google-growth-journey
Skill0 starsChanged 49 days agoMITView on GitHub
End-to-end playbook for taking a website from "no measurement" to a live, correctly-measured Google Ads Performance Max campaign, with GA4, Search Console and Ads all linked. Use when the user wants to "set up Google Ads", "run ads for my site", "connect GA4 / Search Console / Ads", or audit an existing Google marketing setup. Routes to the other gskill skills in the right order.
In this repository7 other files it gives its agents.
- ad-assetsSkill
- conversion-trackingSkill
- google-ui-automationSkill
- growth-auditSkill
- pmax-campaignSkill
- search-console-seoSkill
- tag-managerSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 25
If the work is done through a browser (Google's UIs have no practical API for most of this), read google-ui-automation first — it lists the …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Google growth journey
This is the map. Each stage has its own skill; do them in this order, because every later stage depends on the earlier one being correct. Advertising before measurement works is the most expensive mistake in this whole field: the bidding algorithm learns from whatever conversions it sees, and if it sees none (or the wrong ones) it optimises toward noise with real money.
| # | Stage | Skill | Done when |
| 1 | Audit what exists | growth-audit | You know which IDs are live on the site and which accounts own them |
| 2 | Site hygiene for search | search-console-seo | Sitemap read OK, private pages noindex, canonical clean, property verified |
| 3 | Measurement | conversion-tracking | A real conversion shows up in GA4 DebugView/Realtime with value + transaction id |
| 4 | Tag management (only if needed) | tag-manager | One loader, no duplicate config calls, no PII leaving the page |
| 5 | Creative | ad-assets | Images in 3 ratios, logo, 3 videos (16:9, 1:1, 9:16), copy within limits |
| 6 | Campaign | pmax-campaign | Draft reviewed on the summary step; the user presses Publish |
| 7 | Link everything | growth-audit (link section) | GA4↔Ads, GA4↔Search Console, Ads↔Search Console all "linked" |
| 8 | Verify after 48 h | growth-audit (post-launch) | Conversion actions move from "pending" to recording |
If the work is done through a browser (Google's UIs have no practical API for most of
this), read google-ui-automation first — it lists the traps that cost hours.
Ground rules (apply to every stage)
- Verify with a real browser, not curl. Sites behind Cloudflare/bot protection answer
curlwith 403 or a challenge page. A missing tag incurloutput proves nothing. Use a browser user-agent at minimum, ideally a real browser anddataLayerinspection. - Never paste a claim you have not checked. "The site has no Google tag", "the campaign targets Vietnam", "the link exists" — each of these was wrong at least once in the run this playbook came from, because a summary screen or a blocked request lied. Re-read the source of truth (the settings step, the live page, the linked-products list).
- The human does the irreversible and the identity-bound steps: publishing a campaign, paying/topping up, identity verification, accepting terms, deleting campaigns (only after an explicit request). Prepare everything up to that button and stop.
- One domain, one account, one campaign. Duplicating a campaign to point at a second domain inside the same Ads account drags shared assets (sitelinks, conversion goals) along. Give each domain its own Ads account and its own GA4 property.
- No personal data to Google. URL query strings and page titles often contain names,
birth dates, emails. Whitelist what you send (see
conversion-tracking).
What to ask the user up front
- Which domain, and which Ads account / GA4 property belong to it (account IDs look like
123-456-7890; GA4 measurement IDsG-XXXXXXX; Ads tag IDsAW-123456789). - What counts as a conversion with money attached (purchase, top-up, paid signup) and what is a softer signal (signup, key page view). Only the first kind should drive bidding early.
- Daily budget and target country/language.
- Whether they want to publish themselves (default: yes).
Deliverable at the end
A short report listing: live IDs on the site, each conversion action and its status, each product link and its status, campaign settings as read back from the settings step (not only the summary), and the list of things only the user can do (top up, verify identity, publish).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.