--- name: google-analytics-weekly-traffic-report description: "Every Monday: last week's GA4 sessions, users and key events, each channel set against the week before, and the change worth a look. Use when the user asks for this, or when it runs on a schedule." --- # GA4 weekly traffic report Every Monday: last week's GA4 sessions, users and key events, each channel set against the week before, and the change worth a look. Monday morning in GA4 usually means the Reports snapshot, a date comparison and a few minutes of squinting at which line moved. This skill runs that comparison for you: last week's sessions, active users, engagement rate and key events, split by default channel group and set against the week before. Founders get one number and its direction. Marketers get the channel that explains it, so a 6% rise reads as "email did it" instead of a mystery. ## 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 - Viewer access to the GA4 property, and its numeric property ID (Admin > Property settings). - A Google Analytics MCP server or connector, or a service account with the Analytics Data API enabled, added to the property as a Viewer, with `GOOGLE_APPLICATION_CREDENTIALS` pointing at its key. If something is missing, say what and stop; do not invent numbers. ### Steps 1. **Pull both weeks in one call** (Google Analytics): Call the Analytics Data API `runReport` on `properties/` with two `dateRanges` (last Monday to Sunday, and the week before), `dimensions: [{ name: "sessionDefaultChannelGroup" }]` and the metrics `sessions`, `activeUsers`, `newUsers`, `engagementRate` and `keyEvents`. With two ranges, every row also carries a `dateRange` value, so each channel comes back once per week. 2. **Get the property totals separately** (Google Analytics): Run the same report with no dimension for the totals. Users are not additive across channels: someone who arrived by search and later by email counts once in each channel but once overall. 3. **Decide which channels moved**: Work out each metric's change in percent. A channel has moved when its sessions changed by 15% or more and by at least 200 sessions; smaller swings are ordinary week-to-week noise. 4. **Post the tile**: Sessions is the headline with 8 weeks of history. On the first run, backfill it with one call over the last 56 days and the `isoYearIsoWeek` dimension. Users, engagement rate and key events go in the glance, the channel split in the opened card, and one line names the channel behind the change. ## Route B: the browser ### What you need - Signed in to Google with the Viewer role or higher on the GA4 property; the property must be selected before the reports load. If something is missing, say what and stop; do not invent numbers. ### Steps Start at https://analytics.google.com/analytics/web/. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears. 1. **Open Traffic acquisition for the property**: The URL opens the last property you used; switch with the account picker at the top left if it is the wrong one. Go to Reports > Acquisition > Traffic acquisition and make sure the primary dimension is `Session default channel group`. 2. **Set last week against the week before**: In the date picker at the top right, choose a custom range from last Monday to last Sunday, switch on Compare, pick `Previous period (match day of week)` and click Apply. Every metric column now shows both weeks and the percentage change. 3. **Read the channel rows and the totals**: Note Sessions, Engagement rate and Key events for Organic Search, Direct, Paid Search, Referral, Email and Organic Social, and for the Totals row. A channel has moved when its sessions changed by 15% or more and by at least 200. 4. **Take users from the snapshot**: Users in the channel table overlap between channels, so open Reports > Reports snapshot with the same dates and comparison and read Active users and New users from the first card there. 5. **Name the channel behind the week**: Work out which channel accounts for most of the change in total sessions and write one line about it. Reports > Reports snapshot already shows the headline comparison, so the browser is a reasonable route for this report. Note: Sunday can still gain a few sessions for a day or two after the run; a Tuesday read is more final. The channel table shows Active users per channel too, but that column does not add up to the property total, which is why the snapshot supplies it. ## The report - **Sessions** last week and the change on the week before, with an 8-week trend. - **Active users**, **engagement rate** and **key events**, each with its change. - **Sessions by channel**: Organic Search, Direct, Paid Search, Referral, Email and Organic Social, this week and last. - One line on the channel that explains the week, for example "Email +41% after Tuesday's newsletter; the rest within a few percent". ## 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": "google-analytics-weekly-traffic-report/", "item_key": "google-analytics-weekly-traffic-report", "project": "example.com", "source": "GA4 weekly traffic report skill", "kind": "metric", "title": "Sessions up 6.5% on the week before", "metric": { "label": "Sessions", "value": 28640, "better": "higher", "period_start": "", "period_end": "" }, "glance": [ { "type": "Grid", "props": { "columns": 3 }, "children": [ { "type": "Stat", "props": { "label": "Active users", "value": 21350, "delta": 5 } }, { "type": "Stat", "props": { "label": "Engagement rate", "value": 61.2, "unit": "%", "delta": 2 } }, { "type": "Stat", "props": { "label": "Key events", "value": 1136, "delta": 8 } } ] }, { "type": "Badges", "props": { "items": [ { "label": "↑ Email +41%", "tone": "ok" }, { "label": "↑ Organic Search +10%", "tone": "ok" }, { "label": "↓ Paid Search -4%", "tone": "bad" } ] } } ], "widget": [ { "type": "Table", "props": { "columns": [ "Channel", "Sessions", "Week before", "Change" ], "align": [ "left", "right", "right", "right" ], "rows": [ [ "Organic Search", 12480, 11390, "+10%" ], [ "Direct", 6210, 6050, "+3%" ], [ "Paid Search", 4320, 4500, "-4%" ], [ "Referral", 2150, 2090, "+3%" ], [ "Email", 1880, 1330, "+41%" ], [ "Organic Social", 1600, 1520, "+5%" ], [ "Total", 28640, 26880, "+6.5%" ] ] } }, { "type": "Callout", "props": { "text": "Organic Search and Email account for 1,640 of the 1,760 extra sessions. Email jumped 41% after Tuesday's newsletter.", "tone": "info" } } ], "size": "md" } JSON ``` Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report. Set `period_start` and `period_end` to the window the number covers: the tile names it from them ("Last 7 days", "Today", "This month"). For a count right now (open items, a balance), send `"period_label": ""` so no period shows; for another name, send it as `period_label` (at most 24 characters). ## Run it on a schedule Suggested: Mondays at 8:00 (`0 8 * * 1`). From cron, with the agent signed in on that machine: ```bash 0 8 * * 1 claude -p "Run the google-analytics-weekly-traffic-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/google-analytics-weekly-traffic-report#schedule Source: https://briefkin.dev/skills/google-analytics-weekly-traffic-report