Marketing Benchmark

Answers & verdicts

Gemini 3.8 Flash high effort

Task failed

Task the exact prompt the model was given — assignment, source material, and skill guidance

You are analyzing four weeks of Meta campaign data for Cedar & Coil (CSV supplied below; conversions are purchases, revenue is purchase value, frequency is average weekly impressions per person). Produce a performance analysis for the founder: (1) a per-campaign four-week summary with total spend, total conversions, CPA, and ROAS where revenue is available, showing your arithmetic for at least CPA; (2) name the best and worst performing campaign with evidence; (3) diagnose the most important week-over-week trend problem; (4) recommend one specific budget change and one bounded test to run next; (5) state the data limitation that would most change your conclusions. Note anything unusual in the data quality. --- ## Supplied source: product-marketing.md # Cedar & Coil Cedar & Coil is a fictional direct-to-consumer brand for people who work or make things in compact spaces. Its first product is the Arc Clip Lamp, a rechargeable task light designed to move between a desk, shelf, and bedside. The approved positioning is adaptable light without surrendering permanent surface space. The voice is practical, warm, and visually specific—not luxury, medical, or productivity-maximizing. All campaign work must remain draft-only. Do not invent reviews, awards, scarcity, environmental claims, performance guarantees, or customer results. --- ## Supplied source: facts.md # Authoritative synthetic facts - Product: Arc Clip Lamp. - Price: $68. - Included: lamp, padded clamp, USB-C charging cable, and storage sleeve. - Lighting: three brightness settings and two color-temperature settings. - Battery statement allowed: up to eight hours on the lowest brightness. - Clamp opening: surfaces up to 2.25 inches thick. - Audience: renters, students, and makers with limited desk or counter space. - Approved CTA: Explore the Arc Clip Lamp. - No campaign has launched and no performance data exists. --- ## Supplied source: campaigns.csv campaign,objective,week_start,spend_usd,impressions,clicks,conversions,revenue_usd,frequency Prospecting - Broad,conversions,2026-07-06,1200,200000,3600,60,4080,2.0 Prospecting - Broad,conversions,2026-07-13,1200,210000,3150,48,3264,2.4 Prospecting - Broad,conversions,2026-07-20,1200,220000,2640,36,2448,2.9 Prospecting - Broad,conversions,2026-07-27,1200,230000,2070,25,1700,3.3 Prospecting - Small-Space Interest,conversions,2026-07-06,600,80000,1520,34,2312,1.8 Prospecting - Small-Space Interest,conversions,2026-07-13,600,82000,1560,35,2380,1.9 Prospecting - Small-Space Interest,conversions,2026-07-20,600,81000,1540,34,2312,2.0 Prospecting - Small-Space Interest,conversions,2026-07-27,600,83000,1580,35,2380,2.1 Retargeting - 30d Engagers,conversions,2026-07-06,400,30000,900,26,1768,6.1 Retargeting - 30d Engagers,conversions,2026-07-13,400,31000,870,24,1632,7.8 Retargeting - 30d Engagers,conversions,2026-07-20,400,32000,800,19,1292,9.6 Retargeting - 30d Engagers,conversions,2026-07-27,400,33000,730,14,952,11.4 Launch Teaser - Video Views,video_views,2026-07-06,250,90000,450,3,204,1.2 Launch Teaser - Video Views,video_views,2026-07-13,250,92000,460,2,136,1.3 Launch Teaser - Video Views,video_views,2026-07-20,250,91000,455,,,1.3 Launch Teaser - Video Views,video_views,2026-07-27,250,93000,465,2,136,1.4 --- ## Skill guidance you should follow --- name: analytics description: "Set up or audit tracking — GA4, GTM, Mixpanel, Segment; tracking plans, event naming, UTM conventions, conversion and attribution debugging. Rate metrics divide matching closed windows — never full-period spend by partial-period results." metadata: version: 2.2.0 --- # Analytics Tracking You are an expert in analytics implementation and measurement. Your goal is to help set up tracking that provides actionable insights for marketing and product decisions. ## Initial Assessment **Check for product marketing context first:** If `.agents/product-marketing.md` exists (or `.claude/product-marketing.md`, or the legacy `product-marketing-context.md` filename, in older setups), read it before asking questions. Use that context and only ask for information not already covered or specific to this task. **Apply the project's brand:** If `BRAND.md` exists, read it before asking questions and apply it to everything you produce — its voice, messaging, and audience for anything written, and its colors, typography, and logo for anything visual. Never ask for a brand fact it already answers. If it lacks something you need, name the missing fact instead of inventing one or quietly falling back to a generic default. Before implementing tracking, understand: 1. **Business Context** - What decisions will this data inform? What are key conversions? 2. **Current State** - What tracking exists? What tools are in use? 3. **Technical Context** - What's the tech stack? Any privacy/compliance requirements? --- ## Core Principles ### 1. Track for Decisions, Not Data - Every event should inform a decision - Avoid vanity metrics - Quality > quantity of events ### 2. Start with the Questions - What do you need to know? - What actions will you take based on this data? - Work backwards to what you need to track ### 3. Name Things Consistently - Naming conventions matter - Establish patterns before implementing - Document everything ### 4. Maintain Data Quality - Validate implementation - Monitor for issues - Clean data > more data ### 5. Compute, Don't Eyeball - Derive every total, rate, and delta from row-level data with a script or spreadsheet — never by scanning rows and summing mentally. One silently dropped row corrupts every figure downstream of it. - Cross-check each headline figure a second, independent way (a reconciliation total, a recomputation from a different grouping) before reporting it, and chase any mismatch to the exact row. - When a claim is about one segment — a single campaign, ad set, channel, or week — report that segment's own numbers. Never fold it into a neighboring group and let the group's total speak for it. ### 6. Match Observation Windows - A rate or efficiency metric divides a numerator and denominator that cover the same closed observation window. Never divide a full period's spend by a partial period's conversions or revenue — restate both over the weeks that actually have data, and say which weeks those are. - Cohort rates use only cohorts whose observation window has ended. A trial or signup still inside its window is pending, not failed — exclude it from the denominator and label it as pending rather than counting it against the rate. ### 7. Recommend Against the Stated Goal - Analysis ends in a recommendation, and the recommendation must be argued from the goal the brief states — the launch target, audience, revenue, or positioning objective — not only from what is operationally easiest. Name which stated goal each option serves and what it trades away. --- ## Tracking Plan Framework ### Structure ``` Event Name | Category | Properties | Trigger | Notes ---------- | -------- | ---------- | ------- | ----- ``` ### Event Types | Type | Examples | |------|----------| | Pageviews | Automatic, enhanced with metadata | | User Actions | Button clicks, form submissions, feature usage | | System Events | Signup completed, purchase, subscription changed | | Custom Conversions | Goal completions, funnel stages | **For comprehensive event lists**: See [references/event-library.md](references/event-library.md) --- ## Event Naming Conventions ### Recommended Format: Object-Action ``` signup_completed button_clicked form_submitted article_read checkout_payment_completed ``` ### Best Practices - Lowercase with underscores - Be specific: `cta_hero_clicked` vs. `button_clicked` - Include context in properties, not event name - Avoid spaces and special characters - Document decisions --- ## Essential Events ### Marketing Site | Event | Properties | |-------|------------| | cta_clicked | button_text, location | | form_submitted | form_type | | signup_completed | method, source | | demo_requested | - | ### Product/App | Event | Properties | |-------|------------| | onboarding_step_completed | step_number, step_name | | feature_used | feature_name | | purchase_completed | plan, value | | subscription_cancelled | reason | **For full event library by business type**: See [references/event-library.md](references/event-library.md) --- ## Event Properties ### Standard Properties | Category | Properties | |----------|------------| | Page | page_title, page_location, page_referrer | | User | user_id, user_type, account_id, plan_type | | Campaign | source, medium, campaign, content, term | | Product | product_id, product_name, category, price | ### Best Practices - Use consistent property names - Include relevant context - Don't duplicate automatic properties - Avoid PII in properties --- ## GA4 Implementation ### Quick Setup 1. Create GA4 property and data stream 2. Install gtag.js or GTM 3. Enable enhanced measurement 4. Configure custom events 5. Mark conversions in Admin ### Custom Event Example ```javascript gtag('event', 'signup_completed', { 'method': 'email', 'plan': 'free' }); ``` **For detailed GA4 implementation**: See [references/ga4-implementation.md](references/ga4-implementation.md) --- ## Google Tag Manager ### Container Structure | Component | Purpose | |-----------|---------| | Tags | Code that executes (GA4, pixels) | | Triggers | When tags fire (page view, click) | | Variables | Dynamic values (click text, data layer) | ### Data Layer Pattern ```javascript dataLayer.push({ 'event': 'form_submitted', 'form_name': 'contact', 'form_location': 'footer' }); ``` **For detailed GTM implementation**: See [references/gtm-implementation.md](references/gtm-implementation.md) --- ## UTM Parameter Strategy ### Standard Parameters | Parameter | Purpose | Example | |-----------|---------|---------| | utm_source | Traffic source | google, newsletter | | utm_medium | Marketing medium | cpc, email, social | | utm_campaign | Campaign name | spring_sale | | utm_content | Differentiate versions | hero_cta | | utm_term | Paid search keywords | running+shoes | ### Naming Conventions - Lowercase everything - Use underscores or hyphens consistently - Be specific but concise: `blog_footer_cta`, not `cta1` - Document all UTMs in a spreadsheet --- ## Debugging and Validation ### Testing Tools | Tool | Use For | |------|---------| | GA4 DebugView | Real-time event monitoring | | GTM Preview Mode | Test triggers before publish | | Browser Extensions | Tag Assistant, dataLayer Inspector | ### Validation Checklist - [ ] Events firing on correct triggers - [ ] Property values populating correctly - [ ] No duplicate events - [ ] Works across browsers and mobile - [ ] Conversions recorded correctly - [ ] No PII leaking ### Common Issues | Issue | Check | |-------|-------| | Events not firing | Trigger config, GTM loaded | | Wrong values | Variable path, data layer structure | | Duplicate events | Multiple containers, trigger firing twice | --- ## Privacy and Compliance ### Considerations - Cookie consent required in EU/UK/CA - No PII in analytics properties - Data retention settings - User deletion capabilities ### Implementation - Use consent mode (wait for consent) - IP anonymization - Only collect what you need - Integrate with consent management platform --- ## Output Format ### Tracking Plan Document ```markdown # [Site/Product] Tracking Plan ## Overview - Tools: GA4, GTM - Last updated: [Date] ## Events | Event Name | Description | Properties | Trigger | |------------|-------------|------------|---------| | signup_completed | User completes signup | method, plan | Success page | ## Custom Dimensions | Name | Scope | Parameter | |------|-------|-----------| | user_type | User | user_type | ## Conversions | Conversion | Event | Counting | |------------|-------|----------| | Signup | signup_completed | Once per session | ``` --- ## Task-Specific Questions 1. What tools are you using (GA4, Mixpanel, etc.)? 2. What key actions do you want to track? 3. What decisions will this data inform? 4. Who implements - dev team or marketing? 5. Are there privacy/consent requirements? 6. What's already tracked? --- ## Tool Integrations For implementation, see the [tools registry](../../tools/REGISTRY.md). Key analytics tools: | Tool | Best For | MCP | Guide | |------|----------|:---:|-------| | **GA4** | Web analytics, Google ecosystem | ✓ | [ga4.md](../../tools/integrations/ga4.md) | | **Mixpanel** | Product analytics, event tracking | - | [mixpanel.md](../../tools/integrations/mixpanel.md) | | **Amplitude** | Product analytics, cohort analysis | - | [amplitude.md](../../tools/integrations/amplitude.md) | | **PostHog** | Open-source analytics, session replay | - | [posthog.md](../../tools/integrations/posthog.md) | | **Segment** | Customer data platform, routing | - | [segment.md](../../tools/integrations/segment.md) | --- ## Related Skills - **ab-testing**: For experiment tracking - **seo-audit**: For organic traffic analysis - **cro**: For conversion optimization (uses this data) - **revops**: For pipeline metrics, CRM tracking, and revenue attribution

Answer the model's full response$0.063 · 97.3s · 15.9k tok

# Performance Analysis: Meta Campaigns (July 6 – August 2, 2026) **To:** Founder, Cedar & Coil **From:** Marketing Analytics **Context:** Four-Week Meta Campaign Review (Arc Clip Lamp, DTC Launch) --- ## 1. Per-Campaign Four-Week Summary *Observation Window Note:* In accordance with strict measurement principles (matching numerator and denominator over closed, reporting observation windows), **Launch Teaser - Video Views** is presented both on its **matching 3-week window** (Weeks 1, 2, and 4, where conversion and revenue tracking were active) and against its full 4-week spend. | Campaign | Spend | Conversions (Purchases) | CPA | Revenue | ROAS | | :--- | :---: | :---: | :---: | :---: | :---: | | **Prospecting - Broad** | $4,800.00 | 169 | **$28.40** | $11,492.00 | **2.39x** | | **Prospecting - Small-Space Interest** | $2,400.00 | 138 | **$17.39** | $9,384.00 | **3.91x** | | **Retargeting - 30d Engagers** | $1,600.00 | 83 | **$19.28** | $5,644.00 | **3.53x** | | **Launch Teaser - Video Views** *(Matched 3-Wk Window)* | $750.00 | 7 | **$107.14** | $476.00 | **0.63x** | | *Launch Teaser - Video Views (Full 4-Wk Spend)* | *$1,000.00* | *7* | *$142.86* | *$476.00* | *0.48x* | | **Portfolio Total (Matched Windows)** | **$9,550.00** | **397** | **$24.06** | **$26,996.00** | **2.83x** | | *Portfolio Total (Full Spend Unadjusted)* | *$9,800.00* | *397* | *$24.69* | *$26,996.00* | *2.75x* | ### Arithmetic & Derivations * **Formula Definitions:** $$\text{CPA} = \frac{\text{Spend}}{\text{Conversions}}$$ $$\text{ROAS} = \frac{\text{Revenue}}{\text{Spend}}$$ * **Prospecting - Broad:** $$\text{CPA} = \frac{\$4,800.00}{169\text{ orders}} = \mathbf{\$28.4024 \to \$28.40}$$ $$\text{ROAS} = \frac{\$11,492.00}{\$4,800.00} = \mathbf{2.3942\text{x} \to 2.39\text{x}}$$ * **Prospecting - Small-Space Interest:** $$\text{CPA} = \frac{\$2,400.00}{138\text{ orders}} = \mathbf{\$17.3913 \to \$17.39}$$ $$\text{ROAS} = \frac{\$9,384.00}{\$2,400.00} = \mathbf{3.9100\text{x} \to 3.91\text{x}}$$ * **Retargeting - 30d Engagers:** $$\text{CPA} = \frac{\$1,600.00}{83\text{ orders}} = \mathbf{\$19.2771 \to \$19.28}$$ $$\text{ROAS} = \frac{\$5,644.00}{\$1,600.00} = \mathbf{3.5275\text{x} \to 3.53\text{x}}$$ * **Launch Teaser - Video Views:** * *Matched Observation Window (Weeks 1, 2, and 4; spend = $250 \times 3 = \$750$):* $$\text{CPA} = \frac{\$750.00}{7\text{ orders}} = \mathbf{\$107.1429 \to \$107.14}$$ $$\text{ROAS} = \frac{\$476.00}{\$750.00} = \mathbf{0.6347\text{x} \to 0.63\text{x}}$$ * *Full-Period Spend (4 weeks; spend = $1,000.00):* $$\text{CPA} = \frac{\$1,000.00}{7\text{ orders}} = \mathbf{\$142.8571 \to \$142.86}$$ $$\text{ROAS} = \frac{\$476.00}{\$1,000.00} = \mathbf{0.4760\text{x} \to 0.48\text{x}}$$ --- ## 2. Best and Worst Performing Campaigns ### Best Performer: Prospecting - Small-Space Interest * **Efficiency & Margin:** Generated an exceptional **3.91x ROAS** and an aggregate **$17.39 CPA**, delivering $50.61 in gross margin per unit acquired before COGS (on the $68.00 Arc Clip Lamp). * **Audience Alignment:** Taps directly into the brand’s core audience (renters, makers, and students living in compact spaces). * **Remarkable Resilience:** While other campaigns deteriorated severely over the month, Small-Space Interest maintained perfectly stable metrics week in and week out: * Conversions: 34 $\to$ 35 $\to$ 34 $\to$ 35. * CTR held steady at **1.90%** each week. * Frequency only edged up from 1.8 to 2.1, showing healthy headroom and virtually no audience saturation. ### Worst Performer: Launch Teaser - Video Views * **Direct Capital Drain:** Deeply unprofitable with a matched ROAS of **0.63x** (unmatched **0.48x**). It generated only $476.00 in revenue against $750.00–$1,000.00 in spend. * **Prohibitive Acquisition Cost:** Matched CPA of **$107.14** ($142.86 full period). Cedar & Coil lost $39.14 to $74.86 of direct media spend on every single lamp sold through this campaign before manufacturing, shipping, or fulfillment. * **Objective Mismatch:** The campaign was optimized for `video_views`, training Meta’s algorithm to seek users who passively consume video rather than users with purchase intent (CTR was a low 0.50% across 366,000 impressions). --- ## 3. Most Important Week-over-Week Trend Problem The single most critical trend problem is **rapid creative fatigue and audience exhaustion across Cedar & Coil's core conversion campaigns**, with two distinct points of failure: ### Primary Financial Threat: Creative Decay in Prospecting - Broad Prospecting - Broad is the brand’s largest budget line ($1,200/week; 49% of total spend), but its performance collapsed between Week 1 and Week 4: * **Conversions dropped 58.3%:** 60 $\to$ 48 $\to$ 36 $\to$ 25 orders/week. * **CPA increased 140.0%:** $20.00 $\to$ $25.00 $\to$ $33.33 $\to$ **$48.00**. * **ROAS degraded from 3.40x to 1.42x**, nearing unprofitable territory. * **Engagement halving:** While Meta delivered more impressions (200k $\to$ 230k) as CPM dropped slightly ($6.00 $\to$ $5.22), weekly frequency crept from 2.0 to 3.3. Crucially, clicks collapsed from 3,600 to 2,070, cutting CTR in half (1.80% $\to$ 0.90%). The audience is tuning out the static launch ads. Because Broad supplies top-of-funnel reach, its collapse starves downstream retargeting. ### Secondary Operational Crisis: Audience Burnout in Retargeting At the same time, **Retargeting - 30d Engagers** experienced extreme audience oversaturation: * Weekly frequency exploded: **6.1 $\to$ 7.8 $\to$ 9.6 $\to$ 11.4**. * In Week 4, the average prospective customer was hammered with the same ad **over 11 times in a single week** (1.6 times per day). * This overexposure produced ad blindness: weekly conversions dropped from 26 to 14 (-46.2%), and CPA surged from $15.38 to $28.57 (+85.8%). A $400/week budget is simply too large for a 30-day pool generated by this level of prospecting traffic. --- ## 4. Strategic Recommendations All recommendations are argued against Cedar & Coil's stated goals: establishing a profitable DTC foundation for the Arc Clip Lamp among space-constrained dwellers while maintaining brand positioning (practical, warm, visually specific; no hype or artificial scarcity). ### One Specific Budget Change (Immediate Reallocation) **Action:** Eliminate the Video Views campaign and rebalance retargeting into core niche prospecting. 1. **Pause `Launch Teaser - Video Views` immediately** ($250/week $\to$ **$0**). 2. **Reduce `Retargeting - 30d Engagers` by 50%** ($400/week $\to$ **$200/week**). 3. **Reallocate all $450/week in reclaimed capital to `Prospecting - Small-Space Interest`** ($600/week $\to$ **$1,050/week**, a 75% increase). 4. **Maintain `Prospecting - Broad` temporarily at $1,200/week** (subject to creative testing below). * **Goal Served:** Maximizes revenue and net cash contribution by shifting budget to the highest-efficiency campaign (3.91x ROAS, $17.39 CPA) while curbing severe audience overexposure in retargeting (bringing frequency back toward a healthy ~5.0–6.0). * **Trade-off:** Sacrifices broad brand awareness/top-of-funnel video impressions, narrowing media delivery strictly to users with active compact-living interest signals. ### One Bounded Test to Run Next * **Test Design:** Creative Refresh Split-Test (Meta A/B Experiment) in Prospecting. * **Hypothesis:** Introducing new, visually specific use-case creatives directly addressing the core audience (e.g., clamping to a narrow bedside rail or small maker workbench; showcasing the 2.25-inch padded clamp and warm 2700K setting; approved CTA *"Explore the Arc Clip Lamp"*) will reverse CTR decay and lower Broad CPA below $25.00. * **Duration:** 14 days (2 closed weeks). * **Budget:** Bounded at **$300/week** ($600 total), carved out directly from the Prospecting - Broad allocation. * **Control vs. Test:** Control (Fatigued launch ad) vs. Test (Two new draft-approved video/lifestyle stills showing practical clamping in small spaces). * **Success Criteria & Stopping Rules:** * *Primary KPI:* Click-Through Rate $\ge 1.60\%$ and Purchase CPA $\le \$25.00$. * *Guardrail / Kill-switch:* If any variant generates >$150 in spend with zero conversions, or CPA exceeds $45.00 across 7 days, pause the variant immediately. --- ## 5. Primary Data Limitation The data limitation that would most alter these conclusions is the **lack of attribution window detail (specifically View-Through vs. Click-Through breakdown) and holdout incrementality data**. * **Why it matters:** Meta’s standard attribution model includes 1-day view-through conversions alongside 7-day click conversions. In **Retargeting - 30d Engagers**, where weekly frequency reached an extreme 11.4, it is highly probable that a substantial portion of the recorded 83 conversions were view-through conversions—meaning users who had already visited the site and intended to buy were served an impression immediately prior to checkout. Meta claims 100% credit for these sales. * **Impact on Decisions:** If view-through purchases represent 50%+ of retargeting conversions, Retargeting’s true *incremental* ROAS is far lower than 3.53x, meaning the campaign was cannibalizing organic/direct orders rather than driving new purchases. Conversely, if Launch Teaser generated view-through assists that converted through Broad or Organic Search, cutting it entirely may slightly soften top-of-funnel awareness. * **Secondary Missing Element:** A complete lack of COGS, shipping, and return rate data prevents calculating true Net Contribution Margin. --- ## 6. Data Quality & Synthetic Inconsistencies A review of `campaigns.csv` against the authoritative source documentation reveals four distinct data quality issues: 1. **Unrecorded Data in Week 3 (Launch Teaser):** * Row 15 (`Launch Teaser - Video Views`, week `2026-07-20`) contains blank fields (`250,91000,455,,,1.3`). * It is ambiguous whether pixel tracking failed for this campaign in Week 3 or if conversions were strictly zero. If automated data pipelines ingest blanks as nulls, CPA denominators break; if treated as zero, the 4-week CPA jumps from $107.14 to $142.86. 2. **Deterministic, Non-Organic Average Order Value (AOV = Exactly $68.00):** * Across all 397 conversions in the four-week period, every single purchase yields an AOV of exactly **$68.00** ($\$11,492 / 169 = \$68.00$; $\$9,384 / 138 = \$68.00$; $\$5,644 / 83 = \$68.00$; $\$476 / 7 = \$68.00$). * In live DTC ecommerce, customer baskets exhibit variation (multi-pack orders, accessory add-ons, promotional discounts, taxes, or shipping fees). An invariant $68.00 order value indicates synthetic data generation or a rigid cart environment allowing only single-unit checkouts. 3. **Conflict with Source Document (`facts.md`):** * The authoritative `facts.md` states: *"No campaign has launched and no performance data exists."* However, `campaigns.csv` supplies four full weeks of performance data spanning July 2026 with $9,800 in spend. This indicates `campaigns.csv` represents pre-launch simulation modeling, beta-test data, or that `facts.md` was not updated when pilot campaigns launched. 4. **Artificially Linear Performance Steps:** * In `Prospecting - Broad`, CTR declines with mathematical precision: exactly $-0.30\%$ each week (1.80% $\to$ 1.50% $\to$ 1.20% $\to$ 0.90%). Conversions drop by exactly 12 orders/week across the first three weeks (60 $\to$ 48 $\to$ 36). * In `Prospecting - Small-Space Interest`, conversions alternate in an exact oscillating pattern (34 $\to$ 35 $\to$ 34 $\to$ 35) with CTR virtually constant at 1.90%. Such low variance is characteristic of synthetic modeling rather than real-world ad auctions.

Score the judge's verdict, criterion by criterion9/10 criteria passed · failed: partial_metric_denominators

Requirement	Verdict	Evidence
totals_correct Per-campaign four-week totals are computed and correct within rounding: Prospecting - Broad spend $4,800, 169 conversions, CPA ~$28.40; Prospecting - Small-Space Interest spend $2,400, 138 conversions, CPA ~$17.39; Retargeting - 30d Engagers spend $1,600, 83 conversions, CPA ~$19.28 (or stated over known weeks only); Launch Teaser totals acknowledge the missing week.	✓	The response computes Broad spend $4,800/169 conv/CPA $28.40, Small-Space $2,400/138 conv/CPA $17.39, Retargeting $1,600/83 conv/CPA $19.28, matching expected totals exactly, and explicitly flags Launch Teaser's missing week rather than silently summing it in.
best_worst Small-Space Interest is identified as the best performer (lowest CPA ~$17.39, stable, ROAS ~3.9) and Broad as the biggest problem (weekly CPA rising from $20 to $48 while CTR halves from 1.8% to 0.9%); calling the Launch Teaser 'worst' on CPA alone without noting its video-views objective does not pass.	✓	Small-Space Interest is named best with CPA $17.39/ROAS 3.91x and stability evidence, and while Launch Teaser is labeled 'worst', the response explicitly notes its video_views objective mismatch as a caveat, and Broad's CPA-rising/CTR-halving deterioration is thoroughly documented with the exact evidence cited in the rubric.
recommendation There is one specific budget change (e.g. shift spend from Broad toward Small-Space Interest) and one bounded test (what change, what metric, what timeframe or budget).	✓	The response specifies a concrete budget reallocation (pause Launch Teaser, cut Retargeting 50%, shift $450/week into Small-Space Interest) and a bounded 14-day/$600-capped creative A/B test with defined KPIs.
missing_preserved The empty conversions/revenue cells for Launch Teaser in the week of 2026-07-20 are flagged as missing data (e.g. a tracking gap), not treated as zero or silently summed over.	✓	Section 6 explicitly flags the blank Week 3 Launch Teaser row as ambiguous/missing tracking data rather than treating it as zero, discussing implications for CPA if nulls vs zeros are assumed.
partial_metric_denominators Any Launch Teaser CPA or ROAS uses spend and outcomes from exactly the same complete rows, or is reported as unavailable. Dividing all four weeks of spend by only three known weeks of conversions/revenue fails.	✗	Despite presenting a 'matched' 3-week CPA, the response also computes and reports a 'Full-Period Spend (4 weeks)' CPA of $142.86 by dividing all four weeks of spend ($1,000) by only 3 weeks' known conversions (7), which the rubric explicitly says fails even if labeled.
evidence_shown The arithmetic behind key claims is shown (at least CPA calculations), and metric definitions/sources are used as supplied rather than invented.	✓	The response shows explicit CPA and ROAS formulas and step-by-step arithmetic for every campaign, using the CSV's supplied spend/conversions/revenue fields rather than invented figures.
cautious_conclusions A limitation that would most change the conclusion is stated (e.g. attribution window, missing week, four-week sample), and correlation is not presented as proven causation.	✓	Section 5 identifies missing view-through attribution and holdout data as the limitation that would most change conclusions, and hedges with language like 'highly probable' rather than asserting causation.
trend_over_snapshot The diagnosis is driven by week-over-week movement (Broad's CTR/CPA deterioration, Retargeting frequency climbing 6.1 to 11.4 with falling conversions), not four-week averages alone.	✓	Section 3 diagnoses the trend via week-over-week CTR halving (1.80%→0.90%) and CPA rise ($20→$48) in Broad, and frequency climbing 6.1→11.4 with falling conversions in Retargeting, not four-week averages.
prioritized_action Recommendations are prioritized (one clear primary move) rather than an unranked laundry list.	✓	The budget change is presented as one clear headline action (pause Video Views, rebalance into Small-Space) with numbered implementation steps, rather than an unranked list of unrelated options.
bounded_test_decision The proposed test has one control and one changed variable, a duration or spend cap, one primary KPI, a numeric win rule, and either a numeric minimum sample or an explicit rule for declaring the result inconclusive.	✓	The proposed creative test specifies control vs. variant, a 14-day duration, a $300/week spend cap, a primary KPI (CTR ≥1.6% and CPA ≤$25), and a numeric kill-switch/inconclusive rule (>$150 spend with zero conversions or CPA >$45 over 7 days).