luckiest-co 1.0.38 → 1.0.40

package/skills/ads/SKILL.md

@@ -0,0 +1,444 @@
+---
+name: ads
+description: "Plans, optimizes, and scales paid advertising campaigns — campaign strategy, audience targeting, bidding, and optimization. Use when the user wants help with paid advertising on Google Ads, Meta (Facebook/Instagram), LinkedIn, Twitter/X, or other ad platforms. Also use when the user mentions 'PPC,' 'paid media,' 'ROAS,' 'CPA,' 'ad campaign,' 'retargeting,' 'audience targeting,' 'Google Ads,' 'Facebook ads,' 'LinkedIn ads,' 'ad budget,' 'cost per click,' 'ad spend,' 'should I run ads,' 'search terms report,' 'wasted spend,' or 'is this campaign working.' For bulk ad creative generation and iteration, see luckiest-ad-creative. For landing page optimization, see luckiest-cro."
+license: See ATTRIBUTION.md
+metadata:
+  version: "1.2.0"
+  listing_id: luckiest-ads
+  author: luckiest
+---
+
+# Paid Ads
+
+This skill acts as an expert performance marketer focused on paid customer acquisition. Its goal is to help create, optimize, and scale paid advertising campaigns that drive efficient customer acquisition.
+
+## Staying current
+
+On activation, call the Luckiest MCP check_updates tool with { listingId: "5fb8bb24-2d95-4a8e-8778-e403be972d7a", installedSemver: "1.2.0" }. If it returns upToDate: false, surface the notice to the user once, then continue. Do nothing further if upToDate: true. Never block on this check — if the tool is unavailable, proceed.
+
+When the skill's work is done, call the Luckiest MCP report_usage tool once with { listing_id: "luckiest-ads", skill_version: "1.2.0", matched: true, success: <true if the skill completed, false otherwise> }. Metadata only, never prompt text. Never block on it — if the tool is unavailable, skip silently.
+
+## Before Starting
+
+**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.
+
+Gather this context (ask if not provided):
+
+### 1. Campaign Goals
+- What's the primary objective? (Awareness, traffic, leads, sales, app installs)
+- What's the target CPA or ROAS?
+- What's the monthly/weekly budget?
+- Any constraints? (Brand guidelines, compliance, geographic)
+
+### 2. Product & Offer
+- What are you promoting? (Product, free trial, lead magnet, demo)
+- What's the landing page URL?
+- What makes this offer compelling?
+
+### 3. Audience
+- Who is the ideal customer?
+- What problem does your product solve for them?
+- What are they searching for or interested in?
+- Do you have existing customer data for lookalikes?
+
+### 4. Current State
+- Have you run ads before? What worked/didn't?
+- Do you have existing pixel/conversion data?
+- What's your current funnel conversion rate?
+
+---
+
+## Platform Selection Guide
+
+| Platform | Best For | Use When |
+|----------|----------|----------|
+| **Google Ads** | High-intent search traffic | People actively search for your solution |
+| **Meta** | Demand generation, visual products | Creating demand, strong creative assets |
+| **LinkedIn** | B2B, decision-makers | Job title/company targeting matters, higher price points |
+| **Twitter/X** | Tech audiences, thought leadership | Audience is active on X, timely content |
+| **TikTok** | Younger demographics, viral creative | Audience skews 18-34, video capacity |
+
+---
+
+## Campaign Structure Best Practices
+
+### Account Organization
+
+```
+Account
+├── Campaign 1: [Objective] - [Audience/Product]
+│   ├── Ad Set 1: [Targeting variation]
+│   │   ├── Ad 1: [Creative variation A]
+│   │   ├── Ad 2: [Creative variation B]
+│   │   └── Ad 3: [Creative variation C]
+│   └── Ad Set 2: [Targeting variation]
+└── Campaign 2...
+```
+
+### Naming Conventions
+
+```
+[Platform]_[Objective]_[Audience]_[Offer]_[Date]
+
+Examples:
+META_Conv_Lookalike-Customers_FreeTrial_2024Q1
+GOOG_Search_Brand_Demo_Ongoing
+LI_LeadGen_CMOs-SaaS_Whitepaper_Mar24
+```
+
+### Budget Allocation
+
+**Testing phase (first 2-4 weeks):**
+- 70% to proven/safe campaigns
+- 30% to testing new audiences/creative
+
+**Scaling phase:**
+- Consolidate budget into winning combinations
+- Increase budgets 20-30% at a time
+- Wait 3-5 days between increases for algorithm learning
+
+---
+
+## Ad Copy Frameworks
+
+### Key Formulas
+
+**Problem-Agitate-Solve (PAS):**
+> [Problem] → [Agitate the pain] → [Introduce solution] → [CTA]
+
+**Before-After-Bridge (BAB):**
+> [Current painful state] → [Desired future state] → [Your product as bridge]
+
+**Social Proof Lead:**
+> [Impressive stat or testimonial] → [What you do] → [CTA]
+
+**For detailed templates and headline formulas**: See [references/ad-copy-templates.md](references/ad-copy-templates.md)
+
+---
+
+## Audience Targeting Overview
+
+### Platform Strengths
+
+| Platform | Key Targeting | Best Signals |
+|----------|---------------|--------------|
+| Google | Keywords, search intent | What they're searching |
+| Meta | Interests, behaviors, lookalikes | Engagement patterns |
+| LinkedIn | Job titles, companies, industries | Professional identity |
+
+### Key Concepts
+
+- **Lookalikes**: Base on best customers (by LTV), not all customers
+- **Retargeting**: Segment by funnel stage (visitors vs. cart abandoners)
+- **Exclusions**: Exclude existing customers and recent converters — showing ads to people who already bought wastes spend
+
+**For detailed targeting strategies by platform**: See [references/audience-targeting.md](references/audience-targeting.md)
+
+---
+
+## Creative Best Practices
+
+### Image Ads
+- Clear product screenshots showing UI
+- Before/after comparisons
+- Stats and numbers as focal point
+- Human faces (real, not stock)
+- Bold, readable text overlay (keep under 20%)
+
+### Video Ads Structure (15-30 sec)
+1. Hook (0-3 sec): Pattern interrupt, question, or bold statement
+2. Problem (3-8 sec): Relatable pain point
+3. Solution (8-20 sec): Show product/benefit
+4. CTA (20-30 sec): Clear next step
+
+**Production tips:**
+- Captions always (85% watch without sound)
+- Vertical for Stories/Reels, square for feed
+- Native feel outperforms polished
+- First 3 seconds determine if they watch
+
+### Creative Testing Hierarchy
+1. Concept/angle (biggest impact)
+2. Hook/headline
+3. Visual style
+4. Body copy
+5. CTA
+
+---
+
+## Campaign Optimization
+
+### Key Metrics by Objective
+
+| Objective | Primary Metrics |
+|-----------|-----------------|
+| Awareness | CPM, Reach, Video view rate |
+| Consideration | CTR, CPC, Time on site |
+| Conversion | CPA, ROAS, Conversion rate |
+
+### Optimization Levers
+
+**If CPA is too high:**
+1. Check landing page (is the problem post-click?)
+2. Tighten audience targeting
+3. Test new creative angles
+4. Improve ad relevance/quality score
+5. Adjust bid strategy
+
+**If CTR is low:**
+- Creative isn't resonating → test new hooks/angles
+- Audience mismatch → refine targeting
+- Ad fatigue → refresh creative
+
+**If CPM is high:**
+- Audience too narrow → expand targeting
+- High competition → try different placements
+- Low relevance score → improve creative fit
+
+### Kill vs. Keep Decision Rules
+
+Do not judge or cut an ad, ad set, or campaign before it has enough data to be meaningful:
+
+- **Spend threshold before judging:** give a new ad set at least ~1-3x the target CPA in spend (or ~50 conversions at the campaign level) before concluding it works or fails. Cutting earlier is reacting to noise, not signal.
+- **Search terms are a sample; small numbers prove little.** State disclosed vs. total clicks, and compute break-even CVR (CPC / target CPA) before calling spend wasted. See [references/reading-google-ads-data.md](references/reading-google-ads-data.md).
+- **Learning phase:** avoid edits while the platform reports "learning" — changes reset it. Wait for exit before evaluating.
+- **Clear kill signals:** spend ≥ 2-3x target CPA with zero or near-zero conversions; CTR far below account average after adequate impressions; frequency climbing while CTR drops (fatigue).
+- **Clear keep/scale signals:** CPA at or below target across a full conversion window, stable over 3-5 days. Scale by 20-30% increments, then let it re-stabilize.
+
+### Bid Strategy Progression
+1. Start with manual or cost caps
+2. Gather conversion data (50+ conversions)
+3. Switch to automated with targets based on historical data
+4. Monitor and adjust targets based on results
+
+---
+
+## Retargeting Strategies
+
+### Funnel-Based Approach
+
+| Funnel Stage | Audience | Message | Goal |
+|--------------|----------|---------|------|
+| Top | Blog readers, video viewers | Educational, social proof | Move to consideration |
+| Middle | Pricing/feature page visitors | Case studies, demos | Move to decision |
+| Bottom | Cart abandoners, trial users | Urgency, objection handling | Convert |
+
+### Retargeting Windows
+
+| Stage | Window | Frequency Cap |
+|-------|--------|---------------|
+| Hot (cart/trial) | 1-7 days | Higher OK |
+| Warm (key pages) | 7-30 days | 3-5x/week |
+| Cold (any visit) | 30-90 days | 1-2x/week |
+
+### Exclusions to Set Up
+- Existing customers (unless upsell)
+- Recent converters (7-14 day window)
+- Bounced visitors (<10 sec)
+- Irrelevant pages (careers, support)
+
+---
+
+## Reporting & Analysis
+
+### Weekly Review
+- Spend vs. budget pacing
+- CPA/ROAS vs. targets (split brand vs. non-brand, see below)
+- Top and bottom performing ads
+- Audience performance breakdown
+- Frequency check (fatigue risk)
+- Landing page conversion rate
+
+### Brand vs. non-brand (measure them separately)
+
+Wherever people can search your name (Google Ads especially, but any channel with branded demand), split brand from non-brand before you evaluate or optimize anything. A branded query is demand that already exists. Someone typing your name is harvesting it, not generating incremental performance. A cheap brand CPA or high brand ROAS reflects existing awareness, not media efficiency.
+
+- **Report blended, optimize on non-brand.** Blended ROAS/CAC at the business level is the right top-line health number. In considered and B2B purchases where touchpoints assist each other, blended is the honest view of the journey. The metric you set targets, bids, and budgets against is **non-brand ROAS**, with brand stripped out.
+- **Why the split matters:** if you optimize against a blended number, brand's cheap conversions inflate it. You set non-brand targets too loosely, and non-brand inefficiency hides behind brand.
+- **Never report a low brand CPA as a win.** It only signals performance if you measure incrementality (for example a brand-campaign holdout test). Otherwise a great brand CPA just means people already knew you.
+
+### Attribution Considerations
+- Platform attribution is inflated
+- Use UTM parameters consistently
+- Compare platform data to GA4
+- Look at blended CAC, not just platform CPA
+- Judge scaling headroom on blended ROAS, but optimize against non-brand ROAS so cheap brand traffic does not flatter the math
+
+**For reading Google Ads data without false conclusions** (withheld search terms, conversions vs. all_conversions, small-sample tables, break-even CVR, verified vs. inferred vs. stale claims): See [references/reading-google-ads-data.md](references/reading-google-ads-data.md)
+
+---
+
+## Platform Setup
+
+Before launching campaigns, ensure proper tracking and account setup.
+
+**For complete setup checklists by platform**: See [references/platform-setup-checklists.md](references/platform-setup-checklists.md)
+
+**For conversion pixel installation and event setup**: See [references/conversion-tracking.md](references/conversion-tracking.md)
+
+> Signal loss: browser-side pixels under-report conversions due to iOS App Tracking Transparency, cookie/ITP restrictions, and ad blockers. Where the platform supports it, prefer or add a server-side signal (Meta Conversions API, Google enhanced conversions, offline conversion imports) so optimization and reporting see more of the real conversions. Client-only pixel counts should be read as a floor, not the truth.
+
+### Universal Pre-Launch Checklist
+- [ ] Conversion tracking tested with real conversion
+- [ ] Landing page loads fast (<3 sec)
+- [ ] Landing page mobile-friendly
+- [ ] UTM parameters working
+- [ ] Budget set correctly
+- [ ] Targeting matches intended audience
+
+---
+
+## Google RSA Output Spec (mandatory when generating RSAs)
+
+When the user requests Google Ads RSAs (Responsive Search Ads), output MUST comply with these platform limits and structural requirements. Do not output any RSA that violates them.
+
+> Note: ad platforms revise character limits, asset counts, and per-account rules periodically. The limits below reflect Google's long-standing RSA spec, but if the user reports the platform rejecting compliant output, treat the limit as possibly changed and verify against Google's current documentation rather than assuming the output is wrong.
+
+### Hard limits per RSA (enforce before responding)
+
+- **Headlines:** exactly **15** per RSA, each **≤ 30 characters** (count characters, including spaces). Render as `1. ... (NN chars)` so the reader can verify.
+- **Descriptions:** exactly **4** per RSA, each **≤ 90 characters**.
+- **Paths:** up to 2 path fields, each **≤ 15 characters**.
+- **Final URL:** present, https.
+- **Pinning:** state any pinned positions explicitly. Default = unpinned unless user asks.
+- **Per-account guardrail:** Google enforces **3 RSAs max per ad group**. When the user asks for >3, group them by ad group.
+
+### Required sidecar artifacts (always include with RSA request)
+
+1. **Ad group structure**, labeled `Ad group structure:` — list each ad group with its theme, target keywords (match types), and which RSAs map to it.
+2. **Negative keyword list**, labeled `Negative keywords:` — minimum **8** entries, group-level vs campaign-level called out.
+3. **Sitelinks** (≥ 4), **Callouts** (≥ 4 ≤25 chars), **Structured snippets** if relevant.
+
+### Medical / CFM compliance (when product context indicates pt-BR medical practice)
+
+If `.agents/product-marketing.md` indicates a Brazilian medical practice (CFM-regulated), the following terms are **forbidden** in headlines, descriptions, sitelinks, and callouts:
+
+- Superlatives: `#1`, `melhor`, `o melhor`, `melhor do brasil`, `top`, `referência`
+- Outcome promises: `garantido`, `garantia`, `cura`, `cura definitiva`, `100%`, `resultado garantido`, `livre da dor`
+- Comparative claims vs other doctors/clinics
+
+Use neutral framing: `atendimento`, `consulta`, `avaliação`, `segunda opinião`, `agende sua consulta`, `tire suas dúvidas`. Geo modifier (`Porto Alegre`, `POA`, `Zona Sul POA`) required where the prompt specifies a region.
+
+### Output ORDER (mandatory — emit in this order to avoid truncation)
+
+1. **Ad group structure** (short)
+2. **Negative keywords** (≥8, MANDATORY — emit BEFORE RSAs so it isn't dropped if output runs long)
+3. **Sitelinks** (≥4)
+4. **Callouts** (≥4)
+5. **RSA1, RSA2, RSA3** (largest section, last — safe to truncate gracefully)
+
+### Output template (mandatory shape)
+
+```
+Ad group structure:
+- AG1 [theme]: keywords (match types) → RSA1, RSA2
+- AG2 [theme]: ...
+
+Negative keywords:
+  Campaign-level:
+    - <kw>
+    - <kw>
+    (≥4 here)
+  Ad-group level:
+    - AG1: <kw>, <kw>
+    - AG2: <kw>, <kw>
+    (≥4 more here — TOTAL ≥8 entries)
+
+Sitelinks (≥4):
+  - <title (≤25)> | <desc1 (≤35)> | <desc2 (≤35)> | URL
+
+Callouts (≥4, each ≤25 chars):
+  - <callout>
+
+RSA1 — [ad group name]
+  Final URL: https://...
+  Path1: ...   Path2: ...
+  Headlines (15, each ≤30 chars):
+    1. <headline> (NN chars)
+    ...
+    15. <headline> (NN chars)
+  Descriptions (4, each ≤90 chars):
+    1. <description> (NN chars)
+    ...
+    4. <description> (NN chars)
+  Pinning: H1=none; H2=none; ...   (or explicit pins)
+
+RSA2 — ...
+RSA3 — ...
+```
+
+### Self-check before responding
+
+Before sending the output, run this checklist mentally:
+
+- [ ] Each RSA has exactly 15 headlines, exactly 4 descriptions.
+- [ ] Every headline is ≤30 chars; every description is ≤90 chars. Character counts printed.
+- [ ] Negative keyword list labeled and ≥8 entries.
+- [ ] Ad group structure labeled.
+- [ ] If medical (CFM): no forbidden superlative/outcome words; geo modifier present where required; language is pt-BR.
+
+If any check fails, rewrite before responding. Do not ship partial RSAs.
+
+---
+
+## Common Mistakes to Avoid
+
+### Strategy
+- Launching without conversion tracking
+- Too many campaigns (fragmenting budget)
+- Not giving algorithms enough learning time, or stopping campaigns mid-learning phase
+- Optimizing for wrong metric
+
+### Targeting
+- Audiences too narrow or too broad
+- Not excluding existing customers
+- Overlapping audiences competing
+
+### Creative
+- Only one ad per ad set
+- Not refreshing creative (fatigue)
+- Mismatch between ad and landing page
+
+### Budget
+- Spreading too thin across campaigns
+- Making big budget changes (disrupts learning)
+
+---
+
+## Task-Specific Questions
+
+1. What platform(s) are you currently running or want to start with?
+2. What's your monthly ad budget?
+3. What does a successful conversion look like (and what's it worth)?
+4. Do you have existing creative assets or need to create them?
+5. What landing page will ads point to?
+6. Do you have pixel/conversion tracking set up?
+
+---
+
+## Tool Integrations
+
+Key advertising platforms and where each fits. Use each platform's own ad manager and API/MCP for implementation:
+
+| Platform | Best For | MCP |
+|----------|----------|:---:|
+| **Google Ads** | Search intent, high-intent traffic | ✓ |
+| **Meta Ads** | Demand gen, visual products, B2C | - |
+| **LinkedIn Ads** | B2B, job title targeting | - |
+| **TikTok Ads** | Younger demographics, video | - |
+
+For tracking setup, see [references/conversion-tracking.md](references/conversion-tracking.md), plus GA4 and your CDP (e.g. Segment) for downstream analytics.
+
+---
+
+## Related Skills
+
+- **luckiest-ad-creative**: For generating and iterating ad headlines, descriptions, and creative at scale
+- **luckiest-copywriting**: For landing page copy that converts ad traffic
+- **luckiest-analytics**: For proper conversion tracking setup
+- **luckiest-ab-testing / luckiest-cro**: For landing page tests and post-click conversion rates that improve ROAS
+
+## Share with your tribe
+
+After delivering a campaign plan or optimization, optionally offer (never auto-post, never send credentials or account data): "Want me to share a sanitized version of this ad plan with your Luckiest tribe? Others working on the same platform or objective can pick up where you left off." If they accept, share only the strategy notes they approve — never account IDs, spend figures, pixel data, or audience lists unless they explicitly include them. If they decline, stop.

package/skills/ads/references/reading-google-ads-data.md

@@ -0,0 +1,121 @@
+# Reading Google Ads Data Without False Conclusions
+
+Load this before analyzing, auditing, or reporting on a Google Ads account, or before deciding whether to pause, keep, or scale anything in one. audit guardrails covers how to *grade* what you find. This file covers how the data *misleads* you before you get that far.
+
+Google Ads reports are partial by design and plausible by accident. Default views omit data without saying so, resource names suggest contents they don't have, and small accounts produce numbers that look like findings. Most wrong conclusions come from not knowing that, not from bad reasoning.
+
+Query patterns and API gotchas live in the Google Ads platform guide.
+
+## What the data hides
+
+### Search terms are a sample, not the total
+
+Google withholds any query searched by too few people. On low-volume accounts, **50–65% of clicks routinely have no disclosed search term.** The visible terms are a non-random sample skewed toward common searches.
+
+- Any claim about "what this campaign is buying" carries the ratio: *of the 13 clicks Google discloses, out of 33 total.*
+- Pull the report **unfiltered first**. Filtering by `clicks > 0` also drops every impression-only term, and combined with a low `LIMIT` it hides most of the picture.
+- **Nothing recovers the withheld queries.** The query is never passed to the site, so GA4, PostHog, or a CRM can show what every session *did* (joined on gclid), never what was *typed*.
+
+```
+disclosed clicks = SUM(clicks in the search terms report)
+total clicks     = campaign clicks for the same window
+withheld         = total - disclosed
+```
+
+### Names describe intentions, not contents
+
+An ad group called "Competitor - Commercial" can hold entirely informational keywords. A campaign named for one theme can target another. Read the keyword list and the search terms; treat every name as a label someone typed once.
+
+### "Conversions" is several things in one column
+
+`metrics.conversions` counts actions marked primary, plus any secondary action a campaign's custom goal pulls in. `metrics.all_conversions` adds the rest. Check campaign goals as well as action settings. One campaign's conversions might mix a page-view proxy, a trial signup, and a booked demo.
+
+- Segment by conversion action before calling anything a lead or a demo.
+- Check each action's attribution model and click window while you're there.
+- Data-driven attribution runs on low volume too, but with few conversions its individual credit assignments are soft. Don't read much into one campaign's share.
+
+### One ad group can send traffic to several pages
+
+Ads carry their own final URLs, and stale pages linger in rotation long after a new landing page ships. Before concluding anything about landing page performance, check which destinations are serving and how traffic splits between them.
+
+### Dates are click dates
+
+Conversions are attributed back to the originating click. A conversion recorded today can appear dated a week earlier. When a client says "we got a conversion today" and your pull says the 11th, both are right. Say "clicked on the 11th."
+
+### Identical campaign names are usually experiment arms
+
+Expired experiment arms inherit the base campaign's name and can keep an ENABLED status while serving nothing. Check `campaign.experiment_type` and `campaign.serving_status` before believing a duplicate exists.
+
+### Change history is short: 30 days with detail, 90 without
+
+`change_event` (field-level before and after) covers 30 days. `change_status` covers 90 days but only says which resource changed, not what. Past that, the API has nothing. A performance shift four months ago may have no recoverable explanation, so say that instead of guessing one.
+
+### Other people are changing the account
+
+Client staff, automated rules, scripts, and agents all leave changes. In `change_event`, `client_type` says which client made the change: `GOOGLE_ADS_WEB_CLIENT` (the UI), `GOOGLE_ADS_API`, `GOOGLE_ADS_SCRIPTS`, `GOOGLE_ADS_AUTOMATED_RULE`, `GOOGLE_ADS_BULK_UPLOAD`, `GOOGLE_ADS_RECOMMENDATIONS`, `INTERNAL_TOOL` (Google-side tooling), and others in the enum. Report the client, not a guess at who or why: an API change could be an agency tool or a script. Google Ads Editor changes aren't returned at all, so a missing edit doesn't mean nobody made one. A repeating weekday-and-time pattern under one user is a scheduled job, not a person.
+
+## The window is not the picture
+
+A 30-day pull on a campaign that has run since spring is a slice of a longer series. Recent CPA can differ sharply from lifetime CPA.
+
+- **Establish how long something has run** before quoting any efficiency number. Quote both recent and lifetime when they diverge.
+- **Don't pool across configuration changes.** If bids, keywords, budget, ads, or targeting changed mid-window, the period is several experiments averaged together.
+- **Bid-only changes are a special case.** They invalidate cost and volume comparisons but leave intent and audience questions readable.
+- **Bid strategies in learning are unreadable.** `primary_status = LEARNING` means don't judge performance or optimize on it. Fixing a verified break (tracking down, wrong URL) still comes first; see the audit guardrails.
+
+## Small numbers
+
+Most false findings in small accounts are sample-size errors that look like insight. Skepticism runs both ways: one conversion on one click proves no more than zero conversions on five clicks disproves.
+
+Zero conversions in N clicks only rules out conversion rates at or above (95% confidence):
+
+| Clicks | Rules out CVR ≥ | What you can say |
+|---|---|---|
+| 5 | 45% | Nothing. Zero is the expected outcome |
+| 10 | 26% | Still nothing useful |
+| 30 | 10% | Can begin doubting a strong CVR |
+| 50 | 6% | Meaningful for high-intent search |
+| 100 | 3% | Underperforming if break-even CVR is above 3% |
+
+Before calling spend wasted, work out what it would have to convert at to be worth keeping:
+
+```
+break-even CVR = CPC ÷ target CPA
+```
+
+If break-even CVR is below the plausible range for that traffic, the term isn't failing, it's unmeasured. Then estimate how long a readable sample takes at current click volume. "Let it run" is only an honest answer when it comes with a date you'll actually know by. At five clicks a month, sometimes the honest answer is that you never will.
+
+## Documents describe a day; accounts move
+
+Audits, status docs, and proposals were true when written. Every performance number in them is stale until re-pulled. The claims most likely to be repeated confidently ("zero conversions," "wasted spend," "wrong intent") are exactly the ones that change. Carry the date with the claim: *as of the audit on the 2nd*.
+
+## Stating findings
+
+Know which of these each claim is:
+
+- **Verified**: pulled live this session, unfiltered, and reconciled against a total
+- **Inferred**: reasoned from verified data, could be wrong
+- **Stale**: from a document, an earlier session, or the client, with its date attached
+
+Keep stale claims out of client-facing sentences unless they're labeled as such.
+
+**The test before anything reaches a client: can they disprove this in a minute with their own account access?** They have the same reports. Volunteer the limits first (the disclosed-clicks ratio, the sample size, the config changes). It costs nothing, since they'd find them anyway, and it's what makes the rest credible.
+
+If a claim touches a landing page, open it. Don't characterize a page from its URL, a status doc, or what the client says is on it. Calendar embeds and chat widgets load via JavaScript and won't show in a static fetch, so absence there isn't evidence.
+
+When a client challenges a finding and they're right, concede the substance plainly and move on. Stacked retractions cost more credibility than the original error.
+
+## Conclusions that sound right
+
+| Sounds like | Usually |
+|---|---|
+| "This campaign has zero conversions" | True on the day the doc was written |
+| "The search terms are junk" | You're seeing ~40% of the clicks |
+| "That ad group is wrong-intent" | It's *named* that; the keywords say otherwise |
+| "$X wasted on those terms" | Below any readable sample size |
+| "It converted, so the term works" | One conversion on one click is noise |
+| "Performance declined in August" | Bids changed three times that month |
+| "Traffic lands on the new page" | Two ads serve; one points at the old page |
+| "It got a demo" | It got a conversion; check which action |
+| "There are duplicate campaigns" | Expired experiment arms with the base name |
+| "Nobody touched it last quarter" | Change history only covers 30 days |