--- name: mailchimp-campaign-report description: "Every Tuesday: last week's Mailchimp campaigns with opens, clicks, unsubscribes and top links, set against your audience average. Use when the user asks for this, or when it runs on a schedule." --- # Mailchimp campaign report Every Tuesday: last week's Mailchimp campaigns with opens, clicks, unsubscribes and top links, set against your audience average. Mailchimp shows each campaign's report on its own page, and the number everyone reads first, the open rate, has been inflated by Apple Mail Privacy Protection since 2021. Whether a newsletter did well depends on comparing it with your usual results, which takes a spreadsheet nobody keeps up to date. The agent pulls every campaign sent in the last week, uses Mailchimp's open rate with machine opens excluded, and puts each next to your audience's average. It lists the links people actually clicked and calls out unsubscribe spikes, which are the early sign that a send went to the wrong segment. ## Two ways to get the data Use the API or MCP route when you have an MCP server, a CLI or an API token for the app. Otherwise, if you can drive a browser the user is signed in to (Claude in Chrome, a Playwright or Chrome DevTools MCP server), use the browser route. Say which one you used. ## Route A: API or MCP server ### What you need - A Mailchimp user with access to reports, and an API key in `MAILCHIMP_API_KEY`. The suffix of the key (`-us21`) is the data center, which sets the API host (`https://us21.api.mailchimp.com/3.0`). If something is missing, say what and stop; do not invent numbers. ### Steps 1. **List last week's campaigns** (Mailchimp): `GET /3.0/reports?since_send_time=<8 days ago>&before_send_time=<24 hours ago>&count=100`, so every campaign has had at least a day of opens. Keep `campaign_title`, `subject_line`, `emails_sent`, `opens`, `clicks`, `unsubscribed` and `bounces`. 2. **Use the honest open rate**: Report `opens.proxy_excluded_open_rate` (Apple machine opens removed) next to `clicks.click_rate`. Unsubscribe rate is `unsubscribed` over `emails_sent`. 3. **Compare with your average**: Use each report's `list_stats` (`proxy_excluded_open_rate`, `click_rate`, `unsub_rate`) as the audience average. Flag a campaign 20% or more above or below average on clicks, and any unsubscribe rate over twice the average. 4. **Find the top links** (Mailchimp): `GET /3.0/reports/{campaign_id}/click-details?sort_field=unique_clicks&count=5` for each campaign; keep `url`, `unique_clicks` and `click_percentage`. 5. **Write the report**: One row per campaign with sends, open rate, click rate and unsubscribes against average, the top 3 links overall, and one line on what the best campaign did differently. ## Route B: the browser ### What you need - Signed in to Mailchimp with a user at Viewer level or above on the account. If something is missing, say what and stop; do not invent numbers. ### Steps Start at https://admin.mailchimp.com/reports/. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears. 1. **List last week's campaigns**: Open Mailchimp's reports list (**Analytics > Reports**, or **Campaigns > Reports** in older navigation), set the audience if you have more than one, and keep the campaigns sent between 8 days ago and yesterday, so each has had a day of opens. Note `Sent`, `Opens` and `Clicks` from each row. 2. **Open each campaign report**: Click **View report**. On the **Overview** read the open rate, click rate, unsubscribes and bounces; the **List average** shown under each rate is your audience average. Where the report separates Apple machine opens, use that open rate; otherwise lean on clicks. 3. **Read the top links**: Scroll to **Top links clicked**, or open the **Links** tab, and note the first 5 URLs with their unique clicks for each campaign. 4. **Compare and write**: One row per campaign with sends, open rate, click rate and unsubscribe rate against the list average. Flag campaigns 20% or more above or below average on clicks, any unsubscribe rate over twice the average, and hard bounces. Add the top 3 links overall and what the best campaign did differently. Note: The headline open rate in a report page may still count Apple Mail Privacy Protection opens; the API field `proxy_excluded_open_rate` removes them. The browser route is fine for two or three sends a week; with daily sends, the API route is faster. Mailchimp has renamed its reporting menu more than once; the reports URL above has stayed the same. ## The report - **Per campaign**: sends, open rate without machine opens, click rate and unsubscribe rate, each against your audience average. - **Top links** across last week's sends. - **Flags**: clicks well above or below average, unsubscribe spikes, hard bounces. - One line on the campaign worth copying. ## Post it to Briefkin When `BRIEFKIN_TOKEN` is set, or `~/.config/briefkin/env` exists (the installer from https://briefkin.dev/docs writes it), also send the result to the user's Briefkin board as one card. Keep the `item_key` the same on every run so the card updates in place, use a new `event_id` per run, and put today's real numbers in place of the example ones. If the request fails, say so once and carry on. ```bash [ -n "${BRIEFKIN_TOKEN:-}" ] || . ~/.config/briefkin/env curl -sS -X POST "${BRIEFKIN_URL:-https://briefkin.dev}/v1/events" \ -H "authorization: Bearer $BRIEFKIN_TOKEN" \ -H "content-type: application/json" \ -d @- <<'JSON' { "v": 1, "event_id": "mailchimp-campaign-report/", "item_key": "mailchimp-campaign-report", "project": "Acme", "source": "Mailchimp campaign report skill", "kind": "update", "title": "2 campaigns last week, clicks above your average", "summary": "The October product update got a **4.1% click rate** against your usual 2.6%. The webinar invite's unsubscribe rate was **0.62%**, three times normal.", "widget": [ { "type": "Table", "props": { "columns": [ "Campaign", "Sent", "Opens*", "Clicks", "Unsubs" ], "align": [ "left", "right", "right", "right", "right" ], "rows": [ [ "October product update", 18420, "34.8%", "4.1%", "0.14%" ], [ "Webinar invite: pricing teardown", 9310, "29.5%", "2.2%", "0.62%" ], [ "Audience average", "", "31.2%", "2.6%", "0.20%" ] ] } }, { "type": "Table", "props": { "columns": [ "Top link", "Unique clicks" ], "align": [ "left", "right" ], "rows": [ [ "/changelog/recurring-invoices", 412 ], [ "/webinars/pricing-teardown", 168 ], [ "/templates/invoice", 97 ] ] } }, { "type": "Text", "props": { "text": "* Open rate with Apple machine opens excluded.", "variant": "muted" } } ] } JSON ``` Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report. ## Run it on a schedule Suggested: Tuesdays at 9:00 (`0 9 * * 2`). From cron, with the agent signed in on that machine: ```bash 0 9 * * 2 claude -p "Run the mailchimp-campaign-report skill" # Codex: codex exec, Gemini CLI: gemini -p, Copilot CLI: copilot -p, OpenCode: opencode run ``` Or use the agent's own scheduler (Claude Code routines, Codex automations, Claude or ChatGPT scheduled tasks): https://briefkin.dev/skills/mailchimp-campaign-report#schedule Source: https://briefkin.dev/skills/mailchimp-campaign-report