Open-source SEO, GEO, and marketing skills for AI agents.
feat: update Google Ads skills for 2026 policy reality + add policy freshness system
Tongdb09ff577f66
diff --git a/google-ads/ads-audit/SKILL.md b/google-ads/ads-audit/SKILL.mdindex fd0664be..8fce2c61 100644--- a/google-ads/ads-audit/SKILL.md+++ b/google-ads/ads-audit/SKILL.md@@ -29,35 +29,48 @@ Read these reference documents during analysis for expert-level context:Read these before starting Phase 2 analysis. They contain the numeric thresholds that separate a generic audit from an expert one.+## Step 0: Policy Freshness Check++Before auditing, verify that the policy assumptions underpinning this audit are current.++1. Read `../shared/policy-registry.json` and check each entry: if `last_verified` + `stale_after_days` < today's date, the entry is stale.+2. **High-volatility stale entries:** Use WebSearch to check for recent Google Ads changes related to each stale entry's `area` (e.g., "Google Ads broad match behavior changes 2026"). Compare findings against the `assumption` field. If discrepancies are found:+ - Display a warning banner at the top of the audit: `⚠️ Policy drift detected: [area] — [brief description of what changed]. Recommendations in this area may need manual verification.`+ - Suggest updating `policy-registry.json` with corrected assumptions and today's date.+3. **Moderate-volatility stale entries:** Note them in the audit output as an informational line (e.g., "ℹ️ [area] last verified [date] — may warrant a check") but do not block the audit.+4. **Stable stale entries:** Skip — these rarely change.++If no high-volatility entries are stale, proceed directly to Phase 1 with no output from this step.+## Phase 1: Pull Account DataGather everything in parallel before asking the user a single question. The goal is to show up informed.-**Pull all data calls simultaneously to minimize wait time. Use parallel tool calls — every call below is independent and can run at the same time.**+**Use the adaptive data fetching algorithm from `../shared/gaql-cookbook.md`.** The approach varies by account size — read the cookbook's "Adaptive data fetching algorithm" section. Here's the summary:++### Phase 1A: Account basics + sizing probe (parallel)++Pull these simultaneously — they don't require campaign IDs and they tell you the account size:++- `getAccountInfo` — business name, currency, timezone+- `getAccountSettings` — auto-tagging, tracking template, conversion setup+- `listCampaigns(limit: 100)` — all campaigns with spend, clicks, conversions **(also serves as the sizing probe — always pass limit: 100, default is 20)**+- `getConversionActions` — what conversions are set up+- `getRecommendations` — Google's optimization suggestions-**Account basics:**-- `mcp__adsagent__getAccountInfo` — business name, currency, timezone-- `mcp__adsagent__getAccountSettings` — auto-tagging, tracking template, conversion setup+### Phase 1B: Adaptive data pull (depends on account size)-**Campaign structure:**-- `mcp__adsagent__listCampaigns` — all campaigns with spend, clicks, conversions-- `mcp__adsagent__listAdGroups` — ad groups per campaign (focus on top-spend campaigns)+Count the enabled campaigns with spend from Phase 1A, then follow the cookbook:-**Performance data:**-- `mcp__adsagent__getCampaignPerformance` — daily trends for the last 30 days-- `mcp__adsagent__getKeywords` — top keywords with quality scores, bids, conversions-- `mcp__adsagent__getSearchTermReport` — actual search queries triggering ads-- `mcp__adsagent__getImpressionShare` — search IS, top IS, budget-lost IS, rank-lost IS+- **1-3 campaigns:** Run all 7 standard GAQL queries in parallel. Also run `getCampaignSettings(campaignId)` and `listAds(campaignId)` per campaign (needed for Display Network detection, bidding strategy assessment, and ad copy scoring). Done.+- **4-10 campaigns:** Run all 7 GAQL queries + `getCampaignSettings` and `listAds` for top 3-5 campaigns by spend. If keywords or search terms hit the 50-row limit, supplement with `getKeywords`/`getSearchTermReport` for the top 2-3 campaigns.+- **10+ campaigns:** Use the tiered approach — GAQL for account-wide overview (impression share, ad groups, negatives, daily performance), then per-campaign helpers (`getCampaignSettings`, `getKeywords`, `getSearchTermReport`, `listAds`) for the top campaigns that cover 80% of spend.-**Ads and conversions:**-- `mcp__adsagent__listAds` — ad copy, status, and per-ad metrics-- `mcp__adsagent__getConversionActions` — what conversions are set up and how they're configured-- `mcp__adsagent__getNegativeKeywords` — existing negative keyword coverage+### Fallback-**Recommendations:**-- `mcp__adsagent__getRecommendations` — Google's optimization suggestions (evaluate critically, don't blindly follow)+If `runGaqlQuery` errors or is unavailable, fall back to per-campaign helper tools for each active campaign, run in parallel.-**Minimum data for a meaningful audit:** `listCampaigns`, `getKeywords`, and `getImpressionShare` must return data. If the account has zero campaigns or zero spend, skip to Phase 3 (business context) — there's nothing to audit yet, but the context is still valuable for account setup.+**Minimum data for a meaningful audit:** Campaign list, keyword data, impression share, and conversion actions must return data. If the account has zero campaigns or zero spend, skip to Phase 3 (business context).## Phase 2: Analyze and Score@@ -169,7 +182,7 @@ Read `references/account-health-scoring.md` for the detailed rubric per dimensio- Check pin strategy: H1 should typically be pinned to the most relevant service+location headline- Identify ad groups with CTR below industry benchmark — these need copy refresh-**6. Impression share** (Score 0-5)+**6. Impression share** (Score 0-5) — **Data limit:** `getImpressionShare` supports max 90 days (not 365 like other tools). For GAQL impression share queries, the same 90-day practical limit applies.| Score | Criteria ||-------|----------|@@ -492,7 +505,7 @@ If wasted spend > 15% of total:### Landing Page Problems-If CTR is healthy (>3%) but conversion rate is below 2% across multiple ad groups:+If CTR is above industry benchmark (check `../ads/references/industry-benchmarks.md`) but conversion rate is below industry average across multiple ad groups:> "Your ads get clicks but conversions are low — the landing pages likely don't match the ad promise. Run `/ads-landing` to audit keyword-to-landing-page alignment."diff --git a/google-ads/ads-audit/references/account-health-scoring.md b/google-ads/ads-audit/references/account-health-scoring.mdindex 62e69647..8456a710 100644--- a/google-ads/ads-audit/references/account-health-scoring.md+++ b/google-ads/ads-audit/references/account-health-scoring.md@@ -25,19 +25,19 @@ Structured scoring methodology for Google Ads account audits. Seven health dimen| 0 | None | No conversion tracking installed || 1 | Basic | Tag fires but no values assigned; only 1 conversion action || 2 | Partial | Multiple conversion actions with values, but gaps: missing phone calls OR form submissions OR key pages |-| 3 | Solid | All key actions tracked (form, phone, purchase); values assigned; conversion window set correctly |-| 4 | Advanced | Score 3 + enhanced conversions enabled; conversion linker tag present; GCLID auto-tagging on |-| 5 | Best-in-class | Score 4 + offline conversion import OR store visit tracking; Consent Mode v2 implemented; data-driven attribution model |+| 3 | Solid | All key actions tracked (form, phone, purchase); values assigned; conversion window set correctly; Consent Mode v2 implemented (mandatory for EU since March 2024) |+| 4 | Advanced | Score 3 + enhanced conversions enabled; conversion linker tag present; GCLID auto-tagging on; Consent Mode v2 verified across all regions |+| 5 | Best-in-class | Score 4 + offline conversion import OR store visit tracking; server-side tagging implemented; data-driven attribution model |### Conversion Tracking Red Flags| Signal | Severity | What It Means ||--------|----------|---------------|-| 0 conversions in 30 days with spend >$500 | Critical | Tracking likely broken or missing |+| 0 conversions in 30 days with spend >5x account CPA | Critical | Tracking likely broken or missing || Conversion rate >50% | Critical | Tag firing on page load, not actual conversions || All conversions = "Website" (no action names) | High | Default tracking, no meaningful segmentation || "Include in conversions" set to YES for micro-conversions | Medium | Inflated conversion counts, misleading CPA |-| Last-click attribution on high-funnel campaigns | Medium | Under-crediting awareness campaigns |+| Last-click attribution on any campaigns | High | Deprecated since 2023 — data-driven attribution (DDA) is now the only model for new conversion actions. Migrate existing conversion actions to DDA immediately |---@@ -56,7 +56,7 @@ Structured scoring methodology for Google Ads account audits. Seven health dimen| Check | Passing Threshold | Failing Signal ||-------|------------------|----------------|-| Keywords per ad group | 10-20 | >30 per ad group |+| Keywords per ad group | 5-15 (broad match: 5-10; exact/phrase: 10-20) | >30 per ad group || Ad groups per campaign | 5-15 | >25 or just 1 || Brand vs. non-brand separation | Separate campaigns | Mixed in same campaign || Campaign naming convention | Consistent pattern | No pattern (e.g., "Campaign 1", "test") |@@ -70,11 +70,11 @@ Structured scoring methodology for Google Ads account audits. Seven health dimen| Score | Level | Criteria ||-------|-------|----------|| 0 | Failing | Average QS <4; >50% zombie keywords; no match type strategy |-| 1 | Poor | Average QS 4-5; 30-50% zombie keywords; all broad match |-| 2 | Below Average | Average QS 5-6; 20-30% zombie keywords; mostly broad match with some phrase |+| 1 | Poor | Average QS 4-5; 30-50% zombie keywords; broad match without Smart Bidding or negatives |+| 2 | Below Average | Average QS 5-6; 20-30% zombie keywords; broad match with Smart Bidding but insufficient negatives || 3 | Good | Average QS 6-7; <20% zombie keywords; intentional match type strategy; negatives in place |-| 4 | Strong | Average QS 7-8; <10% zombie keywords; match type tiering (exact for high-value, phrase/broad for discovery) |-| 5 | Excellent | Average QS 8+; <5% zombie keywords; SKAGs or tightly themed groups; bid adjustments by match type |+| 4 | Strong | Average QS 7-8; <10% zombie keywords; intentional match type strategy aligned with Smart Bidding |+| 5 | Excellent | Average QS 8+; <5% zombie keywords; tightly themed ad groups aligned to Smart Bidding signals; comprehensive negative coverage |### Keyword Sub-Metric Thresholds@@ -82,7 +82,7 @@ Structured scoring methodology for Google Ads account audits. Seven health dimen|------------|----------|------------|----|---------|| Average Quality Score | <4 | 4-5 | 6-7 | 8+ || Zombie keywords (0 impressions, 90 days) | >50% | 30-50% | 10-30% | <10% |-| % Exact match keywords | 0% | <20% | 20-50% | 50%+ for top converters |+| Match type intentionality | No strategy | Single match type only | Some match type variation | Deliberate match type strategy aligned with campaign goals and Smart Bidding || % Keywords with QS <5 | >40% | 25-40% | 10-25% | <10% || Negative keyword count vs. active keywords | 0 negatives | <10% ratio | 10-30% ratio | 30%+ ratio |@@ -199,7 +199,7 @@ Structured scoring methodology for Google Ads account audits. Seven health dimen| 2 | Below Average | 15-20% wasted spend; CPA 1.5-2x benchmark || 3 | Average | 10-15% wasted spend; CPA within 1-1.5x benchmark; top 20% keywords drive 50%+ conversions || 4 | Good | 5-10% wasted spend; CPA at or below benchmark; strong conversion concentration |-| 5 | Excellent | <5% wasted spend; CPA well below benchmark; top 20% keywords drive 60%+ conversions; no keyword spending >$200 with 0 conversions |+| 5 | Excellent | <5% wasted spend; CPA well below benchmark; top 20% keywords drive 60%+ conversions; no keyword spending >2x account CPA with 0 conversions |### Wasted Spend Calculation@@ -222,7 +222,7 @@ Wasted Spend % = (Spend on keywords with 0 conversions AND >10 clicks) / Total S| Top 20% keywords' share of conversions | >50% | 30-50% | <30% || Top 20% keywords' share of spend | <50% | 50-70% | >70% with low conversion share || Single keyword share of total spend | <15% | 15-30% | >30% (concentration risk) |-| Keywords spending >$100 with 0 conversions | 0-2 | 3-5 | >5 |+| Keywords spending >1x account CPA with 0 conversions | 0-2 | 3-5 | >5 |---@@ -277,7 +277,7 @@ For each dimension scored below 3, the single highest-impact fix:| Search Term Quality | Run first-ever search term review; add 20+ obvious negatives | Build negative keyword lists by theme; block top wasters | Set up bi-weekly review cadence; mine converting terms || Ad Copy | Create RSA with 8+ headlines for every ad group | Add headlines to reach 8+; ensure 3+ distinct angles | Add second RSA for A/B testing; improve headline variety || Impression Share | Check if budget is depleting before noon; adjust bids down or narrow targeting | Identify top budget-loss campaign; increase budget or narrow geo/schedule | Improve QS on rank-lost keywords; test bid strategies |-| Spend Efficiency | Pause all keywords with >$100 spend and 0 conversions immediately | Set max CPC caps; pause keywords with >20 clicks and 0 conversions | Reallocate budget from low-efficiency to high-efficiency keywords |+| Spend Efficiency | Pause all keywords with spend >1x account CPA and 0 conversions immediately | Set max CPC caps; pause keywords with >20 clicks and 0 conversions | Reallocate budget from low-efficiency to high-efficiency keywords |---diff --git a/google-ads/ads-copy/SKILL.md b/google-ads/ads-copy/SKILL.mdindex fdfe2d6c..2a0028f0 100644--- a/google-ads/ads-copy/SKILL.md+++ b/google-ads/ads-copy/SKILL.md@@ -85,14 +85,18 @@ Gather these fields. Don't ask them as a rigid checklist — pull what you can f### Bootstrapping from existing data-Before asking the user anything, try to fill fields from what's already available:+Before asking the user anything, try to fill fields from what's already available. Start with account-level calls, then use campaign IDs from `listCampaigns` for per-campaign tools:-- `mcp__adsagent__getAccountInfo` → business name, location hints-- `mcp__adsagent__listCampaigns` → service categories, geo targeting-- `mcp__adsagent__listAds` → current voice, headlines in use-- `mcp__adsagent__getKeywords` → keyword landscape-- `mcp__adsagent__getSearchTermReport` → real user language, long-tail opportunities-- `mcp__adsagent__getImpressionShare` → competitive pressure signals+1. **Account-level (parallel, no campaignId needed):**+ - `getAccountInfo` → business name, location hints+ - `listCampaigns(limit: 100)` → service categories from campaign names, identify top campaigns by spend++2. **Per-campaign (parallel, requires campaignId from step 1):**+ - `listAds(campaignId)` → current voice, headlines in use+ - `getKeywords(campaignId)` → keyword landscape+ - `getSearchTermReport(campaignId)` → real user language, long-tail opportunities+ - `getImpressionShare(campaignId)` → competitive pressure signals (max 90 days)+ - `getCampaignSettings(campaignId)` → geo targeting, network settingsPresent what you found and ask the user to confirm/correct/fill gaps. This is faster and more accurate than starting from zero.@@ -194,11 +198,19 @@ Cross-reference against the business context — if the user says "write copy foPull data before writing — copy should be grounded in what converts, not guesses.-**Current ad performance:**-- `mcp__adsagent__listAds` — existing headlines/descriptions and their metrics-- `mcp__adsagent__getKeywords` — active keywords-- `mcp__adsagent__getSearchTermReport` — actual user queries (reveals real language and intent)-- `mcp__adsagent__getCampaignPerformance` — CTR/conversion benchmarks to beat+**Current ad performance (use GAQL for multi-campaign research, see `../shared/gaql-cookbook.md`):**++If the user specified a single campaign/ad group, use helper tools directly. If researching across campaigns (common for brand-wide copy refresh), use GAQL:+- GAQL "Ad copy" query → existing headlines/descriptions and their metrics across all campaigns+- GAQL "Keywords with QS" query → active keywords and what's converting+- GAQL "Search terms" query → actual user queries (reveals real language and intent)+- GAQL "Impression share" query → CTR/conversion benchmarks to beat++For single-campaign work, the per-campaign helpers are simpler:+- `listAds(campaignId)` — existing headlines/descriptions+- `getKeywords(campaignId)` — active keywords+- `getSearchTermReport(campaignId)` — actual user queries+- `getCampaignPerformance(campaignId)` — CTR benchmarks**Use seasonality context.** If business context shows peak months, factor that into copy urgency. During slow months, lean on evergreen value props. During peaks, lean on scarcity and timeliness.@@ -311,7 +323,7 @@ After user approves a variant, push it live:- **Update existing:** `mcp__adsagent__updateAdAssets` — replace headlines/descriptions on a live ad- **Enable when ready:** `mcp__adsagent__enableAd`-Always confirm before any write operation. Note the `changeId` returned — user can undo within 7 days via `mcp__adsagent__undoChange`.+Always confirm before any write operation. Note the `changeId` returned — user can undo within 7 days (if the entity hasn't been modified since) via `mcp__adsagent__undoChange`.### 6. A/B test (if running one)@@ -356,7 +368,7 @@ After deciding a winner: pause the loser with `mcp__adsagent__pauseAd`, keep the2. **Research before writing.** Always pull current performance data. Don't write copy in a vacuum.3. **Character limits are hard.** Count every headline (<=30) and description (<=90). No exceptions. When in doubt, count again.4. **Never deploy without confirmation.** Show the exact copy, get a yes, then create/update.-5. **Note changeIds.** Every write returns one. Tell the user they can undo within 7 days.+5. **Note changeIds.** Every write returns one. Tell the user they can undo within 7 days (if the entity hasn't been modified since).6. **Ground copy in conversion data.** If conversion data is available, use the language that converts. Customer language > marketing language.7. **Seasonal awareness.** Check the business context for peak/slow months. Adjust urgency and messaging accordingly.8. **Defer to /ads for account management.** This skill writes copy and deploys ads. For bid/budget/keyword work, use `/ads`.diff --git a/google-ads/ads-copy/references/rsa-best-practices.md b/google-ads/ads-copy/references/rsa-best-practices.mdindex 2b1ebb72..dea4be66 100644--- a/google-ads/ads-copy/references/rsa-best-practices.md+++ b/google-ads/ads-copy/references/rsa-best-practices.md@@ -2,6 +2,8 @@Expert knowledge for writing, testing, and optimizing Responsive Search Ads. Specific character limits, formula tables, pinning strategies, and performance benchmarks.+> **PMax cross-reference:** For accounts running Performance Max campaigns alongside Search, coordinate RSA messaging with PMax asset groups. PMax may serve text ads on Search inventory — ensure your RSA headlines and descriptions complement (not duplicate) PMax asset group text assets. See PMax documentation for asset group best practices.+---## Character Limits@@ -134,7 +136,7 @@ Should I pin this headline?| +-- YES --> Leave unpinned (let Google optimize)| +-- NO --> Continue|-+-- Do you have 100+ conversions/month in this ad group?++-- Do you have 50+ conversions/month at the campaign level?| +-- YES --> Leave unpinned (enough data for Google to optimize)| +-- NO --> Consider pinning your best 2-3 headlines to Positions 1-2|@@ -147,9 +149,9 @@ DEFAULT --> Leave unpinned. Google's ML needs freedom to test combinations.|-----------|----------------------|----------------------|| 0 pins | Neutral to positive | Maximum reach || 1-2 pins | Minor reduction | Minimal impact |-| 3-5 pins | Moderate reduction | 10-15% fewer impressions |-| 6+ pins | Significant reduction | 20-40% fewer impressions |-| All pinned | Severe; ad strength drops to "Poor" | 40-70% fewer impressions |+| 3-5 pins | Moderate reduction | Noticeable impression reduction (Google has not published exact figures) |+| 6+ pins | Significant reduction | Substantial impression reduction — diminishes RSA's optimization ability |+| All pinned | Severe; ad strength drops to "Poor" | Major impression reduction — effectively negates the RSA format's advantages |### Pinning Rules@@ -219,9 +221,9 @@ DEFAULT --> Leave unpinned. Google's ML needs freedom to test combinations.| Metric | Minimum Threshold | Ideal Threshold ||--------|------------------|-----------------|-| Clicks per variant | 100 | 250+ |+| Clicks per variant | 200 | 500+ || Duration | 14 days | 30 days |-| Conversions per variant | 30 | 50+ |+| Conversions per variant | 50 | 100+ || Impressions per variant | 1,000 | 5,000+ |### When to Call a Winner Early@@ -291,7 +293,7 @@ DEFAULT --> Leave unpinned. Google's ML needs freedom to test combinations.| Too many similar headlines | Google shows redundant combinations; wastes ad slots | Write from 5+ distinct angles || All headlines same length | Looks repetitive; Google prefers mixing short + long | Vary between 15-30 chars || No CTA headlines | Misses click intent when Google picks non-CTA combination | Include 2-3 CTA headlines |-| Pinning everything | Reduces Google's optimization ability by 40-70% | Pin only what is legally or brand-required |+| Pinning everything | Substantially reduces Google's optimization ability | Pin only what is legally or brand-required || Ignoring ad strength entirely | Missing easy wins from insufficient asset diversity | Aim for "Good" minimum unless justified || Only 3 headlines provided | Minimum assets = minimum reach and no testing | Always provide 8+ headlines || Duplicate keywords in headlines | Wastes headline slots repeating same keyword | Each headline should add new info |diff --git a/google-ads/ads/SKILL.md b/google-ads/ads/SKILL.mdindex ee435230..bfcd4281 100644--- a/google-ads/ads/SKILL.md+++ b/google-ads/ads/SKILL.md@@ -112,7 +112,7 @@ Manage Google Ads campaigns via the MCP server.- **getTrackingTemplate** — Current tracking template at account/campaign/ad-group/ad level- **listAdGroups** — Ad groups in a campaign with metrics- **listAds** — Ads in a campaign/ad group with copy, URLs, status, metrics-- **getImpressionShare** — Search/top/abs-top IS and budget/rank-lost IS+- **getImpressionShare** — Search/top/abs-top IS and budget/rank-lost IS (max 90 days, not 365)- **getConversionActions** — Conversion actions and settings- **getAccountSettings** — Auto-tagging, tracking template, conversion tracking IDs- **getCampaignSettings** — Bidding, network, locations, schedule@@ -120,20 +120,21 @@ Manage Google Ads campaigns via the MCP server.- **getRecommendations** — Google optimization recommendations### Write (mutates the account — always confirm with user first)-All write tools return a `changeId` on success. Use this with `undoChange` to reverse the operation within 7 days.+All write tools return a `changeId` on success. Use this with `undoChange` to reverse the operation within 7 days (only if the entity hasn't been modified since).- **pauseKeyword** — Stop a keyword-- **enableKeyword** — Re-enable a paused keyword+- **enableKeyword** — Re-enable a paused keyword (needs adGroupId + criterionId only — no campaignId, unlike pauseKeyword)- **addKeyword** — Add a new keyword to an ad group- **updateBid** — Change CPC bid (manual/enhanced CPC only, max 25% change)-- **addNegativeKeyword** — Block irrelevant search terms (phrase match)+- **addNegativeKeyword** — Block irrelevant search terms (always phrase match at campaign level — no exact match or account-level option)- **removeNegativeKeyword** — Remove a negative keyword-- **updateCampaignBudget** — Change daily budget (max 50% change)-- **createCampaign** — Create a full paused search campaign+- **updateCampaignBudget** — Change daily budget (max 50% change, min $1/day)+- **createCampaign** — Create a full paused search campaign. Headlines (3-15, max 30 chars each), descriptions (2-4, max 90 chars each), finalUrl required. Default bidding: MAXIMIZE_CONVERSIONS- **pauseCampaign** — Pause all ads in a campaign- **enableCampaign** — Re-enable a paused campaign+- **removeCampaign** — **PERMANENTLY** delete a campaign and all its ad groups, ads, keywords (cannot be undone — use pauseCampaign instead in almost all cases)- **setTrackingTemplate** — Set/clear tracking template- **createAdGroup** — Create a new ad group-- **createAd** — Create a new Responsive Search Ad+- **createAd** — Create a new Responsive Search Ad. Headlines (3-15, max 30 chars each), descriptions (2-4, max 90 chars each), finalUrl required- **pauseAd** — Pause an ad- **enableAd** — Re-enable an ad- **updateAdFinalUrl** — Change an ad's landing page URL@@ -141,7 +142,7 @@ All write tools return a `changeId` on success. Use this with `undoChange` to re- **bulkUpdateBids** — Update up to 50 keyword bids in one call (each capped at 25% change)- **bulkPauseKeywords** — Pause up to 100 keywords in one call (partial success possible)- **bulkAddKeywords** — Add up to 100 keywords to an ad group in one call (partial success possible)-- **moveKeywords** — Move keywords between ad groups in the same campaign (adds to destination first, pauses source on success, rolls back on failure)+- **moveKeywords** — Move keywords between ad groups in the same campaign (adds to destination first, pauses source on success, rolls back on failure). **Match type defaults to PHRASE — does NOT inherit from source.** Specify matchType explicitly to preserve original match types. Max 100 keywords.- **renameCampaign** — Rename a campaign- **renameAdGroup** — Rename an ad group- **updateCampaignSettings** — Update network targeting (Google Search, Search Partners, Display Network) and/or location targeting (add/remove geo targets by geo target constant ID, e.g. '2840' for US, '200840' for Seattle-Tacoma DMA). Also supports negative location targeting (exclusions).@@ -280,8 +281,8 @@ When interpreting Google Ads data, apply these specific rules. Every recommendat| QS Range | Monthly Spend | Action | Priority ||----------|--------------|--------|----------|-| 1-4 | >$100/month | Priority fix — read `references/quality-score-framework.md` for diagnostic tree | Critical |-| 1-4 | <$100/month | Fix if keyword is strategically important, otherwise pause and reallocate budget | Medium |+| 1-4 | >1x account CPA/month | Priority fix — read `references/quality-score-framework.md` for diagnostic tree | Critical |+| 1-4 | <1x account CPA/month | Fix if keyword is strategically important, otherwise pause and reallocate budget | Medium || 5-6 | Any | Monitor — improve landing page relevance and ad copy match. Check QS subcomponents (expected CTR, ad relevance, landing page experience) to identify the bottleneck | Low || 7-8 | Any | Healthy — focus on scaling. Small QS gains here have diminishing returns | None || 9-10 | Any | Excellent — do not touch QS factors. Focus entirely on bid and budget optimization | None |@@ -303,10 +304,10 @@ Evaluate every keyword against the account's average CPA. If the account has no| CPA 50-100% of account avg | Maintain current bid. Monitor weekly | Healthy, contributing keyword || CPA 100-150% of account avg | Review search terms for this keyword. Tighten match type or add negatives | Borderline — often fixable with better targeting || CPA > 150% of account avg | Decrease bid 15-25%. If CPA > 200% avg after 2 weeks, pause | Underperformer dragging down account |-| 0 conversions, >$200 spend | Pause immediately OR move to exact match with 25% lower bid | Enough data to conclude this keyword doesn't convert at current targeting |-| 0 conversions, $100-200 spend, QS > 6 | Give 2 more weeks. Check landing page alignment and search term relevance | May need more data — QS suggests the ad/page are relevant |-| 0 conversions, $100-200 spend, QS < 5 | Pause. QS + no conversions = wrong keyword or wrong landing page | Two signals pointing the same direction |-| 0 conversions, <$100 spend | Too early to judge on conversions. Evaluate CTR and search term quality instead | Insufficient data for conversion-based decisions |+| 0 conversions, spend >2x account CPA | Pause immediately OR move to exact match with 25% lower bid | Enough data to conclude this keyword doesn't convert at current targeting |+| 0 conversions, spend 1-2x account CPA, QS > 6 | Give 2 more weeks. Check landing page alignment and search term relevance | May need more data — QS suggests the ad/page are relevant |+| 0 conversions, spend 1-2x account CPA, QS < 5 | Pause. QS + no conversions = wrong keyword or wrong landing page | Two signals pointing the same direction |+| 0 conversions, spend <1x account CPA | Too early to judge on conversions. Evaluate CTR and search term quality instead | Insufficient data for conversion-based decisions || 0 impressions for 30+ days | Pause — this is a zombie keyword. Check: is the bid too low? Match type too restrictive? Keyword paused at ad group level? | Dead weight cluttering the account |**Accounts WITHOUT conversion data (no conversion tracking or <10 total conversions):**@@ -315,21 +316,23 @@ Evaluate every keyword against the account's average CPA. If the account has no|-----------|--------|| CTR > 5% and CPC < account avg | Likely high-intent — prioritize for conversion tracking setup || CTR < 1% after 500+ impressions | Poor relevance — pause or rewrite ad copy |-| Spend > $500 total with no conversion tracking | Flag as critical: "You're spending $X with no way to measure results. Set up conversion tracking before any optimization." |+| Significant spend with no conversion tracking | Flag as critical: "You're spending $X with no way to measure results. Set up conversion tracking before any optimization." |### Search TermsAnalyze every search term report with these rules. Cross-reference `references/search-term-analysis-guide.md` for the full relevance scoring methodology.+**Tool constraint:** `addNegativeKeyword` only supports **phrase match at campaign level**. No exact match or account-level negatives are available via the tool. Keep this in mind when applying the rules below.+| Condition | Action | Match Type ||-----------|--------|------------|| 3+ conversions, not already a keyword | Add as keyword | Phrase match initially — let it prove itself before going broad || 1-2 conversions, relevant to business | Flag for review — add if CPA is acceptable | Exact match to control spend |-| 0 conversions, 10+ clicks | Add as negative | Phrase match at campaign level |+| 0 conversions, 10+ clicks | Add as negative | Phrase match at campaign level (only option via `addNegativeKeyword`) || 0 conversions, 5-9 clicks | Flag for review — check: is it relevant? Is the landing page right? | May need more data OR a landing page fix, not a negative || 0 conversions, <5 clicks | Too early — skip unless clearly irrelevant | — |-| Clearly irrelevant (competitor name, wrong service, wrong location) | Add as negative immediately regardless of click count | Exact match for competitor names, phrase match for wrong services |-| Contains "free", "DIY", "jobs", "salary" (non-commercial intent) | Add as negative unless the business serves that intent | Phrase match at account level (shared negative list) |+| Clearly irrelevant (competitor name, wrong service, wrong location) | Add as negative immediately regardless of click count | Phrase match at campaign level (covers competitor names sufficiently since phrase match blocks any query containing the term) |+| Contains "free", "DIY", "jobs", "salary" (non-commercial intent) | Add as negative unless the business serves that intent | Phrase match at campaign level — add to each relevant campaign individually || Brand misspelling or variation | Add as keyword if not already covered | Exact match |### Impression Share@@ -345,7 +348,7 @@ Impression share tells you WHY you're not showing for searches. The combination| **Budget-Lost IS > 40%** | Severe budget constraint — must address before any other optimization. Either double the budget or cut keyword count by 50%+ | Priority: fix rank issues first to get more value from existing budget, then increase budget | Account is fundamentally misaligned — targeting too many expensive keywords with too little budget and too low quality. Restructure: pick 10-20 best keywords, pause everything else, fix QS, then expand |**Campaign-level impression share rules:**-- Search IS < 50% on a campaign spending >$1,000/month → this campaign is underserving demand. Investigate why before increasing budget+- Search IS < 50% on a campaign consuming >10% of account budget → this campaign is underserving demand. Investigate why before increasing budget- Abs Top IS < 10% on brand campaigns → competitors are outbidding on your brand. Increase brand campaign bids or improve brand ad QS- Top IS dropped >15 points month-over-month → new competitor or QS degradation. Check auction insights if available@@ -355,9 +358,9 @@ CTR varies dramatically by industry, match type, and ad position. Always compare| Condition | Diagnosis | Action ||-----------|-----------|--------|-| Search CTR < 2% (any industry) | Ad copy relevance problem — the ad doesn't match what the searcher expects | Rewrite headlines to include the keyword or closest synonym. Check if ad group is too broad (mixed intent keywords) |+| Search CTR < 2% (varies by industry — check `references/industry-benchmarks.md`) | Ad copy relevance problem — the ad doesn't match what the searcher expects | Rewrite headlines to include the keyword or closest synonym. Check if ad group is too broad (mixed intent keywords) || Search CTR 2-4% | Acceptable for most industries. Check industry benchmark — some industries (legal, B2B SaaS) index higher | Compare to `references/industry-benchmarks.md`. If below industry avg, test new ad copy |-| Search CTR > 5% but conversion rate < 2% | Ad attracts clicks but landing page doesn't deliver on the ad's promise | Audit landing page: does the headline match the ad? Is the CTA clear? Is the page mobile-friendly? Offer `/ads-landing` |+| Search CTR > 5% but conversion rate below industry average | Ad attracts clicks but landing page doesn't deliver on the ad's promise | Audit landing page: does the headline match the ad? Is the CTA clear? Is the page mobile-friendly? Offer `/ads-landing` || Search CTR > 8% | Excellent — but verify this isn't inflated by brand terms mixing with non-brand in the same campaign | Segment brand vs non-brand. If non-brand CTR is also >8%, this is genuinely strong copy || Display CTR < 0.5% | Normal for display. Only flag if display is eating significant budget with no conversions | Consider pausing display network in campaign settings if it's not converting || CTR declining month-over-month on stable keywords | Ad fatigue or new competitor in auction | Test new ad variants with `/ads-copy`. Check auction insights for new entrants |@@ -366,12 +369,12 @@ CTR varies dramatically by industry, match type, and ad position. Always compare| Condition | Diagnosis | Action ||-----------|-----------|--------|-| Daily budget < $10 with 20+ active keywords | Budget spread too thin — each keyword gets pennies | Reduce to 5-10 highest-performing keywords OR increase budget to give each keyword $0.50-1.00/day minimum |-| Daily budget < $10 with <10 keywords | Acceptable for testing or very low-CPC niches | Monitor — ensure at least 10-15 clicks/day for meaningful data |+| Daily budget < 10x average CPC with 20+ active keywords | Budget spread too thin — each keyword gets pennies | Reduce to 5-10 highest-performing keywords OR increase budget to give each keyword at least 1 click/day |+| Daily budget < 10x average CPC with <10 keywords | Acceptable for testing or very low-CPC niches | Monitor — ensure at least 10-15 clicks/day for meaningful data || One campaign consuming >60% of budget with <40% of conversions | Budget misallocation — money flowing to the wrong campaign | Shift 20-30% of that campaign's budget to the higher-converting campaign. If no other campaign converts better, the problem is the campaign itself, not the budget split || Campaign with conversions hitting budget limit daily (budget-lost IS > 30%) | Proven campaign being starved | Increase budget 25-50% (within server guardrail). This is the lowest-risk budget increase |-| Campaign with 0 conversions after $500+ total spend | Not a budget problem — it's a targeting or conversion tracking problem | Do NOT increase budget. Audit keywords, search terms, landing pages, and conversion tracking first |-| Account spending <$50/day total across all campaigns | Low-data environment — statistical significance takes weeks | Consolidate into fewer campaigns/ad groups. Avoid A/B tests until daily volume supports them (min 30 clicks/day per variant) |+| Campaign with 0 conversions after spending >5x account CPA | Not a budget problem — it's a targeting or conversion tracking problem | Do NOT increase budget. Audit keywords, search terms, landing pages, and conversion tracking first |+| Account generating <30 clicks/day total across all campaigns | Low-data environment — statistical significance takes weeks | Consolidate into fewer campaigns/ad groups. Avoid A/B tests until daily volume supports them (min 30 clicks/day per variant) |## Wasted Spend Calculation@@ -412,11 +415,11 @@ Wasted Spend Breakdown (Last 30 Days):### "How are my ads doing?" — Performance Summary-**Step 1: Pull data (parallel)**+**Step 1: Pull data (parallel — 4 calls total)**- `getAccountInfo` — business name, currency-- `listCampaigns` — all campaigns with spend, clicks, conversions-- `getImpressionShare` — traffic coverage-- `getCampaignPerformance` — daily trends (last 30 days)+- `listCampaigns(limit: 100)` — all campaigns with spend, clicks, conversions+- `runGaqlQuery` — impression share for all campaigns (see `../shared/gaql-cookbook.md` "Impression share" pattern)+- `runGaqlQuery` — daily performance (use LAST_7_DAYS for 2+ campaigns to stay under 50-row limit, see `../shared/gaql-cookbook.md`)**Step 2: Analyze**- Calculate account-level CPA, CTR, conversion rate@@ -429,12 +432,18 @@ Wasted Spend Breakdown (Last 30 Days):### "Find wasted spend" — Waste Audit-**Step 1: Pull data (parallel)**-- `listCampaigns` → identify top 3-5 campaigns by spend-- `getKeywords` for each top campaign → all keywords with spend, conversions, QS-- `getSearchTermReport` for each top campaign → actual queries-- `getCampaignSettings` for each → check if Display Network is enabled-- `getNegativeKeywords` for each → current negative coverage+**Step 1: Pull data (2 phases, see `../shared/gaql-cookbook.md`)**++*Phase 1 (parallel — 4 calls):*+- `listCampaigns(limit: 100)` → all campaigns + identify top 3 by spend+- `runGaqlQuery` → "Zero-conversion high-spend keywords" pattern (directly surfaces waste)+- `runGaqlQuery` → "Search terms" pattern (ordered by spend, for irrelevant term detection)+- `runGaqlQuery` → "Negative keywords" pattern (current coverage)++*Phase 2 (parallel, depends on Phase 1):*+- `getCampaignSettings` for top 3 campaigns → check if Display Network is enabled (major waste source)++If the zero-conversion query returns 50 rows (hit the limit), there's significant waste — supplement with `getKeywords` for the top 2-3 campaigns by spend to get the full picture.**Step 2: Analyze**- Apply the Wasted Spend Calculation above@@ -454,17 +463,22 @@ For each waste source, show:### "Optimize bids" — Bid Optimization-**Step 1: Pull data (parallel)**-- `getKeywords` → all keywords with CPA, CPC, conversions, QS-- `getImpressionShare` → identify where bid increases would capture more traffic-- `getCampaignSettings` → confirm bid strategy (manual/enhanced CPC — bid changes only work with these)+**Step 1: Pull data (2 phases, see `../shared/gaql-cookbook.md`)**++*Phase 1 (parallel):*+- `listCampaigns(limit: 100)` → size the account+- GAQL "Keywords with QS" query → all keywords with CPA, CPC, conversions, QS across campaigns+- GAQL "Impression share" query → where bid increases would capture more traffic++*Phase 2 (depends on Phase 1):*+- `getCampaignSettings` for target campaigns → confirm bid strategy (manual/enhanced CPC only)**Step 2: Analyze using keyword performance heuristics**- Segment keywords into tiers:- **Scale** (CPA < 50% avg): increase bid 15-25%- **Maintain** (CPA 50-100% avg): no change- **Reduce** (CPA 100-150% avg): decrease bid 10-15%, add negatives- - **Pause** (CPA > 200% avg or $200+ spend with 0 conversions): pause+ - **Pause** (CPA > 200% avg or spend >2x account CPA with 0 conversions): pause- Cross-reference with impression share: only increase bids on keywords where rank-lost IS > 20% (there's traffic to capture)- Check bid strategy compatibility: if using Target CPA or Maximize Conversions, manual bid changes are blocked — recommend bid strategy adjustment instead (see `references/bid-strategy-decision-tree.md`)@@ -478,11 +492,16 @@ For each waste source, show:### "Scale winning keywords" — Growth Optimization-**Step 1: Pull data (parallel)**-- `getKeywords` → find keywords with: conversions > 2, CPA < account avg, QS > 6-- `getSearchTermReport` → find converting search terms not yet added as keywords-- `getImpressionShare` → check how much more traffic is available-- `getCampaignSettings` → check budget headroom+**Step 1: Pull data (2 phases, see `../shared/gaql-cookbook.md`)**++*Phase 1 (parallel):*+- `listCampaigns(limit: 100)` → size the account + identify top campaigns+- GAQL "Keywords with QS" query → find keywords with conversions > 2, CPA < avg, QS > 6+- GAQL "Converting search terms" query → search terms with conversions (cross-reference against keyword list to find gaps)+- GAQL "Impression share" query → how much more traffic is available++*Phase 2 (depends on Phase 1):*+- `getCampaignSettings` for target campaigns → check budget headroom**Step 2: Identify scaling opportunities**- **Bid increases**: Keywords with CPA < 50% avg AND rank-lost IS > 20% — room to grow@@ -497,11 +516,12 @@ For each action, estimate the impact:### "Fix quality scores" — QS Diagnostic-**Step 1: Pull data (parallel)**-- `getKeywords` → all keywords with QS and QS subcomponents-- `listAdGroups` → ad group structure (keyword count per group)-- `listAds` → ad copy per ad group-- `getSearchTermReport` → search term relevance to ad groups+**Step 1: Pull data (parallel, see `../shared/gaql-cookbook.md`)**+- `listCampaigns(limit: 100)` → size the account+- GAQL "Keywords with QS" query → all keywords with QS across campaigns+- GAQL "Ad groups" query → ad group structure+- GAQL "Ad copy" query → RSA headlines/descriptions per ad group+- GAQL "Search terms" query → search term relevance to ad groups**Step 2: Diagnose using `references/quality-score-framework.md`**- Group keywords by QS: count in each 1-4, 5-6, 7-8, 9-10 bucket@@ -512,16 +532,20 @@ For each action, estimate the impact:- Check ad group sizes: any ad group with >25 keywords likely has QS problems from mixed intent**Step 3: Present action plan prioritized by spend**-Fix high-spend, low-QS keywords first — they waste the most money. A QS improvement from 4 to 6 typically reduces CPC by 15-25%.+Fix high-spend, low-QS keywords first — they waste the most money. A QS improvement from 4 to 6 can approximately reduce CPC by 15-25% (varies by auction dynamics and competition).### "Restructure campaigns" — Account Restructure-**Step 1: Pull data (parallel)**-- `listCampaigns` → all campaigns-- `listAdGroups` for each → ad group count and themes-- `getKeywords` for top campaigns → keyword count and themes per ad group-- `getCampaignSettings` → current targeting and bid strategies-- `getNegativeKeywords` → current negative coverage+**Step 1: Pull data (2 phases, see `../shared/gaql-cookbook.md`)**++*Phase 1 (parallel):*+- `listCampaigns(limit: 100)` → all campaigns + size the account+- GAQL "Ad groups" query → ad group structure across all campaigns+- GAQL "Keywords with QS" query → keyword themes per ad group+- GAQL "Negative keywords" query → current coverage (note: truncates at 50 rows, supplement with `getNegativeKeywords` per campaign for full picture)++*Phase 2 (depends on Phase 1):*+- `getCampaignSettings` for top campaigns → targeting and bid strategies**Step 2: Diagnose structural issues using `references/campaign-structure-guide.md`**Common problems:@@ -593,6 +617,19 @@ Use this structure for every performance summary. Consistent formatting helps us- [Specific positive finding with numbers]```+### Freshness Notes for High-Volatility Recommendations++When making recommendations in these areas, prefix the recommendation with a freshness disclaimer sourced from `../shared/policy-registry.json`:++- Bid strategy recommendations (`bid-strategy-behavior`)+- Match type behavior advice (`match-type-behavior`)+- Experiment/testing guidance (`experiment-testing`)+- PMax configuration advice (`pmax-configuration`)++Format: _"Based on Google Ads behavior as of [last_verified date from registry]. Verify current behavior if this recommendation is critical to your strategy."_++Only add this for the high-volatility areas listed above. Do not add freshness notes for stable knowledge like QS components, impression share metrics, or CTR benchmarks.+**Rules for the report:**- Every issue must have a dollar amount or conversion count attached- Every action must reference a specific campaign, keyword, or ad group by name@@ -625,7 +662,7 @@ Present the terms with their conversion data and let the user approve before add### Landing Page Misalignment-If CTR is strong (>4%) but conversion rate is below 2% on multiple ad groups:+If CTR is above industry benchmark but conversion rate is below industry average on multiple ad groups:> "Your ads are getting clicks but conversions are low — this usually means the landing page doesn't match what the ad promises. Run `/ads-landing` to audit keyword-to-landing-page alignment."diff --git a/google-ads/ads/references/bid-strategy-decision-tree.md b/google-ads/ads/references/bid-strategy-decision-tree.mdindex a9a37e70..e78abbfe 100644--- a/google-ads/ads/references/bid-strategy-decision-tree.md+++ b/google-ads/ads/references/bid-strategy-decision-tree.md@@ -1,6 +1,10 @@# Bid Strategy Decision Tree-Systematic framework for selecting, migrating between, and troubleshooting Google Ads bidding strategies. Every recommendation includes data thresholds and concrete transition criteria.+Systematic framework for selecting, migrating between, and troubleshooting Google Ads bidding strategies. Every recommendation includes transition criteria based on campaign learning status and performance signals.++> **Why automated bidding is strongly preferred (2026):** Google's bidding engine is powered by an LLM-based auction system that evaluates hundreds of real-time signals per auction — device type, precise location, time of day, audience segments, query intent, competitive density, browser, OS, and more. No human bidder can access or process these signals. Manual bidding means forfeiting this advantage entirely, which is why Google's automated strategies consistently outperform manual approaches across virtually all verticals and account sizes. The recommendations below reflect this reality.++> **⚠️ eCPC Deprecation Notice:** Google deprecated Enhanced CPC (eCPC) for new Search campaigns in 2024. Existing campaigns using eCPC may still run, but Google no longer recommends it, and it cannot be selected for new campaigns. All references to eCPC below are for legacy migration purposes only. Do not adopt eCPC for any new campaign.---@@ -11,25 +15,22 @@ START: What is your optimization goal?|+-- Goal: Maximize conversions within a budget| |-| +-- Conversion volume >= 50/month AND stable CPA (< 20% variance)+| +-- Campaign has exited Learning status AND stable CPA (< 20% variance)| | --> Target CPA (tCPA)| |-| +-- Conversion volume 30-49/month-| | --> Maximize Conversions (budget-capped) — transition to tCPA once 50+/mo+| +-- Campaign still in Learning or recently launched+| | --> Maximize Conversions (budget-capped) — transition to tCPA once Learning exits| |-| +-- Conversion volume 15-29/month-| | --> Enhanced CPC (eCPC) — build conversion history-| |-| +-- Conversion volume < 15/month-| --> Manual CPC — gather data before automating+| +-- Very low volume / new campaign with no conversion history+| --> Maximize Conversions (budget-capped) — let the algorithm gather data|+-- Goal: Maximize conversion value / ROAS| |-| +-- Conversion value tracking in place AND 50+ conv/month+| +-- Conversion value tracking in place AND campaign exited Learning with sufficient volume| | --> Target ROAS (tROAS)| |-| +-- Conversion value tracking in place AND 30-49 conv/month-| | --> Maximize Conversion Value — transition to tROAS once 50+/mo+| +-- Conversion value tracking in place AND still building volume+| | --> Maximize Conversion Value — transition to tROAS once Learning exits| || +-- No conversion value tracking| --> Set up value tracking first, use tCPA in the meantime@@ -37,57 +38,59 @@ START: What is your optimization goal?+-- Goal: Drive traffic / brand awareness| || +-- Brand campaign (own brand terms)-| | --> Manual CPC or Target Impression Share (target: 95%+ top of page)+| | --> Target Impression Share (target: 95%+ top of page)| || +-- New market / awareness play| | --> Maximize Clicks (with max CPC cap)| || +-- Competitor conquesting-| --> Manual CPC (tight control on spend) or Target Impression Share+| --> Target Impression Share (with max CPC cap) or Maximize Conversions (budget-capped)|+-- Goal: Testing / new campaign launch|- +-- Existing conversion data from other campaigns- | --> Maximize Conversions for 2-4 weeks, then transition+ +-- Existing conversion data from other campaigns in the account+ | --> Maximize Conversions for 2-4 weeks, then transition to tCPA+ | (or tCPA directly if account has strong conversion history)|+-- No conversion history at all- --> Manual CPC for first 30 days to establish baselines+ --> Maximize Conversions (budget-capped) for first 30 days to establish baselines```---## Strategy-by-Strategy Reference-### Manual CPC+### Maximize Conversions| Attribute | Detail ||-----------|--------|-| **When to use** | New campaigns, < 15 conversions/month, testing keyword viability, competitor campaigns needing tight spend control |-| **Prerequisites** | None — works with zero historical data |-| **Typical use case** | Campaign launch, niche B2B with low volume, brand defense |-| **Pros** | Full control over every keyword bid; no learning period; predictable spend |-| **Cons** | Time-intensive to manage; cannot react to real-time auction signals; misses user-level optimization |-| **Key setting** | Set bids at keyword level; use ad schedule and device bid adjustments |-| **Exit criteria** | 15+ conversions/month accumulated over 30 days --> move to eCPC |+| **When to use** | New campaigns, budget-constrained campaigns, campaigns with some conversion history; when the goal is "spend this budget, get as many conversions as possible" |+| **Prerequisites** | Conversion tracking active; defined daily budget you are willing to fully spend |+| **Typical use case** | Campaign launch with $50-200/day budget; seasonal pushes; clearing remaining budget at end of month; replacing Manual CPC for new campaigns |+| **Pros** | Simple setup; spends full budget; good for learning what CPA is achievable; accesses Google's full real-time auction signals from day one |+| **Cons** | No CPA control — will spend entire budget regardless of CPA; can overspend early in the day; may pursue low-quality conversions |+| **Key setting** | Set daily budget carefully — the system WILL spend it all. Optionally set a max CPA bid limit |+| **Learning period** | 1-2 weeks. Do not judge results until 2 full weeks of data |+| **Exit criteria** | Campaign exits Learning status and CPA stabilizes --> move to tCPA |-### Enhanced CPC (eCPC)+### Maximize Clicks| Attribute | Detail ||-----------|--------|-| **When to use** | 15-29 conversions/month; transitioning from Manual CPC; want some automation with guardrails |-| **Prerequisites** | Conversion tracking active; 15+ conversions in last 30 days |-| **Typical use case** | Mid-volume lead gen, local services, building toward full automation |-| **Pros** | Adjusts bids up/down based on conversion likelihood while respecting your base bid; lower risk than full automation |-| **Cons** | Limited optimization ceiling; Google can raise bids beyond your manual bid; less effective than tCPA at scale |-| **Key setting** | Set a manual base bid; Google adjusts up or down at auction time |-| **Exit criteria** | 30+ conversions/month sustained for 60 days --> move to tCPA |+| **When to use** | Brand awareness campaigns; new market entry; traffic-focused goals; when you have no conversion tracking yet |+| **Prerequisites** | None — works with zero conversion data |+| **Typical use case** | Brand awareness, content promotion, filling top of funnel, market research |+| **Pros** | Maximizes traffic within budget; no conversion tracking needed |+| **Cons** | No conversion optimization; attracts clicks regardless of quality; can burn budget on low-intent queries |+| **Key setting** | ALWAYS set a max CPC bid cap (start at 50% of your target CPC); without it, Google may pay $15+ per click |+| **Exit criteria** | Once conversion tracking is active and sufficient conversions accumulated --> switch to Maximize Conversions (budget-capped) |### Target CPA (tCPA)| Attribute | Detail ||-----------|--------|-| **When to use** | 30+ conversions/month (Google recommendation: 50+ for best results); stable CPA over 60 days; known target CPA |-| **Prerequisites** | Conversion tracking active; 30+ conversions in last 30 days; CPA variance < 20% week-over-week |+| **When to use** | Campaign has exited Learning status; stable CPA over recent weeks; known target CPA |+| **Prerequisites** | Conversion tracking active; campaign exited Learning status; CPA variance < 20% week-over-week || **Typical use case** | Mature lead gen, SaaS trials, service bookings, e-commerce with uniform AOV || **Pros** | Fully automated bidding optimized for conversions at your target cost; learns from auction signals you cannot access || **Cons** | Needs learning period (1-2 weeks); aggressive targets = zero spend; requires sufficient conversion volume |@@ -95,7 +98,7 @@ START: What is your optimization goal?| **Common pitfalls** | Setting tCPA 30%+ below current average --> campaign stops spending. Setting tCPA too high --> overspending on low-quality conversions |**tCPA Setup Checklist:**-- [ ] Minimum 30 conversions in last 30 days+- [ ] Campaign has exited Learning status (do NOT transition while still in Learning)- [ ] Set initial tCPA = current average CPA (do NOT set aspirational target)- [ ] Reduce tCPA by no more than 10-15% per 2-week period- [ ] Do not make changes during the 1-2 week learning period@@ -105,38 +108,14 @@ START: What is your optimization goal?| Attribute | Detail ||-----------|--------|-| **When to use** | E-commerce with variable order values; 50+ conversions/month with value tracking; known ROAS target |-| **Prerequisites** | Conversion value tracking (revenue per conversion); 50+ conversions in last 30 days; ROAS variance < 25% week-over-week |+| **When to use** | E-commerce with variable order values; sufficient conversion volume with value tracking; known ROAS target |+| **Prerequisites** | Conversion value tracking (revenue per conversion); campaign exited Learning status with stable ROAS; ROAS variance < 25% week-over-week || **Typical use case** | E-commerce, multi-product retailers, travel bookings, real estate leads with scored values || **Pros** | Optimizes for revenue/profit not just conversion count; smart for variable AOV businesses || **Cons** | Needs more data than tCPA; sensitive to conversion value accuracy; can ignore low-value but important conversions || **Key setting** | Set target ROAS at or slightly below (10%) your current average ROAS to start || **When tROAS beats tCPA** | When conversion values vary 3x+ between transactions (e.g., $50 order vs. $500 order); tCPA would treat these equally, tROAS will prioritize the $500 order |-### Maximize Conversions--| Attribute | Detail |-|-----------|--------|-| **When to use** | Budget-constrained campaigns; new campaigns with some conversion history; when the goal is "spend this budget, get as many conversions as possible" |-| **Prerequisites** | Conversion tracking active; defined daily budget you are willing to fully spend |-| **Typical use case** | Campaign launch with $50-200/day budget; seasonal pushes; clearing remaining budget at end of month |-| **Pros** | Simple setup; spends full budget; good for learning what CPA is achievable |-| **Cons** | No CPA control — will spend entire budget regardless of CPA; can overspend early in the day; may pursue low-quality conversions |-| **Key setting** | Set daily budget carefully — the system WILL spend it all. Optionally set a max CPA bid limit |-| **Learning period** | 1-2 weeks. Do not judge results until 2 full weeks of data |--### Maximize Clicks--| Attribute | Detail |-|-----------|--------|-| **When to use** | Brand awareness campaigns; new market entry; traffic-focused goals; when you have no conversion tracking yet |-| **Prerequisites** | None — works with zero conversion data |-| **Typical use case** | Brand awareness, content promotion, filling top of funnel, market research |-| **Pros** | Maximizes traffic within budget; no conversion tracking needed |-| **Cons** | No conversion optimization; attracts clicks regardless of quality; can burn budget on low-intent queries |-| **Key setting** | ALWAYS set a max CPC bid cap (start at 50% of your target CPC); without it, Google may pay $15+ per click |-| **Exit criteria** | Once conversion tracking is active and 15+ conversions accumulated --> switch to Manual CPC or eCPC |-### Target Impression Share| Attribute | Detail |@@ -149,25 +128,53 @@ START: What is your optimization goal?| **Cons** | Expensive if targeting high IS on competitive terms; no conversion optimization || **Key setting** | Set max CPC cap to prevent runaway costs; target IS: 95% for brand, 50-70% for competitive |+### Enhanced CPC (eCPC) — DEPRECATED++> **⚠️ Legacy strategy.** eCPC was deprecated for new Search campaigns in 2024. It remains available only on existing campaigns that already use it. Do not adopt eCPC for new campaigns. Migrate existing eCPC campaigns to Maximize Conversions or Target CPA.++| Attribute | Detail |+|-----------|--------|+| **When to use** | Legacy only — existing campaigns that have not yet migrated |+| **Prerequisites** | N/A — do not adopt for new campaigns |+| **Typical use case** | Migration candidate: move to Maximize Conversions (budget-capped) or tCPA |+| **Pros** | Slightly more automation than Manual CPC while respecting base bid |+| **Cons** | Deprecated; limited optimization ceiling; Google can raise bids beyond your manual bid; far less effective than tCPA or Maximize Conversions |+| **Key setting** | N/A — focus on migration planning |+| **Exit criteria** | Migrate to Maximize Conversions (budget-capped) or directly to tCPA if conversion volume supports it |++### Manual CPC — NOT RECOMMENDED++> **⚠️ Not recommended for any new campaign.** Manual CPC forfeits access to Google's real-time LLM-based auction signals. Use Maximize Clicks (with CPC cap) for traffic goals or Maximize Conversions (budget-capped) for conversion goals instead.++| Attribute | Detail |+|-----------|--------|+| **When to use** | Extremely rare edge cases only — e.g., regulatory environments requiring fixed bids |+| **Prerequisites** | None — works with zero historical data |+| **Typical use case** | Not recommended for standard use |+| **Pros** | Full control over every keyword bid; no learning period; predictable spend |+| **Cons** | Cannot react to real-time auction signals; misses user-level optimization; time-intensive to manage; consistently underperforms automated strategies |+| **Key setting** | Set bids at keyword level; use ad schedule and device bid adjustments |+| **Exit criteria** | Migrate immediately to Maximize Conversions (budget-capped) or Maximize Clicks (with CPC cap) |+---## Migration Paths### Standard Migration Ladder-| Step | From | To | Minimum Data Threshold | Timeline |-|------|------|----|----------------------|----------|-| 1 | Manual CPC | Enhanced CPC | 15+ conversions in 30 days | After 30-60 days of data |-| 2 | Enhanced CPC | Target CPA | 30+ conversions in 30 days, CPA variance < 20% | After 60 days on eCPC |-| 3 | Target CPA | Target ROAS | 50+ conversions in 30 days + value tracking active | When conversion values vary 3x+ |-| Alt | Manual CPC | Maximize Conversions | Conversion tracking active, fixed budget | When budget is set and you want to skip eCPC |-| Alt | Maximize Conversions | Target CPA | 30+ conversions in 30 days from Max Conv data | After 2-4 weeks of Max Conv data |+| Step | From | To | Transition Criteria | Timeline |+|------|------|----|---------------------|----------|+| 1 | New campaign | Maximize Conversions (budget-capped) | Conversion tracking active, budget set | Day 1 |+| 2 | Maximize Conversions | Target CPA | Campaign exits Learning status, CPA stabilizes (< 20% variance) | After 2-4+ weeks |+| 3 | Target CPA | Target ROAS | Sufficient conversion volume + value tracking active, conversion values vary 3x+ | When value optimization matters |+| Alt | New campaign (account has conversion history) | Target CPA | Account has strong conversion data from other campaigns | Day 1 (direct) |+| Legacy | eCPC (deprecated) | Maximize Conversions or tCPA | Any time — eCPC should be migrated off | Immediately |### Migration Safety Rules| Rule | Why ||------|-----|-| Never skip more than one step in the ladder (e.g., Manual --> tROAS) | Insufficient data leads to erratic bidding and wasted spend |+| Allow direct tCPA adoption if the account has conversion history from other campaigns | Google's algorithm can leverage cross-campaign data; rigid step-skipping rules are outdated || Never change bid strategy AND campaign settings simultaneously | Impossible to diagnose what caused performance change || Allow 2 full weeks after any strategy change before evaluating | Learning period needs time; early data is unreliable || Never change bid strategy during seasonal peaks (Black Friday, etc.) | Algorithm relearns during your highest-value period |@@ -187,8 +194,8 @@ START: What is your optimization goal?| Max Conversions | No budget cap awareness | Entire monthly budget spent in first week | Set daily budget = monthly budget / 30; monitor daily spend || Max Conversions | Low-quality conversions | High conversion count but poor lead quality | Add a max CPA bid limit; consider switching to tCPA || Max Clicks | No max CPC cap set | Single clicks costing $10-20+ | Always set max CPC cap at 50-75% of target CPC |-| eCPC | Ad groups with mixed intent keywords | Google overbids on irrelevant close variants | Tighten ad groups; add negative keywords |-| Manual CPC | Bids not updated in 30+ days | Missing auction changes, losing impression share | Review bids weekly or switch to automated strategy |+| eCPC (legacy) | Still running on campaigns that should migrate | Underperformance vs. automated strategies | Migrate to Maximize Conversions or tCPA |+| Manual CPC (legacy) | Still in use, missing real-time signals | Losing impression share, CPA higher than automated competitors | Migrate to Maximize Conversions (budget-capped) or Maximize Clicks (with CPC cap) || Target IS | No max CPC cap on competitive terms | CPC spirals to 3-5x normal | Set max CPC cap; accept lower IS on expensive terms |---@@ -198,7 +205,7 @@ START: What is your optimization goal?| Attribute | Standard (Campaign-level) | Portfolio (Shared across campaigns) ||-----------|--------------------------|-------------------------------------|| **Scope** | One campaign | Multiple campaigns share one strategy |-| **When to use** | Campaign has 50+ conversions/month alone | Individual campaigns have < 30 conversions/month but combined total is 50+ |+| **When to use** | Campaign has sufficient conversion volume on its own to exit Learning | Individual campaigns lack volume but combined total is enough to exit Learning || **Data pooling** | Campaign data only | Pools conversion data across all campaigns in the portfolio || **Budget** | Per-campaign budget | Per-campaign budgets still; strategy optimizes across them || **Best for** | High-volume campaigns | Low-volume campaigns that share a common goal (same CPA target, same ROAS target) |@@ -208,16 +215,16 @@ START: What is your optimization goal?### Portfolio Strategy Decision Rule```-IF single campaign conversions >= 50/month+IF single campaign has sufficient volume to exit Learning on its own--> Use standard (campaign-level) strategy-IF single campaign conversions < 30/month+IF single campaign cannot exit Learning on its ownAND multiple campaigns share the same conversion goal- AND combined conversions across those campaigns >= 50/month+ AND combined volume across those campaigns is sufficient to exit Learning--> Use portfolio strategy-IF total conversions across all candidates < 30/month- --> Use Manual CPC or eCPC until volume builds+IF total volume across all candidates is too low to exit Learning even when pooled+ --> Use Maximize Conversions (budget-capped) across campaigns until volume builds```---@@ -228,13 +235,13 @@ IF total conversions across all candidates < 30/month| Strategy | Primary Metric | Secondary Metric | Evaluation Window ||----------|---------------|-----------------|-------------------|-| Manual CPC | CPA or ROAS | Impression share, avg. position | Weekly |-| eCPC | CPA vs. pre-eCPC baseline | Conversion volume change | Bi-weekly |-| tCPA | Actual CPA vs. target CPA | Conversion volume, search IS | 2-week rolling average |+| Maximize Conversions | CPA trend | Total conversions, budget utilization | Weekly |+| Maximize Clicks | CPC trend, traffic volume | Click quality (bounce rate, time on site) | Weekly |+| tCPA | Actual CPA vs. target CPA | Conversion volume, search impression share | 2-week rolling average || tROAS | Actual ROAS vs. target ROAS | Revenue, conversion value | 2-week rolling average |-| Max Conversions | CPA trend | Total conversions, budget utilization | Weekly |-| Max Clicks | CPC trend, traffic volume | Click quality (bounce rate, time on site) | Weekly || Target IS | Actual IS vs. target IS | CPC trend, budget utilization | Weekly |+| eCPC (legacy) | CPA vs. pre-eCPC baseline | Conversion volume change | Bi-weekly (migrate off) |+| Manual CPC (legacy) | CPA or ROAS | Top impression rate, impression share | Weekly (migrate off) |### When to Intervenediff --git a/google-ads/ads/references/campaign-structure-guide.md b/google-ads/ads/references/campaign-structure-guide.mdindex 0d56d4ac..b09a8033 100644--- a/google-ads/ads/references/campaign-structure-guide.md+++ b/google-ads/ads/references/campaign-structure-guide.md@@ -10,9 +10,9 @@ Account structure best practices for Google Ads. Covers campaign organization, a| Era | Strategy | Keywords per Ad Group | Rationale | Current Status ||-----|----------|-----------------------|-----------|----------------|-| 2012-2017 | **SKAG** (Single Keyword Ad Groups) | 1 | Maximize ad relevance; exact control over which ad shows for which keyword | **Deprecated** — close variants and smart bidding made this counterproductive |-| 2017-2020 | **STAG** (Single Theme Ad Groups) | 3-7 | Group close variants; reduce account complexity while maintaining relevance | **Transitional** — still viable but often too granular |-| 2020-present | **Themed Ad Groups** | 10-20 | Leverage smart bidding; consolidate conversion data; match Google's AI-driven auction system | **Current best practice** |+| 2012-2017 | **SKAG** (Single Keyword Ad Groups) | 1 | Maximize ad relevance; exact control over which ad shows for which keyword | **Detrimental** — Google's LLM-based bidding and matching systems actively perform worse with single-keyword isolation; theme-based clustering provides the signal density these systems need |+| 2017-2020 | **STAG** (Single Theme Ad Groups) | 3-7 | Group close variants; reduce account complexity while maintaining relevance | **Outdated** — too granular for modern LLM-based bidding; consolidate into themed ad groups of 10-20 keywords |+| 2020-2026+ | **Themed Ad Groups** | 10-20 | Leverage smart bidding; consolidate conversion data; match Google's AI-driven auction system | **Current best practice** |### Why SKAGs No Longer Work@@ -20,6 +20,7 @@ Account structure best practices for Google Ads. Covers campaign organization, a|---------|------------|| Close variants eliminated keyword-level control | Google matches exact match keywords to synonyms, paraphrases, and implied-intent queries regardless of ad group structure || Data fragmentation kills smart bidding | A SKAG with 5 conversions/month cannot feed tCPA effectively; the same 50 conversions consolidated into one themed ad group can |+| LLM-based systems need signal density | Google's bidding and matching now use LLM models that require diverse, theme-rich keyword clusters to build accurate intent representations; single-keyword ad groups starve these models of contextual signal || Account management overhead | 500 SKAGs = 500 ad groups to manage, each needing unique ads; 25 themed groups covers the same keywords with 95% less maintenance || Google rewards consolidation | Google's recommendation engine and smart bidding algorithms perform better with fewer, larger ad groups |@@ -31,7 +32,7 @@ Account structure best practices for Google Ads. Covers campaign organization, a| Theme definition | Keywords should share the same intent AND could be answered by the same landing page || Ads per ad group | 2-3 RSAs (Responsive Search Ads) || Landing page | 1 primary landing page per ad group (all keywords should be relevant to that page) |-| Match types | Mix of exact and phrase match within the same ad group is fine; broad match only with smart bidding |+| Match types | Exact + broad match is the standard pairing; phrase and broad now overlap ~100% in many verticals, so phrase match adds minimal incremental value. Use broad match only with smart bidding |### Theme Grouping Examples@@ -65,7 +66,7 @@ Create a new campaign when you need any of these to differ:| Scenario | Why Splitting Hurts | Better Approach ||----------|-------------------|-----------------|-| Splitting by match type (Exact campaign, Phrase campaign, Broad campaign) | Fragments conversion data; smart bidding needs all data in one place | Mix match types within ad groups |+| Splitting by match type (Exact campaign, Phrase campaign, Broad campaign) | Fragments conversion data; smart bidding needs all data in one place. Phrase and broad now overlap ~100% in many verticals, making three-way splits pointless | Mix match types within ad groups; use exact + broad as the standard pairing || Splitting every service into its own campaign | 20 campaigns with 5 conversions/month each = too thin for smart bidding | Group related services into one campaign with themed ad groups || Splitting brand + non-brand into 10+ campaigns each | Management overhead; inconsistent settings | 1 brand campaign + 2-4 non-brand campaigns organized by service line || Splitting by device (mobile campaign, desktop campaign) | Fragments data; smart bidding adjusts bids by device automatically | Use device bid adjustments within one campaign if needed |@@ -76,11 +77,11 @@ Create a new campaign when you need any of these to differ:| Campaign | Purpose | Bid Strategy | Budget Allocation ||----------|---------|-------------|-------------------|-| **Brand** | Own brand terms | Target IS (95% top of page) or Manual CPC | 5-10% of total budget |+| **Brand** | Own brand terms | Target IS (95% top of page) | 5-10% of total budget || **Non-Brand Core** | Highest-value services/products | tCPA or tROAS | 50-60% of total budget || **Non-Brand Secondary** | Lower-priority or newer services | tCPA or Maximize Conversions | 15-20% of total budget |-| **Competitor** | Competitor brand terms (optional) | Manual CPC (tight control) | 5-10% of total budget |-| **Testing** | New keywords, audiences, ad copy tests | Maximize Conversions or Manual CPC | 10-15% of total budget |+| **Competitor** | Competitor brand terms (optional) | Target IS (with max CPC cap) or Maximize Conversions (budget-capped) | 5-10% of total budget |+| **Testing** | New keywords, audiences, ad copy tests | Maximize Conversions (budget-capped) or Maximize Clicks (with max CPC cap) | 10-15% of total budget || **Remarketing (Display/Search)** | Re-engage past visitors | tCPA | 5-10% of total budget |---@@ -215,7 +216,7 @@ Review these settings for every new campaign. Incorrect defaults waste budget.| **Ad schedule** | Set to business hours if B2B; 24/7 if B2C with online conversion | 24/7 (default) | B2B: Why pay for 2am clicks? B2C: online conversions happen at all hours || **Device targeting** | All devices (use bid adjustments if needed) | All devices (default is fine) | Smart bidding adjusts by device; manual campaigns may need mobile bid adjustment || **Brand exclusions** | Exclude own brand from non-brand campaigns | None (default) | Prevent brand queries from inflating non-brand campaign performance metrics |-| **Dynamic search ads** | Off unless intentionally using DSA | Sometimes on by default | DSA can match to unexpected pages; use only when deliberately set up |+| **Dynamic search ads** | Off unless intentionally using DSA. Note: DSA is being superseded by Performance Max, which provides broader automation with asset-group-level control | Sometimes on by default | DSA can match to unexpected pages; use only when deliberately set up. For new campaigns, prefer Performance Max over DSA |### Location Targeting Detaildiff --git a/google-ads/ads/references/industry-benchmarks.md b/google-ads/ads/references/industry-benchmarks.mdindex c9a4de24..50d2cdef 100644--- a/google-ads/ads/references/industry-benchmarks.md+++ b/google-ads/ads/references/industry-benchmarks.md@@ -1,6 +1,8 @@# Google Ads Industry Performance Benchmarks-Industry-specific benchmarks for Google Ads performance evaluation. Data ranges based on widely-published WordStream, Google, and LOCALiQ industry reports (2024-2025). Use these as diagnostic baselines, not absolute targets.+Industry-specific benchmarks for Google Ads performance evaluation. Data ranges based on widely-published WordStream, Google, and LOCALiQ industry reports (2024-2025 vintage). Use these as diagnostic baselines, not absolute targets.++> **Inflation adjustment (2026)**: These benchmarks reflect 2024-2025 data. Google Ads CPCs have been rising 10-20% year-over-year due to increased competition, AI-driven bidding adoption, and SERP layout changes (AI Overviews reducing organic clicks, pushing more demand to paid). When comparing your 2026 performance to these tables, adjust CPC and CPA benchmarks upward by 10-20% to account for market inflation.---@@ -12,6 +14,8 @@ These benchmarks represent **averages across all advertisers**, including poorly## Search Network Benchmarks by Industry+> **2026 note**: CPC and CPA figures below are 2024-2025 averages. For 2026 planning, adjust upward by 10-20%. AI Overviews and other SERP layout changes are compressing organic CTR, increasing advertiser demand for paid placements, and driving costs higher across most verticals.+| Industry | Avg CPC | Avg CTR | Avg Conv Rate | Avg CPA | Notes ||----------|---------|---------|---------------|---------|-------|| **Legal** | $6.75 | 4.4% | 7.5% | $90.00 | Highest CPC vertical; personal injury drives the average up |@@ -41,6 +45,8 @@ These benchmarks represent **averages across all advertisers**, including poorly## Display Network Benchmarks by Industry+> **2026 note**: Display CPC figures below are 2024-2025 averages. Display costs have inflated less than Search (5-10% YoY) but still trend upward. Adjust accordingly.+| Industry | Avg CPC (Display) | Avg CTR (Display) | Avg Conv Rate (Display) | Notes ||----------|-------------------|-------------------|------------------------|-------|| **Legal** | $0.85 | 0.60% | 1.8% | Display is rarely primary channel for legal |@@ -77,7 +83,7 @@ These benchmarks represent **averages across all advertisers**, including poorly|-------------|--------------|-----|| **Lost IS (Budget)** > 20% | Budget runs out before end of day | Increase daily budget or reduce keyword scope || **Lost IS (Budget)** 5-20% | Mild budget constraint | Acceptable if intentional; increase budget for high-ROI campaigns |-| **Lost IS (Rank)** > 30% | Ad Rank too low (QS or bid issue) | Improve QS first (free); increase bids second (costs money) |+| **Lost IS (Rank)** > 30% | Ad Rank too low (QS or bid issue) | Improve QS first (free); for Smart Bidding accounts, loosen target CPA/ROAS to allow higher bids; for manual bidding, increase bids (costs money) || **Lost IS (Rank)** 10-30% | Moderate rank issues | Focus QS improvement on highest-volume keywords || **Lost IS (Rank)** < 10% | Healthy competitive position | Monitor; no urgent action needed |@@ -92,7 +98,9 @@ These benchmarks represent **averages across all advertisers**, including poorly| Position 3 (Top) | 2.5-3.5% | 4.0-5.5% | 1.5-2.0% || Position 4+ (Other) | 1.0-2.0% | 2.0-3.0% | 0.5-1.0% |-**Interpretation**: If your ads are in Position 1 but CTR is < 5%, your ad copy is underperforming. If your CTR is 8%+ in Position 1, your ad copy is strong; focus optimization on CPA and conversion rate instead.+**Interpretation**: If your ads are in Absolute Top position but CTR is < 5%, your ad copy is underperforming. If your CTR is 8%+ in Absolute Top, your ad copy is strong; focus optimization on CPA and conversion rate instead.++> **SERP layout changes (2026)**: AI Overviews (formerly SGE) now appear on a significant percentage of informational and commercial queries. When AI Overviews are present, they push ads and organic results further down the page, compressing CTR for all ad positions. CTR benchmarks above may overstate expected performance for queries where AI Overviews dominate. Monitor your Absolute Top impression share and CTR trends together — declining CTR with stable impression share may indicate AI Overview interference rather than ad copy issues.---@@ -184,8 +192,12 @@ Different conversion types have inherently different values. Compare your CPA to| CPA | 10-30% higher than desktop | Baseline | Mobile more expensive per conversion | Despite cheaper clicks, lower conv rate drives CPA up || Bounce Rate | 55-65% | 35-45% | Mobile 50% higher | Mobile pages must load fast and be thumb-friendly |+> **2026 update**: The mobile-desktop conversion rate gap has been narrowing steadily as mobile checkout, autofill, and payment experiences improve (Apple Pay, Google Pay, one-tap checkout). In verticals with strong mobile UX (e-commerce, local services, click-to-call), the gap is now closer to 20-30% rather than the 40-60% shown above. B2B and complex SaaS remain desktop-dominant.+### Device Strategy Decision+> **Smart Bidding note (2026)**: Device bid adjustments are **ignored** by Smart Bidding strategies (tCPA, tROAS, Maximize Conversions, Maximize Conversion Value). Smart Bidding automatically adjusts bids per device based on real-time conversion signals. The recommendations below apply only to Manual CPC or Enhanced CPC campaigns. For Smart Bidding campaigns, optimize the mobile experience (page speed, form simplicity) instead of adjusting bids — the algorithm will find the right bid per device.+| Scenario | Recommended Device Strategy ||----------|---------------------------|| E-commerce with mobile-optimized checkout | Bid equally on all devices; mobile can perform well |diff --git a/google-ads/ads/references/quality-score-framework.md b/google-ads/ads/references/quality-score-framework.mdindex 5aaa9769..2aa3bd80 100644--- a/google-ads/ads/references/quality-score-framework.md+++ b/google-ads/ads/references/quality-score-framework.md@@ -10,17 +10,21 @@ Quality Score (QS) is a 1-10 keyword-level metric composed of three sub-componen| Component | Weight (approx.) | Rating Scale | What Google Evaluates ||-----------|------------------|--------------|----------------------|-| **Expected CTR** | ~39% | Below Average / Average / Above Average | Historical CTR of the keyword normalized for ad position, extensions, and format |-| **Ad Relevance** | ~22% | Below Average / Average / Above Average | How closely your ad copy matches the intent and language of the keyword |-| **Landing Page Experience** | ~39% | Below Average / Average / Above Average | Page load speed, mobile-friendliness, content relevance, navigation clarity, trust signals |+| **Expected CTR** | ~35-40% | Below Average / Average / Above Average | Historical CTR of the keyword normalized for ad position, extensions, and format |+| **Ad Relevance** | ~20-25% | Below Average / Average / Above Average | How closely your ad copy matches the intent and language of the keyword |+| **Landing Page Experience** | ~35-40% | Below Average / Average / Above Average | Page load speed, mobile-friendliness, content relevance, navigation clarity, trust signals |-**Key insight**: Landing Page Experience and Expected CTR together account for ~78% of QS. Ad Relevance is the smallest factor but often the easiest to fix.+**Disclaimer**: Google has never published official QS component weights. The ranges above are estimates derived from third-party regression studies and may shift over time. Use them directionally, not as exact figures.++**Key insight**: Landing Page Experience and Expected CTR together account for the majority (~70-80%) of QS. Ad Relevance is the smallest factor but often the easiest to fix.---## QS Impact on Cost-Per-Click-Quality Score directly modifies your actual CPC through the Ad Rank formula. Below are the approximate CPC modifiers relative to QS 5 (the neutral baseline):+Quality Score directly modifies your actual CPC through the Ad Rank formula. Below are the approximate CPC modifiers relative to QS 5 (the neutral baseline).++> **Caveat**: These CPC impact figures are approximate estimates from third-party studies (notably WordStream and Optmyzr), not Google-published data. Actual CPC modifiers vary by auction, vertical, and account history.| Quality Score | CPC Modifier vs. QS 5 Baseline | Effective CPC at $2.00 Benchmark | Impact on Budget ||--------------|-------------------------------|----------------------------------|-----------------|@@ -37,6 +41,8 @@ Quality Score directly modifies your actual CPC through the Ad Rank formula. Bel**Rule of thumb**: Every 1-point QS increase above 5 saves ~16% on CPC. Every 1-point decrease below 5 costs ~50% more on CPC.+> **Smart Bidding context**: Throughout this playbook, where the advice says "increase bids," accounts using Smart Bidding (tCPA, tROAS, Maximize Conversions, Maximize Conversion Value) should adjust their target CPA or target ROAS instead. Smart Bidding sets bids automatically per auction — manual bid changes are overridden.+---## Diagnostic Decision Tree@@ -81,14 +87,17 @@ START: Keyword Quality ScoreExpected CTR is position-adjusted, meaning Google accounts for your average position. A "Below Average" rating means your ads get fewer clicks than competitors in similar positions.-### Position-Adjusted CTR Benchmarks (Search Network)+### CTR Benchmarks by Impression Location (Search Network)++Average Position was deprecated by Google in September 2019. Use Impression Location segments instead:++| Impression Location | Expected CTR Floor (Below Average threshold) | Good CTR | Excellent CTR |+|--------------------|---------------------------------------------|----------|---------------|+| **Absolute Top** (first position above organic) | < 5.0% | 6.0-9.0% | > 9.0% |+| **Top** (above organic results, positions 2-4) | < 2.5% | 3.5-6.0% | > 6.0% |+| **Other** (bottom of page or secondary placements) | < 1.0% | 1.5-3.0% | > 3.0% |-| Average Position | Expected CTR Floor (Below Average threshold) | Good CTR | Excellent CTR |-|-----------------|---------------------------------------------|----------|---------------|-| 1.0 - 1.5 | < 4.0% | 5.0-8.0% | > 8.0% |-| 1.5 - 2.5 | < 2.5% | 3.5-6.0% | > 6.0% |-| 2.5 - 3.5 | < 1.5% | 2.5-4.5% | > 4.5% |-| 3.5+ | < 1.0% | 1.5-3.0% | > 3.0% |+> Use the "Top vs. Other" and "Abs. Top vs. Top" impression share metrics to understand where your ads appear. These replaced average position as the standard diagnostic.### Fix Actions (Priority Order)@@ -124,7 +133,7 @@ Ad Relevance measures how well your ad copy matches the keyword's intent and lan|----------|---------|-----|| Keyword not in any headline | Google sees weak topical match | Add keyword or close synonym to Headline 1 or 2 || Keyword in headline but different intent | Ad promises something the keyword doesn't seek | Rewrite headline to match the search intent (informational, commercial, transactional) |-| Ad group has 30+ keywords | Impossible for 1 ad to be relevant to all keywords | Split into ad groups of 10-20 tightly themed keywords |+| Ad group has 30+ keywords | Impossible for 1 ad to be relevant to all keywords | Split into tighter ad groups: 5-10 keywords for broad match accounts (broad needs fewer keywords per group since each keyword matches widely), 10-20 for exact/phrase match accounts || Generic ad copy across all ad groups | Same ads everywhere = low relevance for specific terms | Write unique RSAs per ad group with keyword-specific headlines || Using only broad match keywords | Google matches queries far from original keyword | Review search terms; add phrase/exact match for high-value terms |@@ -201,7 +210,7 @@ When you have limited time, prioritize fixes by impact and effort:| P1 | Expected CTR | CTR < position-adjusted floor | Rewrite headlines, add extensions | +1 to +2 QS points || P1 | Landing Page | No HTTPS or broken mobile layout | Fix technical issues | +1 to +2 QS points || P2 | Ad Relevance | Ad group has 30+ keywords | Split into tighter ad groups | +1 to +2 QS points |-| P2 | Expected CTR | Broad match diluting CTR | Add negative keywords from search terms | +1 QS point |+| P2 | Expected CTR | Broad match diluting CTR | Add negative keywords from search terms (note: negatives have reduced precision under Search Themes / PMax — see search term analysis guide) | +1 QS point || P3 | Landing Page | Generic trust signals | Add reviews, certifications, guarantees | +0.5 to +1 QS point || P3 | Expected CTR | Ad extensions not showing | Enable all relevant extension types | +0.5 to +1 QS point |diff --git a/google-ads/ads/references/search-term-analysis-guide.md b/google-ads/ads/references/search-term-analysis-guide.mdindex 894a5798..57f13349 100644--- a/google-ads/ads/references/search-term-analysis-guide.md+++ b/google-ads/ads/references/search-term-analysis-guide.md@@ -8,14 +8,16 @@ Systematic framework for interpreting search term reports, mining keywords, builUnderstanding what Google actually matches to each match type is the foundation of search term analysis.-### Match Type Definitions (Current Behavior, 2024-2025)+### Match Type Definitions (Current Behavior, 2025-2026)| Match Type | Syntax | What It Matches | Example Keyword | Example Matched Queries ||-----------|--------|----------------|-----------------|------------------------|| **Broad Match** | `keyword` | Themes, intent, related concepts; uses AI to determine relevance | `plumber services` | "plumber near me", "fix leaky faucet", "emergency pipe repair", "plumbing contractor reviews" |-| **Phrase Match** | `"keyword"` | Queries that include the meaning of the keyword; word order matters for meaning | `"plumber services"` | "plumber services near me", "affordable plumber services in austin", "residential plumber services" |+| **Phrase Match** | `"keyword"` | Queries that include the meaning of the keyword; word order matters for meaning. In practice, semantic overlap with Broad match is nearly complete in most verticals (2025-2026). | `"plumber services"` | "plumber services near me", "affordable plumber services in austin", "residential plumber services" || **Exact Match** | `[keyword]` | Queries that match the exact meaning including close variants | `[plumber services]` | "plumber services", "plumber service", "plumbing services", "services plumber" |+> **Note (2026):** Phrase and Broad match overlap is nearly 100% in many verticals. Running both on the same keyword is redundant. In most cases, choose one: Broad with Smart Bidding for maximum reach, or Exact for maximum control. Adding Phrase match alongside Broad rarely changes which queries you match.+### Close Variants (All Match Types)Google applies close variants to ALL match types. These include:@@ -99,6 +101,19 @@ FOR EACH search term in the report:---+## Important: Negatives Are Now a Supplementary Layer (2026)++Traditional negative keyword lists remain useful but are no longer the primary lever for query control. Several Google Ads changes have reduced their precision:++- **Brand Exclusions**: Campaign-level brand exclusion lists (launched 2023, expanded 2024-2025) are now the recommended way to block competitor or irrelevant brand traffic. They work at the brand-entity level, not the keyword level, so they catch variations that keyword negatives miss.+- **Account-Level Suitability Settings**: Google's account-level content suitability and brand safety controls filter categories of queries before negative keywords are even evaluated. Check these settings before building extensive negative lists.+- **Search Themes (Performance Max / Broad Match)**: Google's AI-driven query expansion (Search Themes in PMax, and Broad match's intent matching) can serve ads on queries that don't contain any of your keyword text. Negative keywords only block queries containing the negative term literally — they cannot block conceptually related queries that the AI matches to. This means negative keyword lists give **false confidence** in query control when running Broad match or PMax campaigns.+- **Google's LLM-powered bidding** has real-time signal advantages (device, location, time, audience, query context) that static negative lists cannot replicate. Smart Bidding can effectively "negative out" low-value queries by bidding them down to near-zero, often more precisely than manual negatives.++**Bottom line**: Use negatives for clear-cut exclusions (job seekers, DIY, competitors you don't want to conquest). But don't rely on them as your primary quality control mechanism — Brand Exclusions, suitability settings, and Smart Bidding now carry more of that weight.++---+## Negative Keyword Strategy### Negative Match Type Selection@@ -134,7 +149,7 @@ Is only this specific query irrelevant?### Universal Negative Keyword Lists-Build these shared lists and apply across all campaigns in the account.+Build these shared lists and apply across all campaigns in the account. **Caveat (2026):** These lists are a useful starting point for clear-cut exclusions, but they give false confidence when running Broad match or Performance Max campaigns. Google's Search Themes and AI-driven query expansion can serve ads on conceptually related queries that don't literally contain any of these negative terms. Treat these lists as a supplementary layer, not a complete shield.#### General Exclusions (Apply to All Commercial Campaigns)@@ -226,7 +241,7 @@ Categorize search terms by intent to determine campaign-level strategy.| Phrase Match | CPA <= 130% of account average | Review search terms; add negatives for irrelevant phrase matches || Broad Match | CPA <= 160% of account average | Review search terms aggressively; consider tightening to phrase; ensure smart bidding is active |-**Rule**: If broad match CPA is > 200% of exact match CPA for the same keyword, switch to phrase match. Broad match should only be used with smart bidding (tCPA or tROAS) which adjusts bids per query.+**Rule**: If broad match CPA is > 200% of exact match CPA for the same keyword, review search terms and add negatives or Brand Exclusions. Switching to Phrase match is unlikely to help given the near-complete semantic overlap between Phrase and Broad in most verticals (2026). Instead, ensure Smart Bidding (tCPA or tROAS) is active — Google's LLM bidding has real-time signal advantages that adjust bids per query more precisely than match type restrictions. If CPA remains unacceptable after negative refinement and Smart Bidding, consider pausing the broad match variant and running exact match only.---@@ -248,7 +263,7 @@ Critical: Wasted spend > 20% of total spend| Pattern | How to Identify | Fix ||---------|----------------|-----|-| Broad match runaway | High spend, low relevance search terms, mostly from broad match keywords | Add negatives; consider switching to phrase match |+| Broad match runaway | High spend, low relevance search terms, mostly from broad match keywords | Add negatives and Brand Exclusions; ensure Smart Bidding is active; if CPA remains unacceptable, switch to exact match only || Wrong-intent matches | Commercial keyword matching to informational queries | Add informational signal words as negatives || Competitor term leakage | Your ads showing for competitor brand names (unintentionally) | Add competitor names as exact negatives (unless conquesting) || Geographic mismatch | Queries from locations you don't serve | Tighten location targeting; add location negatives |diff --git a/google-ads/shared/policy-registry.json b/google-ads/shared/policy-registry.jsonnew file mode 100644index 00000000..96440bce--- /dev/null+++ b/google-ads/shared/policy-registry.json@@ -0,0 +1,436 @@+{+ "$schema": "policy-registry/v1",+ "description": "Versioned registry of Google Ads policy and behavior assumptions used across toprank ads skills. Each entry tracks a specific assumption, its volatility, verification source, and when it should be re-verified.",+ "last_full_review": "2026-04-04",+ "entries": [+ {+ "id": "auction-ad-rank-equation",+ "category": "bidding",+ "rule": "Ad Rank = f(Bid, Quality Score, Expected Impact of Extensions, Ad Rank Thresholds, Auction Context). This is the core auction equation governing ad placement.",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/1722122",+ "notes": "The components feeding into Ad Rank evolve (e.g., context signals expand), but the fundamental equation structure has been stable for years.",+ "stale_after_days": 365+ },+ {+ "id": "object-hierarchy",+ "category": "campaign-structure",+ "rule": "Google Ads object hierarchy: Account > Campaign > Ad Group > Ad/Keyword. Budget and bid strategy are set at campaign level; match types, bids, and ads at ad group level.",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/6324971",+ "notes": "PMax campaigns flatten this hierarchy (no ad groups in the traditional sense), but the core hierarchy remains for Search campaigns.",+ "stale_after_days": 365+ },+ {+ "id": "metric-definitions-ctr-cpc-cvr",+ "category": "benchmarks",+ "rule": "CTR = Clicks / Impressions. CPC = Cost / Clicks. Conversion Rate = Conversions / Clicks. CPA = Cost / Conversions. ROAS = Conversion Value / Cost.",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/6396841",+ "notes": "Definitions are stable. What counts as a 'conversion' depends on the account's conversion action setup and attribution model.",+ "stale_after_days": 365+ },+ {+ "id": "gaql-syntax-stability",+ "category": "campaign-structure",+ "rule": "GAQL (Google Ads Query Language) uses SQL-like SELECT/FROM/WHERE/ORDER BY/LIMIT syntax against resource types (campaign, ad_group, ad_group_criterion, etc.).",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads API docs",+ "source_url": "https://developers.google.com/google-ads/api/docs/query/overview",+ "notes": "New fields and resources are added with API versions, but the query syntax itself is stable. Field names may change across API versions.",+ "stale_after_days": 365+ },+ {+ "id": "rsa-character-limits",+ "category": "ad-copy",+ "rule": "RSA character limits: Headlines max 30 chars (up to 15 per RSA), Descriptions max 90 chars (up to 4 per RSA). Display paths max 15 chars each. Minimum: 3 headlines, 2 descriptions.",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/7684791",+ "notes": "Character limits have not changed since RSA introduction. Google may expand max headlines/descriptions in the future.",+ "stale_after_days": 365+ },+ {+ "id": "qs-three-components",+ "category": "quality-score",+ "rule": "Quality Score (1-10) is composed of three sub-components: Expected CTR, Ad Relevance, and Landing Page Experience. Each rated Below Average / Average / Above Average.",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/6167118",+ "notes": "The three-component structure has been consistent. QS is a keyword-level metric. The internal weighting of components is not published by Google.",+ "stale_after_days": 365+ },+ {+ "id": "conversion-tracking-prerequisite",+ "category": "conversion-tracking",+ "rule": "Conversion tracking is a prerequisite for all Smart Bidding strategies (tCPA, tROAS, Maximize Conversions, Maximize Conversion Value). Without it, only Manual CPC, Maximize Clicks, and Target Impression Share are available.",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/1722054",+ "notes": "This fundamental requirement is unlikely to change. Enhanced conversions and server-side tagging are expanding the ways to track, but the requirement itself is stable.",+ "stale_after_days": 365+ },+ {+ "id": "budget-is-campaign-level",+ "category": "campaign-structure",+ "rule": "Daily budget is set at the campaign level. Google may spend up to 2x the daily budget on any given day but will not exceed 30.4x the daily budget in a calendar month (monthly spending limit).",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/1704443",+ "notes": "The 2x daily / 30.4x monthly overdelivery policy has been stable. Shared budgets across campaigns are also available.",+ "stale_after_days": 365+ },+ {+ "id": "impression-share-metrics-available",+ "category": "benchmarks",+ "rule": "Search IS, top IS, abs top IS, budget-lost IS, and rank-lost IS are standard metrics available via GAQL and the Google Ads UI.",+ "volatility": "stable",+ "last_verified": "2026-04-04",+ "source": "Google Ads API docs",+ "source_url": "https://developers.google.com/google-ads/api/fields/v17/metrics",+ "notes": "These metrics have been stable across API versions. Field names may change between API versions.",+ "stale_after_days": 365+ },+ {+ "id": "qs-component-weights",+ "category": "quality-score",+ "rule": "Approximate QS component weights: Expected CTR ~35-40%, Ad Relevance ~20-25%, Landing Page Experience ~35-40%. Landing Page + Expected CTR together account for ~70-80% of QS.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "industry reports",+ "source_url": null,+ "notes": "Google has never published official weights. These are estimates from third-party regression studies (WordStream, Optmyzr). Use directionally, not as exact figures.",+ "stale_after_days": 180+ },+ {+ "id": "qs-cpc-impact-approximations",+ "category": "quality-score",+ "rule": "QS CPC impact relative to QS 5 baseline: QS 10 = ~50% discount, QS 7 = ~28% discount, QS 3 = ~67% premium, QS 1 = ~400% premium. Rule of thumb: +1 QS above 5 saves ~16% CPC; -1 below 5 costs ~50% more.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "industry reports",+ "source_url": null,+ "notes": "Approximate estimates from WordStream and Optmyzr studies. Actual CPC modifiers vary by auction, vertical, and account history. Not Google-published data.",+ "stale_after_days": 180+ },+ {+ "id": "rsa-pinning-impact",+ "category": "ad-copy",+ "rule": "Pinning RSA headlines/descriptions to specific positions reduces the number of combinations Google can test, which may lower ad strength and reduce serving. Pin only when regulatory or brand requirements demand it.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/9023565",+ "notes": "Google's RSA rotation algorithm evolves. The trade-off between pinning control and combination reduction remains, but the severity of the penalty may shift.",+ "stale_after_days": 180+ },+ {+ "id": "ab-testing-statistical-minimums",+ "category": "experiments",+ "rule": "A/B testing minimum: at least 30 clicks/day per variant for meaningful data. Below 30 clicks/day total across all campaigns, avoid A/B tests — statistical significance takes weeks. Allow 2-4 weeks minimum for any test.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": null,+ "notes": "The 30 clicks/day threshold is a practical guideline, not a Google-enforced rule. Lower volume accounts need longer test windows.",+ "stale_after_days": 180+ },+ {+ "id": "themed-ad-group-keyword-count",+ "category": "campaign-structure",+ "rule": "Current best practice: themed ad groups with 10-20 closely themed keywords per ad group. SKAGs (single keyword ad groups) are detrimental — LLM-based bidding needs signal density from keyword clusters.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": null,+ "notes": "This shifted from SKAG (2012-2017) to STAG (2017-2020) to themed (2020+). Google's recommendation engine and smart bidding perform better with fewer, larger ad groups.",+ "stale_after_days": 180+ },+ {+ "id": "themed-ad-group-best-practices",+ "category": "campaign-structure",+ "rule": "Theme definition: keywords should share the same intent AND could be answered by the same landing page. 2-3 RSAs per ad group. 1 primary landing page per ad group. Exact + broad match is the standard pairing.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": null,+ "notes": "The exact+broad pairing reflects phrase and broad match now overlapping ~100% in many verticals, making phrase match redundant alongside broad.",+ "stale_after_days": 180+ },+ {+ "id": "target-impression-share-brand",+ "category": "bidding",+ "rule": "For brand campaigns, use Target Impression Share bidding targeting 95%+ top of page. Set max CPC cap to prevent runaway costs. Brand campaigns should get 5-10% of total budget.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/7381968",+ "notes": "The strategy itself is stable, but optimal impression share targets may shift with competitive dynamics and SERP layout changes.",+ "stale_after_days": 180+ },+ {+ "id": "negative-keyword-match-type-behavior",+ "category": "match-types",+ "rule": "Negative keywords do NOT expand to close variants, synonyms, or paraphrases like positive keywords do. Negative exact blocks only that exact term. Negative phrase blocks queries containing the exact phrase. Must add negatives for each variant explicitly.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/2453972",+ "notes": "This asymmetry between positive and negative match types is a frequent source of wasted spend. Google could change this behavior, which would significantly alter negative keyword strategy.",+ "stale_after_days": 180+ },+ {+ "id": "consent-mode-requirements",+ "category": "privacy",+ "rule": "Google Consent Mode v2 is required for EEA/UK traffic to maintain conversion measurement and remarketing. Must implement consent_mode with ad_storage and analytics_storage parameters. Without it, conversion data gaps in EU markets.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/10000067",+ "notes": "Privacy regulations evolve. Consent Mode implementation requirements may expand to other regions. Google may add new consent signal types.",+ "stale_after_days": 180+ },+ {+ "id": "close-variants-all-match-types",+ "category": "match-types",+ "rule": "Google applies close variants (misspellings, plurals, stemming, synonyms, paraphrases, implied words) to ALL match types including exact match. Exact match keywords can match to synonyms and paraphrases Google considers same-intent.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/7478529",+ "notes": "Close variant expansion has been one-directional (always expanding). The scope of what constitutes a 'close variant' grows over time. Exact match is no longer truly exact.",+ "stale_after_days": 180+ },+ {+ "id": "bid-strategy-migration-ladder",+ "category": "bidding",+ "rule": "Standard migration path: New campaign -> Maximize Conversions (2-4 weeks) -> Target CPA (once learning exits) -> Target ROAS (if value tracking active and values vary 3x+). Allow direct tCPA if account has cross-campaign conversion history.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": null,+ "notes": "Google increasingly encourages starting with automated bidding from day one. The rigid step-by-step ladder may become unnecessary as algorithms improve.",+ "stale_after_days": 180+ },+ {+ "id": "display-network-baseline-metrics",+ "category": "benchmarks",+ "rule": "Display network baseline: CTR is typically 8-12x lower than Search CTR. CPC is typically 3-5x lower than Search CPC. Conversion rates are typically 3-4x lower than Search. Display is an awareness and remarketing channel, not direct response.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "industry reports",+ "source_url": null,+ "notes": "Display performance ratios relative to Search have been broadly consistent. Specific numbers shift with market conditions.",+ "stale_after_days": 180+ },+ {+ "id": "impression-share-benchmarks-by-type",+ "category": "benchmarks",+ "rule": "Impression share targets: Brand campaigns 90-95%+, Non-brand competitive 40-60%, Non-brand long-tail 60-80%, Competitor terms 20-40%. Brand IS < 80% is a red flag (competitors stealing brand traffic).",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": null,+ "notes": "These are diagnostic benchmarks, not hard rules. Competitive dynamics and budget constraints affect achievable IS.",+ "stale_after_days": 180+ },+ {+ "id": "campaign-split-by-match-type-harmful",+ "category": "campaign-structure",+ "rule": "Splitting campaigns by match type (separate Exact, Phrase, Broad campaigns) is harmful. It fragments conversion data and starves Smart Bidding. Phrase and broad overlap ~100% in many verticals, making three-way splits pointless.",+ "volatility": "moderate",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": null,+ "notes": "This was common practice pre-2020. The shift to consolidated structures reflects Smart Bidding's need for signal density.",+ "stale_after_days": 180+ },+ {+ "id": "smart-bidding-learning-phase",+ "category": "bidding",+ "rule": "Smart Bidding learning phase is now variable, not a fixed duration. Google states 'typically 1-2 weeks' but it depends on conversion volume, account history, and bid strategy. Do not judge results or make changes during the learning period.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/7065882",+ "notes": "Previously cited as a fixed 7-day learning phase. Google now uses variable duration based on multiple signals. The 2-week minimum wait advice is a practical guideline.",+ "stale_after_days": 90+ },+ {+ "id": "broad-match-semantic-reach",+ "category": "match-types",+ "rule": "Broad match uses AI to determine intent and can match to themes, related concepts, and queries with no keyword words present (e.g., 'plumber services' matches 'fix leaky faucet'). Phrase and broad match overlap ~100% in many verticals (2025-2026).",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/7478529",+ "notes": "Broad match reach continues to expand with Google's LLM improvements. The phrase-broad overlap may become complete, potentially making phrase match redundant.",+ "stale_after_days": 90+ },+ {+ "id": "ecpc-deprecation-status",+ "category": "bidding",+ "rule": "Enhanced CPC (eCPC) was deprecated for new Search campaigns in 2024. Existing campaigns may still use it but Google no longer recommends it. Cannot be selected for new campaigns. Migrate to Maximize Conversions or Target CPA.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/2390684",+ "notes": "Google may fully sunset eCPC for existing campaigns at any time. Monitor for forced migration announcements.",+ "stale_after_days": 90+ },+ {+ "id": "pmax-asset-requirements",+ "category": "pmax",+ "rule": "Performance Max campaigns require asset groups with text assets (headlines, descriptions), images, videos, and logos. PMax serves across Search, Display, YouTube, Discover, Gmail, and Maps. PMax may serve text ads on Search inventory.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/10724817",+ "notes": "PMax capabilities and asset requirements change frequently. New asset types and placement options are added regularly. PMax search cannibalization behavior evolves.",+ "stale_after_days": 90+ },+ {+ "id": "industry-cpc-cpa-ctr-benchmarks",+ "category": "benchmarks",+ "rule": "Industry benchmarks (2024-2025 vintage): Legal CPC $6.75/CTR 4.4%/CPA $90, Home Services CPC $3.00/CTR 5.0%/CPA $37.50, E-commerce CPC $1.50/CTR 4.5%/CPA $37.50, SaaS CPC $4.20/CTR 3.8%/CPA $93. Well-managed accounts outperform averages by 30-50%.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "industry reports",+ "source_url": null,+ "notes": "Based on WordStream, Google, and LOCALiQ reports. CPCs rising 10-20% YoY due to increased competition and AI bidding adoption. Adjust 2024-2025 figures upward 10-20% for 2026.",+ "stale_after_days": 90+ },+ {+ "id": "manual-cpc-signal-disadvantage",+ "category": "bidding",+ "rule": "Manual CPC forfeits access to Google's real-time LLM-based auction signals (device, location, time, audience, query intent, competitive density, browser, OS, and hundreds more). Not recommended for any new campaign.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": null,+ "notes": "The signal disadvantage grows as Google's bidding LLM improves. The gap between manual and automated bidding performance is widening over time.",+ "stale_after_days": 90+ },+ {+ "id": "brand-exclusions-account-suitability",+ "category": "pmax",+ "rule": "PMax supports brand exclusions to prevent ads from showing on competitor brand terms or unwanted brand queries. Account-level brand suitability controls allow excluding specific brands from all campaign types.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/12850226",+ "notes": "Brand exclusion capabilities are expanding. Initially PMax-only, now spreading to other campaign types. Exact behavior and coverage evolving.",+ "stale_after_days": 90+ },+ {+ "id": "search-themes-negative-keyword-precision",+ "category": "pmax",+ "rule": "PMax Search Themes guide the algorithm toward specific search intents but reduce negative keyword precision — PMax negative keywords may not block as precisely as Search campaign negatives because the algorithm considers themes holistically.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/13542443",+ "notes": "Search Themes feature is relatively new. Its interaction with negative keywords and the degree of control it provides is still evolving.",+ "stale_after_days": 90+ },+ {+ "id": "experiment-traffic-splitting",+ "category": "experiments",+ "rule": "Google Ads experiments use cookie-based traffic splitting (not search-based), meaning the same user consistently sees the same variant. This avoids cross-contamination but means split is user-based, not query-based.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/6318732",+ "notes": "With cookie deprecation discussions and privacy changes, the traffic splitting mechanism may change. Monitor for shifts to search-based or cohort-based splitting.",+ "stale_after_days": 90+ },+ {+ "id": "campaign-mix-experiments",+ "category": "experiments",+ "rule": "Campaign-level experiments allow A/B testing of bid strategies, ad copy, landing pages, and targeting. Campaign Mix Experiments for testing budget allocation across campaigns may have limited availability.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/6318732",+ "notes": "Google is expanding experiment capabilities. Multi-campaign experiments and portfolio-level testing features are in various stages of rollout.",+ "stale_after_days": 90+ },+ {+ "id": "attribution-model-dda-default",+ "category": "conversion-tracking",+ "rule": "Data-Driven Attribution (DDA) is the default and recommended attribution model. Google removed Last Click, First Click, Linear, Time Decay, and Position-Based models for new conversion actions (2023-2024). Existing actions may still use legacy models.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/6259715",+ "notes": "Google is moving toward DDA-only. Legacy models may be fully sunsetted. DDA uses machine learning to credit touchpoints based on their actual impact.",+ "stale_after_days": 90+ },+ {+ "id": "ai-overviews-ctr-impact",+ "category": "benchmarks",+ "rule": "AI Overviews in Google Search are reducing organic CTR and pushing more demand to paid placements, driving CPC inflation of 10-20% YoY across most verticals. Position 1 paid ads may see higher CTR as AI Overviews push organic results further down.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "industry reports",+ "source_url": null,+ "notes": "AI Overviews rollout is ongoing and impact varies by query type and vertical. The effect on paid vs organic CTR distribution is still being measured.",+ "stale_after_days": 90+ },+ {+ "id": "device-bid-adjustments-smart-bidding",+ "category": "bidding",+ "rule": "Under Smart Bidding (tCPA, tROAS, Maximize Conversions, Maximize Conversion Value), manual device bid adjustments are overridden — Smart Bidding adjusts bids by device automatically per auction. Only Manual CPC and eCPC respect device bid adjustments.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/2732132",+ "notes": "The -100% mobile bid adjustment still works to exclude mobile entirely, even under Smart Bidding. As Smart Bidding adoption approaches universal, device bid adjustments become less relevant.",+ "stale_after_days": 90+ },+ {+ "id": "pmax-search-cannibalization",+ "category": "pmax",+ "rule": "PMax campaigns can serve on Search inventory alongside Search campaigns. When both target the same queries, exact match keywords in Search campaigns take priority over PMax for identical queries. Behavior for broad/phrase is less clear.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": "https://support.google.com/google-ads/answer/10724817",+ "notes": "The priority rules between PMax and Search campaigns evolve. Google has stated exact match priority but the behavior for broad/phrase may change.",+ "stale_after_days": 90+ },+ {+ "id": "cpc-yoy-inflation-rate",+ "category": "benchmarks",+ "rule": "Google Ads CPCs have been rising 10-20% year-over-year (2024-2026) due to increased competition, AI-driven bidding adoption, and SERP layout changes (AI Overviews reducing organic clicks). Adjust historical benchmarks upward accordingly.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "industry reports",+ "source_url": null,+ "notes": "Inflation rate varies by vertical. Some verticals (legal, insurance) see higher inflation. This rate could accelerate or decelerate.",+ "stale_after_days": 90+ },+ {+ "id": "llm-bidding-signal-advantage",+ "category": "bidding",+ "rule": "Google's bidding engine uses an LLM-based auction system evaluating hundreds of real-time signals per auction: device type, precise location, time of day, audience segments, query intent, competitive density, browser, OS, and more. No human bidder can access or process these signals.",+ "volatility": "high",+ "last_verified": "2026-04-04",+ "source": "Google Ads Help",+ "source_url": null,+ "notes": "Google describes their bidding as AI/ML-powered. The 'LLM-based' characterization reflects the current generation. The signal set and model architecture evolve continuously.",+ "stale_after_days": 90+ }+ ]+}diff --git a/google-ads/shared/preamble.md b/google-ads/shared/preamble.mdindex bee9cd27..324edf0c 100644--- a/google-ads/shared/preamble.md+++ b/google-ads/shared/preamble.md@@ -59,6 +59,10 @@ Use whichever MCP server prefix was detected:Always pass `accountId` from `~/.adsagent/config.json` to every tool call (except `listConnectedAccounts`).+### Prefer GAQL for multi-campaign reads++When a workflow needs data from 2+ campaigns, use `runGaqlQuery` with bulk queries instead of per-campaign helper calls. See `../shared/gaql-cookbook.md` for ready-to-use query patterns. This typically reduces API calls from `N × data_types` to just the number of data types (e.g., 7 queries instead of 30+). Fall back to per-campaign helpers if GAQL errors or you need >50 rows for a single campaign.+## Step 5: ProceedConfig is loaded. Hand control back to the invoking skill.