping and returns FAQs.[3][5] Write concise, action-oriented link text. Make the linked page match the sitelink promise. Add sitelinks at the account, campaign, and ad group levels strategically so the most relevant links serve with each theme.

Dynamic Image Assets

Google supports dynamic image assets, which use AI to select relevant images from the landing page.[14] Review automatically selected images carefully — particularly when the landing page contains user-generated content, outdated photography, or imagery from unrelated page sections. Dynamic selection is not a substitute for manually uploaded, purpose-selected images.

Worked example (illustrative)

Adding Structured Snippets and Callouts to a Home Services Account

  • Setup: A Seattle roofing company spending $3,500/month has sitelinks and call assets active but no callout or structured snippet assets. The account’s impression share is 31% (below the 37% cross-industry average[9]), and lost IS to rank is 44% (above the 38% non-brand benchmark[9]), indicating a Quality Score or bid-relevance issue rather than a budget problem.
  • Numbers: At the cross-industry median CPL of $60.98[7] and a $3,500 monthly budget, the account should generate approximately 57 leads/month. If the current conversion rate is 7.32%[7] and CPC is $5.42[6], then: $3,500 ÷ $5.42 = 645 clicks × 7.32% = 47 leads — 10 fewer than the benchmark CPL implies, consistent with a rank-loss penalty reducing eligibility. Adding 6 callouts (“Licensed & Bonded,” “Same-Day Estimates,” “25-Year Warranty,” “No Hidden Fees,” “Storm Damage Experts,” “Financing Available”) and 1 structured snippet header “Services” with 5 values (“Roof Replacement,” “Storm Repair,” “Gutter Installation,” “Skylight Installation,” “Emergency Tarping”) improves expected CTR signals without a bid change.
  • Decision: Publish 6 callout assets and 1 structured snippet (Services header, 5 values under 12 characters each) at the campaign level by November 20, 2026. Monitor lost IS to rank weekly for 4 weeks; target a reduction from 44% toward the 38% non-brand benchmark.[9]
  • Why: Assets affect the expected click-through rate component of Quality Score, which directly drives lost impression share to rank; Google confirms assets serve when predicted to improve performance.[3]

8. Smart Bidding (tCPA and tROAS)

From June 2026, Google has relabeled “Maximize conversions with a Target CPA” as Target CPA and “Maximize conversion value with a Target ROAS” as Target ROAS. The underlying bidding behavior is unchanged.[1] The choice of bid strategy is the most consequential single setting in a Search campaign — it determines what the algorithm optimizes for at every auction.

Bid Strategy Selection

Business objective Recommended strategy Prerequisite
Maximum conversion volume within budget, no efficiency constraint Maximize Conversions Reliable conversion tracking; qualified Primary conversion action set[1]
Conversion volume at a specific average cost per conversion Target CPA Stable historical CPA; sufficient conversion volume; conversion lag accounted for[10]
Maximum conversion value within budget, no ROAS constraint Maximize Conversion Value Reliable conversion values representing actual business outcomes[4]
Conversion value at a specific return on ad spend Target ROAS Reliable values; minimum 3 weeks or 1–2 conversion cycles of value data; stable historical ROAS[4]
Click volume when conversion tracking is not yet usable Maximize Clicks Use only as a temporary measure; less aligned with lead or revenue growth[1]
Brand impression control Target Impression Share Paired with exact or phrase match for brand coverage[4]

Target CPA: Setting and Adjusting

Start the Target CPA near the recent, stable historical CPA rather than an aspirational figure. A target materially below achievable historical performance restricts delivery and reduces conversion volume. Google recommends accounting for conversion lag — avoid using periods where late conversions have not yet been recorded when establishing the baseline.[10] If conversion volume is low or tracking has recently changed, begin with Maximize Conversions (no target) and introduce Target CPA only after the system has accumulated usable data across several conversion cycles.[1]

Target ROAS: Setting and Adjusting

Use Target ROAS only when conversion values are reliable and conversions are sufficiently frequent. Start from recent, stable ROAS based on approximately 30 days of historical data.[4] Upload value data for at least 3 weeks or 1–2 conversion cycles before activating. Do not mix conversions with materially different value definitions unless values accurately represent their business contribution. For lead generation, import qualified-lead or closed-sale values when possible — optimizing only to form submissions can cause the system to favor volume over lead quality.[4]

Learning Period and Performance Evaluation

After any bid strategy change, allow sufficient time for the learning period to complete before evaluating results. Avoid making frequent structural changes that reset learning. Judge performance over complete conversion cycles, not day-to-day fluctuations.[1]

Worked example (illustrative)

Setting a tCPA Target for a New Lead Generation Campaign

  • Setup: A Miami insurance agency launching a new Search campaign for homeowners insurance quotes in November 2026. No prior campaign-level conversion history exists. The agency’s historical average from prior manual CPC campaigns shows a CPL of $88 over the past 90 days across 67 leads, but 15 of those leads came from a one-week promotional period with a 40%-lower CPL not representative of normal traffic.
  • Numbers: Remove the promotional period: 52 leads over approximately 77 days at a non-promotional CPL = (total spend minus promo spend) ÷ 52. If total spend was $5,896 and promo spend was $800, non-promo CPL = $5,096 ÷ 52 = $98.00. The cross-industry median CPL is $60.98[7] but legal/insurance verticals are closer to $131.63[7]; $98 is between those benchmarks and should be treated as a realistic baseline.
  • Decision: Launch on Maximize Conversions (no target) for the first 30 days, with a $4,000/month budget. On December 15, 2026, if the account has accumulated 30 or more conversions, switch to Target CPA set at $102 (a 4% increase above the $98 baseline to avoid over-constraining delivery in the early phase). Do not set the target at $60 or $70 based on the cross-industry median.
  • Why: Google recommends starting with Maximize Conversions when data is limited, then introducing Target CPA from credible historical performance — not an aspirational or industry-average figure — and accounting for conversion lag when measuring the baseline.[1][10]

Worked example (illustrative)

Diagnosing Whether Budget or Rank is Limiting a Campaign

  • Setup: A Boston e-commerce retailer (outdoor gear) spending $8,000/month on a non-brand Search campaign. Impression share is 29%, below the 37% cross-industry average.[9]
  • Numbers: Lost IS to budget = 41%; lost IS to rank = 22%. The Conner Crowe benchmark shows 74.2% of accounts have budget loss exceeding rank loss[8]; this account fits that pattern. At the non-brand benchmark, lost IS to budget should be approximately 32%[9] — this account’s 41% is 9 percentage points above benchmark, meaning budget is the binding constraint, not Quality Score. To close from 29% to 37% impression share on budget alone, the account would need to increase budget by approximately 41% × (37% ÷ 29% − 1) = approximately 28%, implying a budget increase of $8,000 × 0.28 = $2,240/month, bringing the total to $10,240/month.
  • Decision: Increase the campaign budget from $8,000/month to $10,200/month effective December 1, 2026, while holding Target ROAS at its current setting of 320%. Do not lower Target ROAS to chase impression share — the diagnosis is budget-limited, not rank-limited.
  • Why: When lost IS to budget exceeds lost IS to rank, the correct lever is budget, not bid or Quality Score intervention; changing the wrong lever wastes optimization time.[8][9]

“So what do I actually have to hand over?”

Nothing. No access to your Google Ads account, no login, no agency link request. You tell us the business and the offer; we send back a spreadsheet.

You import it yourself, or hand it to whoever runs the account. It is yours to keep and re-use — there is no ongoing fee and nothing to cancel.

How it works — $129

9. Budgets and Bid Management

Budget decisions and bid strategy decisions are distinct but interact directly. A campaign that is budget-limited cannot fully benefit from Target CPA or Target ROAS tightening; a campaign that is rank-limited cannot benefit from budget increases alone.[8][9]

Budget Setting Principles

  • Set budgets according to the business objective and expected demand, then allow Smart Bidding to allocate bids within those limits[1]
  • Google may spend up to 2× the average daily budget on a given day while remaining within the applicable monthly charging limit (daily budget × 30.4)[10]
  • Avoid unnecessary campaign fragmentation that leaves many campaigns “Limited by budget” — consolidation is the preferred fix[1]
  • When increasing budgets, make changes deliberately and allow performance to stabilize; large increases can alter auction participation and traffic mix
  • Use shared budgets only among campaigns with compatible objectives, targets, and economics[1]

Diagnosing Budget vs. Rank Loss

Before changing bids or budgets, determine which type of impression share loss is dominant using the Search Lost IS (budget) and Search Lost IS (rank) columns. The Conner Crowe benchmark found that 74.2% of sampled accounts (23 of 31) had budget loss exceeding rank loss.[8] This means the majority of accounts constrained by impression share are budget-limited, not Quality Score or bid-limited. Apply the correct remedy:

  • Budget-limited: Increase budget or consolidate campaigns; do not lower tCPA/tROAS targets
  • Rank-limited: Improve Quality Score components (ad relevance, expected CTR, landing page experience) or raise tCPA/tROAS targets to compete at higher bids

Smart Bidding Exploration

Google’s October 2026 announcements promote Smart Bidding Exploration, which allows more flexible ROAS targets to pursue additional conversion opportunities beyond existing demand.[3] Use this only when the current campaign is not budget-limited and there is a specific reason to believe untapped demand exists at a modestly lower ROAS. Set a ROAS floor and monitor marginal conversion quality closely.

Worked example (illustrative)

Setting a Realistic Daily Budget for a New tCPA Campaign

  • Setup: A Minneapolis accounting firm launching a Search campaign for tax preparation services in January 2027, targeting a tCPA of $95 based on historical data. The firm wants approximately 25 leads per month from paid search.
  • Numbers: Target monthly leads: 25. Target monthly spend: 25 × $95 = $2,375. Average daily budget required: $2,375 ÷ 30.4 days = $78.12/day. Because Google may spend up to 2× the daily budget on a single day ($78.12 × 2 = $156.24)[10], the firm must be comfortable with occasional daily charges up to $156.24 while the monthly total stays within $2,375 (78.12 × 30.4 = $2,375). Set daily budget at $78, not $39 — a $39 budget would allow only ~12 leads/month at $95 CPA, well below the 25-lead goal.
  • Decision: Set the average daily budget to $78, Target CPA to $95, and Primary conversion action to “Tax Consultation Request” (phone call or form fill, minimum 60 seconds call duration). Review on February 1, 2027 after the first full conversion cycle.
  • Why: Google’s tCPA guidance states advertisers should be comfortable with up to 2× the daily budget spend on a given day, so the daily budget must be set to deliver the monthly volume target, not halved as a safety measure.[10]

10. Audience Signals and Targeting

Search campaigns target by query intent first; audiences layer additional signals that Smart Bidding uses at auction time to adjust bids, personalize messaging, or qualify traffic. As of November 2026, audience signals in Search serve primarily as bid modifiers and observation layers — they do not restrict who sees ads unless Targeting (rather than Observation) mode is selected.[1]

Audience Targeting vs. Observation Mode

Mode Effect Use when
Observation Collects performance data by audience segment; Smart Bidding uses signals without restricting reach Default for most Search campaigns; provides data without limiting eligible queries
Targeting Restricts ad serving to users in the selected audience segment only Remarketing-only campaigns, customer match with a specific re-engagement objective, or where reach restriction is intentional

Recommended Audience Layers for Search

  • Customer Match lists: Upload first-party customer email lists for existing customer suppression, cross-sell, or win-back campaigns. Ensure data is hashed correctly and complies with US privacy requirements (CCPA in California; verify state-level obligations).
  • Remarketing lists (RLSA): Past site visitors, cart abandoners, or previous converters. Use in Observation mode to allow Smart Bidding to bid higher for high-intent return visitors.
  • In-market audiences: Google’s in-market segments layer purchase-intent signals. Add in Observation mode; do not use Targeting mode unless deliberately restricting to in-market users only.
  • Similar segments: Available for some account types; layer in Observation mode to expand Smart Bidding signal coverage.
  • Brand exclusions (AI Max): Google’s October 2026 announcements include controls to exclude people who recently searched for or interacted with a brand — particularly important when AI Max expands matching.[3]

Audience Signal Quality

Audience signals are only as valuable as the conversion data feeding Smart Bidding. Uploading customer lists without confirming reliable conversion tracking is operational will not improve performance. Prioritize measurement quality before adding audience complexity.[1]

Note for non-US markets: Audience targeting capabilities, Customer Match eligibility requirements, and consent/privacy rules differ materially in the EU (GDPR) and other jurisdictions; consult local legal counsel before applying US audience frameworks internationally.

Worked example (illustrative)

Using RLSA Observation to Identify High-Value Audience Segments

  • Setup: A San Francisco B2B software company spending $15,000/month on a non-brand Search campaign, using Target CPA of $210. The account has 3,000 CRM contacts and a site-visitor remarketing list of 12,000 users (30-day window).
  • Numbers: Current account metrics: 2,780 clicks/month at $5.42 CPC[6] = $15,068 spend; 8.18% conversion rate[6] = 227 conversions; CPA = $66.38 (well below the $210 target, suggesting the target is too loose). Adding the 12,000-user remarketing list in Observation mode for 60 days (November 15–January 15, 2027) reveals that users in the remarketing list convert at 14.2% vs. 6.1% for non-list users — a 2.33× higher conversion rate. Smart Bidding’s auction-time bid adjustment for this segment, if given the signal, can direct proportionally more spend toward these users.
  • Decision: Add the 12,000-user visitor remarketing list and the 3,000-contact Customer Match list to the campaign in Observation mode effective November 15, 2026. After 60 days, review segment-level CPA. If remarketing list CPA is ≤$140 vs. the $210 target, tighten the account-level Target CPA to $175 to extract incremental efficiency.
  • Why: Observation mode provides audience performance data without restricting reach, allowing Smart Bidding to use audience signals as additional auction-time inputs without reducing eligible traffic volume.[1]

11. Common Mistakes to Avoid

The following errors consistently undermine Search campaign performance in 2026. Each is directly traceable to a Google guidance principle or benchmark finding.

Mistake 1: Optimizing Toward the Wrong Conversion Action

Setting page views, ad clicks, or unqualified micro-conversions as Primary conversion actions causes Smart Bidding to optimize for the wrong outcome. Audit Primary vs. Secondary conversion action designations before changing any bid strategy.[1][4] Only actions that represent genuine business value — qualified leads, purchases, phone calls of meaningful duration, offline revenue — should be Primary.

Mistake 2: Setting an Aggressive tCPA or tROAS Target Without Historical Basis

A Target CPA or ROAS derived from an aspirational business goal rather than observed historical performance will restrict delivery and can starve the campaign of conversions. The benchmark cross-industry CPL of $60.98[7] is irrelevant if the account’s actual historical CPL in that vertical is $131.63.[7] Use historical performance as the starting point.

Mistake 3: Fragmenting Campaigns by Match Type or Device

Creating separate campaigns for exact, phrase, and broad versions of the same theme — or for mobile vs. desktop — is explicitly discouraged by Google’s 2026 guidance. Smart Bidding evaluates device and match-type signals at auction time. Fragmentation reduces data per campaign and impairs Smart Bidding’s optimization.[1][11]

Mistake 4: Neglecting Search Terms Review After Enabling Broad Match or AI Max

Broad match and AI Max expand query matching significantly. Without weekly search terms review and active negative keyword management, wasted spend on irrelevant queries can consume a material portion of the budget — in the plumbing example above, up to 63.8% in 14 days before intervention.

Mistake 5: Pinning Every RSA Asset into Fixed Positions

Fully pinned RSAs function identically to old expanded text ads — they eliminate the combinatorial testing that RSAs are designed to enable. Google notes that pinning reduces available combinations and typically lowers Ad Strength.[1][12] Reserve pinning for genuine compliance or legal requirements only.

Mistake 6: Ignoring the Budget vs. Rank Loss Distinction

Applying Quality Score fixes to a budget-limited campaign, or increasing budget for a rank-limited campaign, wastes time and money. The Conner Crowe benchmark shows 74.2% of accounts have budget loss exceeding rank loss.[8] Check the Search Lost IS (budget) and Search Lost IS (rank) columns before choosing a remedy.

Mistake 7: Building API Workflows Around Deprecated Features

Campaign-level broad match configurations and standalone Automatically Created Assets have been closed to new creation since August 3, 2026.[3] Search language targeting has been removed from Search campaigns as of late September 2026.[3] API scripts that attempt these operations will fail. Audit all automation workflows for deprecated operations immediately.

Mistake 8: Accepting AI Max Migration Without Preparation

Dynamic Search Ads migrate automatically to AI Max in February 2027.[3] Accepting forced migration without first exporting DSA targets, exclusions, search terms, and landing-page performance data, and without reviewing negative keywords and URL exclusions, risks significant traffic and quality disruption.

Worked example (illustrative)

Catching a Wrong Primary Conversion Action Before a Bid Strategy Change

  • Setup: An Austin real estate agency preparing to switch a lead generation campaign from Maximize Clicks to Target CPA. The account shows 340 “conversions” per month at an apparent CPA of $18.23. Before switching, the manager audits conversion actions.
  • Numbers: Of the 340 monthly “conversions”: 280 are tagged as “Page View — Property Listing” (a Secondary action accidentally marked Primary); 42 are “Contact Form Submitted” (the genuine lead event); 18 are “Phone Call — 60+ seconds.” True qualified leads = 42 + 18 = 60/month. True CPA = ($6,200 monthly spend) ÷ 60 = $103.33 per lead. At the cross-industry median CPL of $60.98[7], $103.33 is above average but plausible for real estate. Setting Target CPA at $18.23 (based on the inflated count) would set an impossible target and collapse delivery within days of activation.
  • Decision: Before switching bid strategy, remove “Page View — Property Listing” from Primary designation (move to Secondary or remove entirely); confirm “Contact Form Submitted” and “Phone Call — 60+ seconds” are the only Primary actions. Set Target CPA at $108 (a 4.5% buffer above the $103.33 true historical CPA). Activate Target CPA on November 18, 2026.
  • Why: Smart Bidding optimizes toward whatever is marked Primary — a wrong designation makes every downstream bid decision incorrect, and the resulting Target CPA would be based on an unachievable figure.[1][4]

12. What Changed Recently (Last 30 Days)

The changes below reflect Google Ads Search developments effective as of November 2, 2026. Each item carries a specific recommended action so the information can be acted on immediately.

AI Max for Search: Becoming the Default Direction

Google is positioning AI Max for Search as a one-click enhancement layer for Search campaigns. It expands query matching beyond existing keywords, customizes ad text, and uses Final URL Expansion to route users to a more relevant landing page than the explicitly supplied final URL.[5] Google’s internal 2025 data reports that non-retail advertisers activating AI Max see 14% more conversions or conversion value at a similar CPA or ROAS. This is a Google-reported vendor claim, not an independently audited figure; treat it as directional rather than guaranteed.

AI Max operates across three core components:

  • Search Term Matching: Finds additional relevant queries using keyword, creative, landing-page, and other campaign signals.
  • Text Customization: Generates or adapts headlines and descriptions from existing ads, keywords, and landing-page content.
  • Final URL Expansion: Sends users to a potentially more relevant page than the campaign’s declared final URL.

Test AI Max in a controlled experiment before enabling it across every campaign. Review search-term reporting weekly during the learning period, maintain comprehensive negative-keyword lists and brand controls, audit all generated assets, and confirm that conversion tracking, value rules, enhanced conversions, and offline conversion imports are reliable before increasing automation.

Worked example (illustrative)

Controlled AI Max Experiment for a Home Services Account

  • Setup: A Phoenix home-services account spending $9,000/month across three Search campaigns (HVAC installation, plumbing, electrical). The account uses Target CPA bidding at $85/lead and tracks phone calls (90+ seconds) and contact form submissions as Primary conversions. The manager wants to test AI Max on the HVAC installation campaign only.
  • Numbers: HVAC campaign: $3,200/month, 38 leads/month, CPA = $3,200 ÷ 38 = $84.21. Google reports a 14% conversion lift benchmark for non-retail AI Max activations (vendor claim). Projected leads if benchmark holds: 38 × 1.14 = 43 leads/month. Projected CPA at same spend: $3,200 ÷ 43 = $74.42. Minimum experiment duration to detect signal: 28 days covering at least 2 full weekly cycles and enough conversion volume for statistical validity (target ≥ 30 conversions per arm, so both control and test arms need $3,200/month or equivalent split). Set success threshold: CPA ≤ $90 (6.9% above current Target CPA) with ≥ 35 leads in the test arm.
  • Decision: Launch a 50/50 Google Ads experiment on the HVAC campaign beginning December 1, 2026, with AI Max enabled in the test arm and the current keyword-only setup retained in the control arm. Set experiment end date to December 28, 2026. Review search-term report every Monday during the test; add any irrelevant query themes as negative keywords within 48 hours.
  • Why: A controlled experiment isolates AI Max’s incremental effect from seasonal variation and prevents budget risk across all three campaigns simultaneously while the account gathers real performance data against the vendor-reported 14% benchmark.[5]

New AI Max Testing and Planning Tools

Google’s October 2026 announcements introduced structured testing capabilities for evaluating AI Max and broader Search changes before full deployment. Supported test areas include budget changes, ROI-target changes, brand controls, location settings, campaign configuration changes, and AI Max activation itself.

Google also introduced Smart Bidding Exploration, which allows more flexible ROAS targets to pursue additional conversion opportunities beyond an advertiser’s existing demand. This is best used when the account is not budget-limited and incremental volume at a slightly looser efficiency target is commercially acceptable.

Use experiments or staged rollouts with predefined success criteria: a primary KPI (conversion value, qualified leads, or profit—not clicks alone), an efficiency guardrail (CPA, ROAS, contribution margin, or lead quality), a minimum duration long enough to cover conversion lag and normal weekly variation, and separate brand and non-brand reporting before evaluating incremental performance.

Dynamic Search Ads Upgrading to AI Max — Migration Deadline February 2027

Google has confirmed that legacy Dynamic Search Ads will automatically upgrade to AI Max, with forced migration scheduled for February 2027. Campaigns using campaign-level broad match migrate with Search Term Matching enabled, while Text Customization and Final URL Expansion remain off. Campaigns migrating from standalone Automatically Created Assets retain Text Customization and Search Term Matching, while Final URL Expansion remains off.

Before migration: export current DSA targets, exclusions, search terms, ads, and landing-page performance; document which pages each DSA campaign drives traffic to; review and expand negative keywords; check URL exclusions and landing-page content quality; and decide whether to voluntarily upgrade and test AI Max before the forced migration date.

Legacy Campaign-Level Broad Match and Automatically Created Assets Closed to New Creation

From August 3, 2026, advertisers can no longer create new campaign-level broad-match configurations or legacy standalone Automatically Created Assets through the Google Ads UI, Editor, or API. Existing configurations have been subject to automatic migration beginning September 2026. Do not build new workflows around these legacy settings. For new automation, use the AI Max controls and APIs supported by the current Google Ads version, and update scripts that assume campaign-level broad match or legacy ACA creation is available.

Search Language Targeting Has Been Removed

Google removed campaign-level language targeting for Search and AI Max for Search campaigns in late September 2026. Google now automatically matches ads based on the language of the ad creatives, the language of the landing pages, and the user’s language preferences. Attempts to set CampaignCriterion.language for Search campaigns now fail with ContextError.OPERATION_NOT_PERMITTED_FOR_CONTEXT.

Stop setting language criteria when creating or updating Search campaigns. Audit ad groups and landing pages for language consistency. Separate materially different language experiences into appropriate campaign or asset structures. Update Google Ads API scripts and validation rules to avoid failed operations. Note: this change applies globally; advertisers outside the US managing multilingual campaigns should audit all Search campaign language criteria immediately.

RSA Headlines Now Eligible for Sitelink Positions

Google announced it is using AI to feature RSA headlines in sitelink positions, while respecting existing pinning controls. This gives existing RSA assets additional opportunities to appear and may improve engagement, but it also means every headline must be treated as a possible navigation element, not merely as ad copy. Write headlines that remain meaningful when shown outside the normal headline sequence. Avoid vague, repetitive, or legally risky headlines. Review sitelink and headline reporting for unintended combinations.

Expanded Search Reporting and Controls for AI Max

Google has expanded access to search-terms reporting, channel-performance reporting, diagnostics, brand controls, and controls to exclude people who recently searched for or interacted with a brand. These controls are especially important as AI Max expands matching beyond explicitly selected keywords. Establish a recurring AI Max review process covering new query categories, brand versus non-brand traffic, search terms with weak conversion quality, queries outside geographic or service eligibility, generated copy quality, expanded landing-page destinations, and incremental versus cannibalized conversions.

Search Lift Measurement Expanded in the API

Google has expanded Search Lift measurement support in the Google Ads API. Advertisers can now query Search Lift keyword-bundle data and additional metrics including relative search lift, normalized incremental searches, and Search Lift impressions. This supports a shift away from judging automation solely by last-click or platform-reported conversions. Where eligible, combine conversion and conversion-value reporting with search-term and asset diagnostics, brand/non-brand analysis, incrementality testing, offline revenue or qualified-lead data, and profit-based bidding targets.

Worked example (illustrative)

Prioritizing the November 2026 Change Checklist for a Multi-Campaign Account

  • Setup: A Chicago B2B software account running 6 Search campaigns: 1 brand, 2 non-brand lead gen, 1 competitor, 1 legacy DSA, and 1 campaign using the now-deprecated campaign-level broad match. Monthly spend: $22,000. Primary conversion: demo request form submission, tracked via Google tag with enhanced conversions enabled.
  • Numbers: Legacy DSA campaign: $2,800/month, 19 demo requests/month, CPA = $2,800 ÷ 19 = $147.37. Forced DSA-to-AI Max migration deadline: February 2027 = approximately 13 weeks away. Campaign-level broad match deprecated August 3, 2026 — new creation already blocked; existing configuration subject to automatic migration from September 2026 onward. Language targeting removal: September 2026 — if any of the 6 campaigns still carry language criteria, API calls will fail. Negative keyword list last reviewed: no documented date in account change history.
  • Decision: Week of November 3, 2026: (1) Export DSA search terms, targets, exclusions, and landing-page performance. (2) Audit all 6 campaigns for language criteria; remove any remaining criteria via Editor before November 10, 2026. (3) Update API scripts to remove language-targeting operations. (4) Review and expand negative keyword lists across all campaigns; document additions with dates. (5) Schedule voluntary AI Max experiment on the DSA campaign beginning December 1, 2026, rather than waiting for the February 2027 forced migration.
  • Why: Acting 13 weeks ahead of the forced migration deadline provides two full 28-day test cycles to validate AI Max performance against the existing DSA CPA of $147.37 before control is removed, and eliminates API failure risk from the already-deprecated language targeting.[5]

Priority Checklist for November 2026

  • Check every Search campaign for AI Max eligibility, migration status, and enabled components.
  • Export and document DSA and legacy broad-match performance before further automatic migration.
  • Update API and automation workflows to remove manual Search language-targeting operations.
  • Audit negative keywords, brand controls, URL exclusions, and Final URL Expansion settings.
  • Run controlled AI Max experiments with conversion-quality guardrails and predefined success thresholds.
  • Review generated RSA assets and landing pages weekly during the learning period.
  • Separate brand and non-brand reporting before evaluating incremental performance claims.
  • Update forecasts to account for potentially higher query volume, CPCs, and conversion lag when AI Max is active.

Related reading

Or skip the work and get professional Search campaign setup, without an agency retainer.

References

  1. [1] https://support.google.com/google-ads/answer/14996489?hl=en support.google.com
  2. [2] https://support.google.com/google-ads/answer/12159290?hl=en support.google.com
  3. [3] https://support.google.com/google-ads/answer/10195720?hl=en support.google.com
  4. [4] https://support.google.com/google-ads/answer/14996994?hl=en-AU support.google.com
  5. [5] https://support.google.com/google-ads/answer/12158267?hl=en support.google.com
  6. [6] https://www.fibbler.co/blog/google-ads-benchmarks www.fibbler.co
  7. [7] https://www.benchmarketing.org/benchmarks/google-ads/by-industry www.benchmarketing.org
  8. [8] https://connercrowe.com/benchmarks/google-ads-account-benchmarks/ connercrowe.com
  9. [9] https://www.benchmarketing.org/benchmarks/metrics/impression-share www.benchmarketing.org
  10. [10] https://services.google.com/fh/files/blogs/google-keywords-to-the-wise.pdf services.google.com
  11. [11] https://support.google.com/google-ads/answer/7478529?hl=en support.google.com
  12. [12] https://support.google.com/google-ads/answer/13389795?hl=en support.google.com
  13. [13] https://business.google.com/us/accelerate/resources/articles/reaching-the-right-customers/ business.google.com
  14. [14] https://support.google.com/google-ads/answer/14998308?hl=en support.google.com
Share the Post:

Related Posts