Données
MCP Kit for Marketers
3 skills included
New- 5 connexions prêtes
- 30 requêtes de reporting
- Sécurité des accès
39€incl. VAT, one-time payment
claude skills / MCP Kit for Marketers / marketing-reporting-queries / en français
The monthly report in one question: what changed, why, and what to do next.
orders of magnitude, depending on your volume
the problem
input, output
What you give Claude: the period, the active connections, your definitions and the month's context
Company: Halden Tile Co. (online tile shop) Period: October 2026, compared with September 2026 and October 2025 Connected: Search Console, GA4, Google Ads, Shopify, Notion Main conversion: purchase, net sales excluding tax and shipping, EUR Brand terms: halden, halden tiles, haldon Context: target ROAS lowered on 3 Oct, newsletter with code AUTUMN10 on 8 Oct, add to cart event renamed on 20 Oct Readers: the owner and the marketing team
$ claude "Run the October 2026 marketing report for Halden Tile Co. with the core questions, and explain what changed."
Marketing report, monthly report excerptexample
Marketing report: Halden Tile Co., October 2026
Data up to: 2026-10-31 in all tools (Search Console final) | Compared with: September 2026 (30 days), October 2025 | EUR, net sales excl. tax and shipping
IN FIVE LINES
1. Net sales 48,300, up 9% on September and up 14% on October 2025. Orders 412, up 6%.
2. Organic clicks 6,120, down 4%. Brand stable, the loss comes from two bathroom guide pages.
3. Ads cost 7,900, 128 paid-search orders in Shopify. Cost per Shopify order 62, against 66.
4. Problem: "Outdoor tiles" loses 41% of impression share to rank on its best-margin products.
5. Opportunity: 14 non-brand queries in positions 8 to 20 on the large format page.
October has 31 days and September 30. Per day, net sales grew about 5%.
RECONCILIATION
Paid search orders: Shopify 128 | GA4 119 | Google Ads 151
Ads counts view-through and cross-device. Ratio Ads to Shopify 1.18, September 1.16: stable, no tracking change.
WHAT CHANGED AND WHY
Non-brand clicks -7% | /guides/bathroom-waterproofing -190 clicks, position 4.1 to 6.8 after 12 Oct | FACT for the drop, cause not explained yet
Ads cost +4%, conversions +9% | target ROAS lowered on 3 Oct | INFERENCE
AUTUMN10 used on 96 orders, 60 planned | referral sessions from one coupon site | HYPOTHESIS
ACTIONS
1. Update the waterproofing guide | Content lead | watch position weekly (question 6)
2. Fix mobile speed on the outdoor landing page | Web developer | watch IS lost to rank (question 17)
NOT CHECKED
Funnel steps: add to cart renamed on 20 Oct, months not comparable.
the method
Claude asks each tool for its latest available date and flags at the top any incomplete month or data not yet final.
The questions marked core run on every connected tool, copied word for word, and each answer is recorded with its period.
Shopify orders, GA4 purchases and Google Ads conversions are set side by side. The gap is explained, or flagged as a possible tracking problem.
Each change is located and labeled fact, inference or hypothesis, then 3 to 5 actions with an owner and the metric to watch.
demo in French / click, try it
An interactive demo of the skill. It is in French, the pack works in any language: ask Claude to answer in English.
Shopify
the film / 44 seconds, with sound
the pack film
the guardrails
Ideally, yes, and the mcp-connections-setup skill in the same pack handles it. Without a connection, the missing tool goes under "Not checked" or is filled in from your exports.
That is normal: attribution, consent, time zones and refunds differ. The report shows the gap, explains it and uses your store or CRM as the reference.
Yes, Claude drafts the page text. Creating or editing a page is a write action, so it asks for your approval every time.
Yes. The same questions work for a week, with the same rule: compare periods of the same length.
the pack
Données
3 skills included
New39€incl. VAT, one-time payment