Skip to content
Syncfusion GTM Playbook

Weekly GTM review

Type
checklist
Status
draft
Owner
COE Team
Last reviewed
2026-11-07
Products
All products

Draft. Not yet reviewed. Check facts before you use it with customers.

Run every Monday per product: Claude pulls last week’s paid, analytics, search and AI-visibility data through MCP (read-only) and writes a short summary with the top 3 actions; done means the summary is posted and each action has an owner. The Claude skill weekly-gtm-review runs this checklist.

Before you start

  • MCP servers for Google Ads, Microsoft Ads (if used), GA4, Search Console, Bing Webmaster and the CRM are connected with your own read-only login. Setup and guardrails: MCP and tools.
  • Claude never changes budgets, bids, campaign status, keywords or settings during the review. If an action needs a change to spend or live campaigns, Claude proposes it as a list and a human approves it in chat first.
  • Period: last full week (Mon–Sun) vs the week before; flag anything that also differs from the 4-week average.
  • Have the latest AI answer monitoring results to hand (monthly).

Checklist

Phase 1: paid (Claude via Google Ads / Microsoft Ads MCP)

  • Spend, clicks, conversions and cost per conversion per campaign type (brand, category, competitor, migration, remarketing).
  • Qualified conversions imported from the CRM vs form fills; cost per qualified opportunity.
  • Top new search terms by spend with no conversions (negative candidates).
  • Any disapprovals, budget-limited campaigns or auto-applied changes.

Phase 2: site and goals (Claude via GA4 MCP)

  • Sessions by channel (organic, paid, YouTube, referral, AI assistants where identifiable as referrers).
  • Key events: demo requests, trial sign-ups, contact-sales, pricing-page views.
  • Landing pages with the largest drop or gain in conversions.

Phase 3: search (Claude via Search Console and Bing Webmaster MCP)

  • Clicks, impressions, average position: brand vs non-brand.
  • Top 10 queries gaining and losing impressions.
  • New coverage errors, manual actions or Core Web Vitals regressions (see setup checklist).

Phase 4: AI visibility (from the monthly log)

  • Mention and citation rate change since the last run, per engine.
  • New wrong facts or new competitor appearances.

Phase 5: decide (marketing lead, 15 minutes)

Results format

Claude prints the result in chat; save it to the team’s shared results log (location TBD). If an output directory is given (for example by pbmaintain), also write the report files defined in reports/schema/README.md there. Otherwise don’t create results files in the repo. Keep it under one screen. Use exactly this format:

## Weekly GTM review: {product slug or "all"}, week of {YYYY-MM-DD}
Run by: {name or "Claude"} | Data sources: {list of MCP servers that responded}


### Headline
{One or two sentences: what changed and why it matters.}


| Area | Metric | This week | Last week | Change | Note |
|---|---|---|---|---|---|
| Paid | Spend | | | | |
| Paid | Qualified conversions | | | | |
| Paid | Cost per qualified opportunity | | | | |
| Site | Demo + trial + contact-sales | | | | |
| Search | Non-brand clicks (GSC + Bing) | | | | |
| AI | Mention rate (last run) | | | | |


### Notable changes
- {query, campaign or page}: {what moved, likely reason}


### Top 3 actions
1. {action} | Owner: {role or name} | Due: {date}
2. ...
3. ...


### Proposed changes needing approval
- {campaign / setting}: {current} → {proposed}. (None applied until approved in chat.)

Report numbers exactly as the source returns them. If a source fails, write “no data” for its rows and name the failed source; never fill gaps with estimates.

Report keys (for the report files): Item IDs are the phase plus the item’s position in it (W1.1 … W5.2). Metric keys: ads.spend, ads.qualified_conversions, ads.cost_per_qualified_opp, ga.key_events, gsc.nonbrand_clicks, bing.nonbrand_clicks, ai.mention_rate.

Done when

  • The summary is posted in the results log and shared with the team.
  • Each of the top 3 actions has an owner and a due date.
  • Any proposed spend or campaign change is approved or rejected in chat.

If something goes wrong

  • An MCP server fails or returns partial data: note it in the summary and fix access before next week (see MCP and tools).
  • A metric moves sharply (spend spike, conversions to zero, traffic drop): alert the channel owner the same day; don’t wait for the next review.