Skill: Analytics Context

Business context for Alaskan Fisherman analytics. Use alongside the Analytics MCP connector — this skill tells Claude what the data means and how to combine sources, not how to operate the tools technically.

Business context

Alaskan Fisherman sells premium fish products in Germany (DE) with a subscription/recurring-order component. All MCP analytics data is DE market only — CZ is not available.

The core funnel:

Unique Visitors — paid and organic traffic
Choose Your Fish — product page (primary campaign landing page)
Order Page — checkout
Created Orders — orders initiated
Paid Orders — orders completed and paid

Primary KPIs:

CR = Paid Orders / Unique Visitors — min > 0.60%, target > 2.00%, ambitious > 4.00%
CPA = Marketing Spend / Paid Orders — baseline around 220 EUR, target toward 100 EUR

Secondary: product→checkout transition (> 40%), unpaid rate (< 20%).

The site can run multiple variants concurrently. Always segment by variant when more than one is active — aggregate numbers across variants are misleading.

Canonical GA4 event names per funnel stage are in the knowledge base (../knowledge/semantic.md).

PostHog pageviews

PostHog-native active-user widgets use $pageview, not the custom page_view event. The custom page_view event is the GTM/GA4-shaped dataLayer page event used for schema parity and funnel joins.

Known timeline:

2026-05-04 20:43 CET: frontend PR #1 (feat/de-gtm-and-posthog-parity, SHA cd49a72) merged. It disabled PostHog auto pageviews and introduced custom page_view.
2026-05-05 through 2026-05-11 before the fix merge: custom page_view is the reliable onsite funnel-entry event; PostHog built-in DAU/WAU widgets based on $pageview undercount.
2026-05-11: frontend branch fix/posthog-native-pageviews restores native $pageview via SPA capture_pageview: 'history_change' and stops mirroring only page_view into PostHog. After it is merged, use $pageview for PostHog DAU/WAU/retention and keep page_view for GTM/GA4-shaped funnel work.

Which source answers which question

Question	Source
How much traffic? Where from?	GA4 or Plausible
Are our ads working? Spend / ROAS / CPA?	Meta
Which creative performs best?	Meta
Conversion rate / funnel drop-off?	PostHog
Where do users abandon checkout?	PostHog + Clarity
UX friction on a specific page?	Clarity
Quick page or source breakdown?	Plausible

Cross-tool workflows

Full funnel review GA4 (traffic + source attribution) → PostHog (drop-off at each funnel step) → Clarity (friction on pages with highest drop-off).

Campaign performance Meta (spend, ROAS, creatives) → GA4 (sessions and conversions from paid) → PostHog (checkout completion rate for paid visitors).

Checkout investigation PostHog checkout dropoff (find the step) → Clarity order/checkout friction (UX signals on that step) → PostHog replay candidates (sessions worth watching).

Variant comparison Segment GA4 and PostHog by variant. Never average across variants.