Illustration for Analytics & reporting marketing skills
Analytics & reporting skill · v0.1.0

Closed-loop measurement skill

Judge every marketing change by what it did in the data: baseline, change log, follow-up read and keep or revert.

Free · MIT licence · Copy needs a free sign-in

Category
Analytics & reporting
Works with
Claude, Codex, Cursor, any agent
Format
SKILL.md · 1,392 words
Version
v0.1.0
Price
Free · MIT

The Closed-loop measurement skill file

cmo-closed-loop-measurement/SKILL.md1,392 words · MIT
---
name: cmo-closed-loop-measurement
description: Judge every marketing change by what it did in the data: baseline, change log, follow-up read and keep or revert.
---

A change is not finished when it ships; it is finished when someone checks whether it worked and the playbook learns from the answer. This skill keeps a change log, fixes a baseline before each change, reads the data again after a set window, and decides: keep, keep testing, revert, or unproven. In CMO.so it starts from what Maggie actually did and how each tracked article performs in Google Search.

When to use

  • After Maggie publishes or rewrites articles, or after any page, campaign or offer change.
  • "Did it work?", "before and after", "keep or roll back?", "which changes moved traffic?".
  • Building a habit of monthly readbacks so future work follows evidence.
  • Not for: designing a controlled split test up front (cmo-ab-testing), watching for ranking or traffic drops (cmo-seo-monitoring), channel-level ROI and funnels (cmo-campaign-analytics) or rewriting the articles themselves (cmo-content-refresh).
Sign in to see, copy and download the whole skillIt's free. Copying also upvotes and saves it, so you'll find it again under Saved.
Preview: the workflow, output, example, quality bar and guardrails unlock when you sign in.
Everyday usage

How people use Closed-loop measurement day to day

Ask in plain words. The agent picks this skill, asks for anything it needs and hands back the result. Words in a highlight are yours to change.

Most common · also on the Analytics & reporting page

Judge a change by its results

A baseline, a change log and a follow-up that shows whether the change worked.

Ask your agentDid last month's homepage change improve signups?
  • A change log with baseline and readback windows
  • A readback per change: baseline vs after, delta, caveats, verdict
  • Updated rules for the playbook (what to do more or less of)

More things to ask

  • Did that change actually work?
  • Check the articles Maggie rewrote last month
  • Set up a before-and-after for this change
  • Which of our changes moved traffic?
  • Should we keep or roll back this update?
  • Build a readback for our marketing experiments
How it works

From input to output

  1. 1You give it
    • The change: what, where (page, post, campaign), when, and the metric it should move
    • From CMO.so: `maggie_actions_list`, `content_performance_list`, `content_asset_get`, `gsc_performance_get`
    • Other analytics the user supplies (GA4, ads, CRM) for changes outside search
    • Known confounders in the window (seasonality, launches, algorithm updates, other changes)
  2. 2It works through the method
    • Asks what's missing, one question at a time
    • Follows the workflow step by step
    • Checks the result against its quality bar and guardrails
  3. 3You get
    • A change log with baseline and readback windows
    • A readback per change: baseline vs after, delta, caveats, verdict
    • Updated rules for the playbook (what to do more or less of)
Why a skill

The skill against the alternatives

Do it yourselfHire helpClosed-loop measurement skill
CostYour own timeAgency or freelancer feesFree, MIT licence
Time to a first resultHours, every timeDays to weeks, plus revisionsMinutes, in your agent
ConsistencyDepends on remembering every stepAs long as the retainer lastsSame steps and checks every run
The methodIn your headTheir process, often hiddenWritten down: read it, change it
Who does the workYou, from scratchSomeone learning your businessYour agent, following the skill
FAQ

Questions about the Closed-loop measurement skill

What it does, what it needs and how to use it with your agent.

Still not sure?Try it free on your own website and see the first posts before you decide.
Start free
What is the Closed-loop measurement skill?

Judge every marketing change by what it did in the data: baseline, change log, follow-up read and keep or revert.

It's one of cmo.so's open Analytics & reporting skills: a SKILL.md file that tells an AI agent when to use it, what to ask you first, the steps to follow, the checks to pass and what to hand back, so the job is done the same careful way every time.

When should I use it?

Reach for it when you'd otherwise say something like “Did that change actually work?”, “Check the articles Maggie rewrote last month”, “Set up a before-and-after for this change”. Your agent reads the skill's description and picks it for requests like these, so you don't need to name it.

If a request is close but not quite right, the skill says so and points to the related skill that fits better, instead of forcing a poor answer.

What do I need to give it?

It works best with the change: what, where (page, post, campaign), when, and the metric it should move; from CMO.so: `maggie_actions_list`, `content_performance_list`, `content_asset_get`, `gsc_performance_get`; other analytics the user supplies (GA4, ads, CRM) for changes outside search; and known confounders in the window (seasonality, launches, algorithm updates, other changes).

You don't have to prepare everything up front. The skill asks for missing pieces one at a time and says plainly when it's working from an assumption instead of your data.

What do I get back?

You get a change log with baseline and readback windows; a readback per change: baseline vs after, delta, caveats, verdict; and updated rules for the playbook (what to do more or less of).

Every result comes from a written method with a quality bar, so numbers name their source, claims are checked and the output is ready to use or share rather than a rough draft.

Is it free?

Yes. The skill is open source under the MIT licence, so you can use it, change it and share it, including for client work.

Seeing, copying and downloading the whole file on cmo.so needs a free sign-in. Copying also upvotes and saves it, so you can find it again under Saved on the skills page.

Which AI agents can use it?

Claude Code installs it from the cmo.so marketplace. Any agent that reads SKILL.md folders, such as Codex or Cursor, can use it once the folder is in its skills directory.

You can also paste the copied skill into ChatGPT, Claude or any assistant at the start of a conversation and ask it to follow the skill.

How do I install it?

In Claude Code, run /plugin marketplace add AI-Marketing-OS-LTD/cmo-so-skills, then /plugin install cmo-so-analytics@cmo-so. That installs this skill together with the rest of the Analytics & reporting group.

For other agents, sign in, press Copy or Download above and put the file in your agent's skills folder as SKILL.md inside a folder named after the skill.

Do I need a cmo.so account to use it?

No. The skill works on its own with whatever your agent can already see and do: your files, the pages you share and the data you paste in.

With cmo.so connected it can go further, using tools such as maggie_actions_list, content_performance_list, gsc_performance_get, content_asset_get through the MCP server. Every change is previewed and confirmed before it happens.

Which skills work well with it?

It pairs well with A/B testing & experiments; and SEO change monitoring. The agent can hand over from one to the next in the same conversation, for example research first, then the plan, then the writing.

Each of them has its own page with what it needs and what it returns, linked below.

Can I change it to fit my business?

Yes. It's a plain Markdown file, so you can edit the steps, add your own checks and examples, and keep your version in your own repo. The MIT licence allows it.

If your change would help others too, suggest it on the skills page or open a pull request on GitHub, and it may be folded into the library for everyone.

Where does the method come from?

It's written and maintained by cmo.so in the public cmo-so-skills repo, alongside the rest of the library.

Improvements ship to the repo and appear here as soon as they're synced, with the version shown at the top of the page.

Is my data sent to cmo.so when the agent uses it?

No. The skill only guides your own agent; your files, documents and data stay wherever your agent and its tools keep them.

cmo.so only sees what you connect to cmo.so. Your votes and saves on this page are stored against your cmo.so sign-in.

Ready when you are!

Rather have Maggie do it for you?

Enter your website. Maggie uses the same playbooks to plan, write and publish your marketing every day.

  • Free plan, no card needed
  • You decide what goes out
  • Cancel any time

Start with your website

Free to start. No card needed.

Welcome back.

Sign in to cmo.so.

or
At least 8 characters.
New to cmo.so? Have an account? Prefer a password?

By continuing, you agree to our Terms of Service and Privacy Policy.