analytics-insights

Files (11)

• SKILL.md
• anomaly-diagnosis.md
• clv-analysis.md
• competitive-intelligence.md
• dark-social-tracking.md
• dashboard-design.md
• incrementality-testing.md
• kpi-frameworks.md
• mmm-framework.md
• privacy-first-measurement.md
• reporting-templates.md

SKILL.md (first lines)

# Analytics & Insights

> **Metrics:** mention rate, citation rate, share of model, share of AI voice, sentiment and the AI Visibility Score are defined, with formulas, in `skills/context-engine/ai-visibility-metrics.md`. Report every score with its formula and its sample size (n).

## GA4 "AI Assistant" channel group (added 13 May 2026)

Google Analytics 4 added a new **default channel group called "AI Assistant"** on 13 May 2026 ([GA4 channel groups doc](https://support.google.com/analytics/answer/9164320?hl=en)). When a referrer matches a recognized AI Assistant (ChatGPT, Gemini, Claude, etc.), GA4 automatically:

- Categorizes the session under the **AI Assistant channel group**
- Sets the **Medium dimension to `ai-assistant`**

This is the **attribution-side counterpart** to the new GSC AI Performance Report (rolled out 3 June 2026 — see `/digital-marketing-pro:gsc-ai-performance`). Because the GSC AI report intentionally excludes click data, the GA4 AI Assistant channel is currently the cleanest path to attribute *actual traffic* coming from generative AI surfaces.

**Recommended GA4 setup checks** when onboarding a brand:

1. **Confirm the channel group is live in the property.** Newer GA4 properties get it automatically; older ones may need it to appear after Google's backfill completes. If the brand reports their channel reports look unchanged after 13 May, check explore reports filtered by `sessionDefaultChannelGroup = "AI Assistant"`.
2. **Add the AI Assistant channel to custom reports + dashboards** — for any brand running an AEO program (`/digital-marketing-pro:aeo-geo`, `/digital-marketing-pro:aeo-audit`), the AI Assistant channel trend is now a primary KPI alongside organic search clicks.
3. **Don't merge AI Assistant into "Organic Search" or "Direct".** Some legacy reporting templates roll AI traffic into Direct (because referrers weren't always present) or Organic Search (because answer engines feel "search-like"). Both are misattributions now — the AI Assistant channel is the authoritative bucket.
4. **Reconcile with `aeo-audit` outputs and the GSC AI report.** Three data sources, three different views:
   - `aeo-audit` (synthetic probing) — what AI engines *could* say about the brand
   - GSC AI Performance Report — actual impressions in Google AI Overviews / AI Mode (no clicks)
   - GA4 AI Assistant channel — actual *traffic* from non-Google AI assistants (clicks materialized)

   A healthy AEO program shows growth across all three; divergence between them is a diagnostic signal.

### The blind spot: Google's own AI surfaces are filed as Organic Search (checked 2026-10-04)

Google's default-channel definitions ([support.google.com/analytics/answer/9756891](https://support.google.com/analytics/answer/9756891)) say three things:
- **AI Assistant** rule: "The medium exactly matches 'ai-assistant'". GA4 sets that medium (and the campaign `(ai-assistant)`) when the referrer matches its list of AI assistants. Google's channel description names ChatGPT, Gemini, Deepseek, Copilot and Grok; the 13 May 2026 release note also names Claude.
- The AI Assistant channel "excludes Google's AI Overviews and AI Mode".
- **Organic Search** is the channel for non-ad links in organic-search results, "including Google's AI Overviews and AI Mode".

So a click from an AI Overview or from AI Mode is counted as Organic Search. The channel page documents no dimension that separates it from a classic blue-link click. Consequences:
- **Never report the AI Assistant channel as "all AI traffic."** It is non-Google assistant traffic only.
- **Never subtract or estimate an "AI Overviews share" of Organic Search** from GA4 data. No first-party metric for it exists, and an estimate presented as data is a fabrication.
- When AI Overviews impressions rise in Search Console while organic clicks fall, report both numbers side by side. Say plainly that GA4 cannot attribute the clicks to AI Overviews or AI Mode.

### One table per AI surface: what you can and cannot measure (checked 2026-10-04)

| Surface | Visibility metric (first-party) | Click / traffic metric | Not available |
|---|---|---|---|
| Google AI Overviews + AI Mode | Search Console generative AI report: impressions by page, country, date, device ([help](https://support.google.com/webmasters/answer/16984139)) | Inside GA4 Organic Search, inseparable | Clicks, CTR, and queries in the AI report; an AI-only slice of GA4 organic |
| Copilot, Bing, and select partner AI experiences | Bing Webmaster Tools AI Performance: citations, grounding queries, **Intents**, **Topics**, **Citation Share** (your share of all citations shown for a grounding query), and **Compare** periods. Intents, Topics, Citation Share and Compare are in preview ([Bing blog](https://blogs.bing.com/search/2026/6/New-AI-Visibility-Insights-in-Bing-Webmaster-Tools-Intents-Topics-Citation-Share-Compare/), 16 Jun 2026) | GA4 AI Assistant channel (Copilot is a named source) | Bing says Citation Share "does not expose competitor domains, represent traffic share, or assign quality scores to content" |
| Google Shopping in AI Mode / AI Overviews / Gemini app | Merchant Center **AI performance insights**: brand share of voice vs similar brands, across discovery, evaluation and purchase, plus popular product terms and specifications ([Merchant Center help](https://support.google.com/merchants/answer/17117204); [blog.google](https://blog.google/products-and-platforms/products/shopping/shopping-updates-google-marketing-live/), 20 May 2026) | Merchant Center / Ads conversion reporting as usual | Availability: Google said it is rolling out in the U.S., Canada, Australia, India and New Zealand "in the coming months" (as of May 2026). Confirm in the account before promising it |
| ChatGPT, Claude, Perplexity, Gemini app (answers) | **No first-party citation report exists.** Use synthetic probes (`/digital-marketing-pro:aeo-audit`, `/digital-marketing-pro:geo-monitor`) and label them as probes | GA4 AI Assistant channel | Any vendor-published citation or impression count |

Report each row in its own units. **Do not add Bing citations, Search Console impressions and GA4 sessions into one "AI visibility" number.** They measure different events on different surfaces.

## When to Use This Skill

Activate this module when the user's request involves any of the following:

- **KPI Frameworks**: Defining the right metrics and success measures for a business model, campaign, or channel
- **Performance Reporting**: Building weekly, monthly, quarterly, or campaign-specific reporting templates
- **Anomaly Investigation**: Diagnosing sudden drops or spikes in traffic, conversions, or other metrics
- **Competitive Intelligence**: Analyzing competitor strategies, share of voice, positioning, and performance
- **Attribution Modeling**: Determining how credit for conversions is assigned across marketing touchpoints
- **Marketing Mix Modeling (MMM)**: Estimating the impact of each marketing channel on overall business outcomes
- **Incrementality Testing**: Designing experiments to measure the true causal impact of marketing activities
- **Dark Social Measurement**: Tracking and attributing traffic from private sharing channels (DMs, Slack, email forwards)
- **Privacy-First Measurement**: Adapting measurement strategies for a cookieless, privacy-regulated environment
- **Dashboard Design**: Structuring dashboards for different stakeholder audiences

**Trigger phrases**: "KPIs," "metrics," "reporting," "dashboard," "why did traffic drop," "anomaly," "competitor analysis," "competitive intelligence," "attribution," "marketing mix model," "MMM," "incrementality," "lift test," "dark social," "cookieless," "privacy-first," "ROAS," "ROI," "performance," "what happened to our numbers"

## Brand Context (Auto-Applied)

Before producing any marketing output from this module:

1. **Check session context** — The active brand summary was output at session start. Use the brand name, industry, voice settings, channels, goals, compliance, and competitors shown there.
2. **If you need the full profile**, read: `~/.claude-marketing/brands/{slug}/profile.json`
3. **Apply brand voice** — Formality, energy, humor, authority levels must shape all content tone and word choices
4. **Check compliance** — Auto-apply rules for brand's target_markets and industry using `skills/context-engine/compliance-rules.md`
5. **Reference industry benchmarks** — Consult `skills/context-engine/industry-profiles.md` for the brand's industry
6. **Use platform specs** — Reference `skills/context-engine/platform-specs.md` for character limits and format requirements
7. **Check campaign history** — Run `python campaign-tracker.py --brand {slug} --action list-campaigns` before planning new work
8. **If no brand exists**, say: "No brand profile found. Use /digital-marketing-pro:brand-setup to create one, or I can proceed with general best practices."
9. **Check brand guidelines** — If `~/.claude-marketing/brands/{slug}/guidelines/_manifest.json` exists, load and enforce: `restrictions.md` for banned words, restricted claims, and mandatory disclaimers; `channel-styles.md` for channel-specific tone overrides (may differ from base voice); `messaging.md` for approved key messages, taglines, and positioning language; `voice-and-tone.md` for detailed voice rules beyond the 4 numeric scores. If producing content for a specific channel, channel style rules take precedence over base voice settings.

Do not ask the user for information that already exists in their brand profile.

## Required Context

Before executing analytics work, gather:

1. **Business Model**: SaaS, e-commerce, lead gen, marketplace, etc. (determines the KPI framework)
2. **Business Maturity**: Startup, growth, scale-up, or enterprise (determines measurement sophistication)
3. **Current Metrics**: What is already being tracked? What tools are in use?
4. **Analytics Stack**: Google Analytics (GA4), ad platforms, CRM, BI tools, CDPs, tag managers
5. **Data Availability**: How much historical data exists? What granularity?
6. **Reporting Audience**: Who receives reports? (Exec/C-suite, marketing team, board, clients)
7. **Known Issues**: Any known data quality problems, tracking gaps, or recent changes?
8. **Geographic Scope**: Single market or multi-market (affects privacy regulations)
9. **Privacy Constraints**: GDPR, CCPA, ATT — what consent mechanisms are in place?
10. **Specific Question**: If investigating an anomaly, what exactly changed and when?

For anomaly investigation, prioritize speed. Ask for the specific metric, timeframe, and any known changes. For strategic measurement work, gather the full context.

## Capabilities

- **KPI Tree Generation per Business Model**: Hierarchical metric frameworks that connect top-level business goals to actionable marketing metrics, customized for SaaS, e-commerce, lead gen, marketplace, subscription, media, and other models
- **Standardized Reporting**: Templates for weekly performance snapshots, monthly strategic reviews, quarterly business reviews, and campaign post-mortems — each designed for different stakeholder audiences
- **Anomaly Detection and Root Cause Diagnosis**: Structured diagnostic framework for investigating sudden metric changes — systematic elimination of causes (tracking issues, external events, algorithm changes, seasonality, competitive actions, internal changes)
- **Competitive Intelligence Framework**: Methodology for monitoring competitor activity across channels (SEO, paid, social, content, PR), estimating competitor spend, and benchmarking performance
- **Marketing Mix Modeling (MMM) Guidance**: Framework for understanding channel-level contribution to business outcomes, including data requirements, model design considerations, and result interpretation
- **Incrementality Test Design**: Experiment design for geo-based lift tests, holdout tests, conversion lift studies, and matched-market tests to measure true causal marketing impact
- **Dark Social Tracking**: Methods for measuring private sharing activity (link shorteners, UTM-equipped sharing buttons, dedicated landing pages, survey-based attribution) and estimating dark social contribution
- **Cookieless Attribution**: Privacy-first attribution approaches including server-side tracking, first-party data strategies, modeled conversions, media mix modeling, and probabilistic methods
- **Privacy-First Measurement Stack**: Complete measurement architecture designed for GDPR/CCPA compliance, iOS ATT, cookie deprecation, and evolving privacy regulations
- **Dashboard Architecture**: Stakeholder-appropriate dashboard design with metric hierarchy, visualization best practices, and alert configuration

## Process

**Primary Workflow: Measurement Framework & Reporting**

1. **Business Context & Goal Alignment**
   - Classify the business model and maturity stage
   - Identify the north star metric (the single metric most tied to business value)
   - Map business goals to marketing objectives to tactical metrics (KPI tree)

From indranilbanerjee/digital-marketing-pro, MIT.

More from indranilbanerjee/digital-marketing-pro