As a diagnostic AI and Google Ads Quality Score specialist, I can help you untangle this. For a D2C coffee subscription like Ritual Roast with a $38 AOV, high CPCs eat into your initial acquisition margins incredibly fast. A mediocre Quality Score (5-6) dragged down primarily by **Expected CTR** tells me your ads are likely bleeding impressions to irrelevant close variants, or your ad copy isn't strongly differentiating your subscription from local roasters or grocery brands on the SERP. Here is your diagnostic and recovery roadmap based on 2026 Google Ads behavior. ## STEP 0: FRAMEWORK ROUTING **Decision:** Search Keyword-QS Framework. Your campaign mix is Search + Shopping. We will run the Search keyword-QS framework for your Search campaigns. Please note that Shopping campaigns do not have keyword-level Quality Score; they run on a completely different set of quality signals (feed health, price competitiveness, historical product conversion rate) which are not visible as a 1-10 metric. This diagnosis applies strictly to your Search campaigns. --- ## STEP 1: QS DIAGNOSTIC FRAMEWORK Quality Score is a live, auction-time computation based on three core components. Here is how they break down for your Search campaigns: **Expected CTR** — Google's prediction of whether your ad will be clicked at the keyword's average position. * Drivers: Ad copy strength, keyword-to-ad alignment, historical CTR data, negative keyword hygiene. * Common Fixes: Benefit-driven headlines, adding robust assets (sitelinks, callouts), and aggressive negative keywords to stop irrelevant impressions from dragging down your averages. **Ad Relevance** — How closely the ad matches the keyword's intent. * Drivers: Keyword presence in the headline, ad group thematic tightness (STAG structure), message match. * Common Fixes: Restructuring to STAGs (Single Theme Ad Groups of 5-15 keywords), ensuring distinct ad groups don't mix "buy coffee beans" intent with "coffee subscription box" intent. **Landing Page Experience** — Quality and relevance of your post-click destination. * Drivers: 2026 Core Web Vitals (LCP under 2.5s, INP under 200ms, CLS under 0.1), mobile usability, trust elements. * Common Fixes: Aligning the landing page H1 with the ad headline, optimizing mobile site speed, providing a frictionless checkout. ### Diagnostic Decision Tree Since your **WEAKEST_COMPONENT** is **Expected CTR**, investigate in this order: 1. **Search Term Report Hygiene:** Are close variants triggering your ads for one-off bag purchases when you only sell subscriptions? Irrelevant impressions destroy CTR. 2. **Ad Copy Differentiation:** Does your copy sound exactly like every other coffee roaster? 3. **Asset Utilization:** Are you missing sitelinks, callouts, or structured snippets that would make your ad physically larger on mobile screens? --- ## STEP 2: ACCOUNT-WIDE QS PROFILE Your account exhibits a "Mostly 5-6" pattern. Here is what that means for Ritual Roast: | Pattern | Likely Root | Where to Look First | Recovery Difficulty | | :--- | :--- | :--- | :--- | | Mostly 5-6 (Mediocre) | Stale creative, loose match types, or poor negative keyword hygiene. | Search Terms report vs. Keyword intent; Ad Asset coverage. | Moderate — usually fixable with structural tightening and copy refreshes. | **Red Flag for this pattern:** A 5-6 average usually means you are paying a slight premium on every click. In a $38 AOV business, this CPC penalty directly impacts your break-even ROAS. It also indicates your account isn't fundamentally broken, but it is leaking efficiency. --- ## STEP 3: AD GROUP TRIAGE FRAMEWORK Do not try to fix the whole account at once. Batching massive changes will force Smart Bidding back into a learning phase and scramble your data. Triage your ad groups by impact: 1. High-spend + Low-QS ad groups → fix first (biggest waste) 2. Medium-spend + Low-QS ad groups → fix second 3. Low-spend + Low-QS ad groups → fix last or prune 4. High-spend + High-QS → protect, do not touch ### Triage Worksheet Template | Ad Group Name | 30-Day Spend | QS Range | Weakest Component | Action Required | | :--- | :--- | :--- | :--- | :--- | | e.g., Best Coffee Subs | $1,200 | 4-5 | Expected CTR | Add negative KW lists, rewrite H1 | | e.g., Dark Roast Beans | $850 | 5-6 | Ad Relevance | Split into separate STAG | | [Fill in] | [Fill in] | [Fill in] | [Fill in] | [Fill in] | --- ## STEP 4: COMPONENT-SPECIFIC RECOVERY PLAYBOOK ### Expected CTR Recovery (Your Primary Focus) * **Critical:** Deploy aggressive negative keywords. In 2026, exact match includes paraphrased queries and same-intent variants by default. If your subscription ad shows for "cheap coffee near me," nobody clicks, and your Expected CTR tanks. * **High:** Rewrite headlines using specific, benefit-driven formulas. Examples: * *Formula 1 (Clarity + Offer):* Fresh Roasted Coffee Delivered | Save 15% On Your First Box * *Formula 2 (Social Proof + Feature):* 10,000+ Happy Subscribers | Single-Origin Beans Weekly * *Formula 3 (Objection Handling):* Pause Or Cancel Anytime | Specialty Coffee Subscription * **Medium:** Add all applicable assets (formerly extensions) like sitelinks, callouts, and structured snippets to dominate SERP real estate. * **Low:** Pause keywords with a sustained low CTR only after they have accrued significant impressions (e.g., >500) without contributing to conversions. ### Ad Relevance Recovery * **High:** Tighten ad groups to a STAG pattern (5-15 closely related keywords). Never use SKAGs (Single Keyword Ad Groups) in 2026; they will cannibalize each other due to close variants. * **Medium:** Pin a high-intent keyword variant to Headline position 1 only if strict message matching is required. Otherwise, let Google's machine learning rotate them. * **Low:** Eliminate ad groups that mix intents (e.g., separating "gift a coffee sub" from "buy coffee sub for myself"). ### Landing Page Experience Recovery * **High:** Ensure your landing page H1 exactly matches the core promise of your ad headline. * **High:** Optimize for 2026 Core Web Vitals: LCP under 2.5s, INP under 200ms, and CLS under 0.1. * **Medium:** Ensure mobile-first design with thumb-friendly CTAs and fast, secure forms. --- ## STEP 5: 30-DAY RECOVERY ROADMAP *Warning: Smart Bidding (tCPA/tROAS) resets its learning phase after meaningful changes (7-14 days). Stagger your updates.* **Days 1-7 (Triage)** * Identify the top 5 high-spend, low-QS ad groups. * Diagnose the weakest component for each. * Document baseline metrics (CPAs, CPCs, QS). * DO NOT make changes yet — observe the baseline. **Days 8-14 (Quick Wins, Staggered)** * Add account-wide assets (sitelinks, callouts). This has a low impact on the learning phase. * Implement aggressive negative keyword lists to cut wasted impressions. * Rewrite underperforming headlines for the top 3 ad groups (staggered over the week). **Days 15-21 (Structural Fixes)** * Restructure one campaign at a time into tighter STAGs for better Ad Relevance. * Update landing page H1s to match your new ad copy. * Address obvious mobile speed bottlenecks. **Days 22-30 (Measurement)** * Compare the QS distribution before and after. * Allow Smart Bidding to exit the learning phase before judging CPA/ROAS results. * Document what worked and plan the next iteration. --- ## STEP 6: MEASUREMENT FRAMEWORK Quality Score updates lag. Real changes usually reflect in the UI 2-3 weeks after action. Focus on group-level patterns, not daily fluctuations. | Metric | Review Cadence | What to Watch For | | :--- | :--- | :--- | | QS Distribution Histogram | Weekly | Track the percentage of keywords moving from the 5-6 tier to the 7+ tier. | | Component-Level QS | Bi-Weekly | Watch for **Expected CTR** moving from "Below Average" to "Average". This is your leading indicator. | | CPC Trend by Ad Group | Weekly | Look for stabilization or reduction in Avg. CPC as QS improves. | | Impression Share Lost to Rank | Weekly | A decrease here indicates your auction competitiveness is improving without raising bids. | | Negative KW Conflict Report | Weekly | Ensure your new negative keywords aren't blocking legitimate subscription traffic. | | Ad Strength | Monthly | Diagnostic only. Industry data shows "Average" ad strength often outperforms "Excellent" on CPA. Do not sacrifice conversion-driving copy just to appease the Ad Strength meter. | --- ## STEP 7: WHAT QS WILL NOT FIX Your primary symptom is that "CPCs feel too high." While fixing **Expected CTR** will improve your auction-time rank and potentially lower CPCs, set realistic expectations. Quality Score recovery alone will not solve: * **Tight Margins on a $38 AOV:** If competitor bids have driven the floor CPC to $3.00, a perfect QS won't make it $0.50. You may need to optimize for Customer Lifetime Value (LTV) rather than immediate ROAS. * **Shopping Campaign Inefficiencies:** Your Shopping campaigns do not use keyword QS. If Shopping CPCs are too high, you need to diagnose your feed titles, price competitiveness, and product imagery. * **Wrong Audience Targeting:** High CTR on bad traffic just spends your money faster. Are your current Search ads pushing users to a general homepage, or do you have dedicated landing pages built specifically for the subscription offer?