--- name: google-analytics-key-events-by-channel description: "Every Monday, key events and session key event rate for each GA4 default channel group, so you see which channels convert and which only bring visits. Use when the user asks for this, or when it runs on a schedule." --- # GA4 conversions by channel Every Monday, key events and session key event rate for each GA4 default channel group, so you see which channels convert and which only bring visits. Sessions by channel tell you who showed up. Whether they did anything is a separate question, and since GA4 renamed conversions to key events in 2024 the answer sits under a new name in Traffic acquisition, next to a rate column that is easy to misread. The agent pulls both for each default channel group every Monday. What a marketer notices first is usually a mismatch: a channel with a fifth of the sessions and a tenth of the key events, or email quietly converting at twice the site rate. ## 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 - The Viewer role on the GA4 property; reading key event settings through the Admin API needs no more than that. - The Analytics Data and Admin APIs, reached through an Analytics MCP server or connector, or a service account with the `analytics.readonly` scope. If something is missing, say what and stop; do not invent numbers. ### Steps 1. **List what counts as a key event** (Google Analytics): Read the property's key events once with the Admin API (`properties.keyEvents.list`) so the report can name them, for example `sign_up`, `generate_lead` and `purchase`. 2. **Pull key events by channel** (Google Analytics): Call `runReport` for the last 7 full days and the 7 before, with `dimensions: [{ name: "sessionDefaultChannelGroup" }]` and the metrics `sessions`, `keyEvents` and `sessionKeyEventRate`. To follow one conversion only, use `keyEvents:sign_up` and `sessionKeyEventRate:sign_up` instead. 3. **Compare shares and rates**: Set each channel's share of sessions next to its share of key events, and compare its rate with last week. Flag channels whose rate moved 20% or more and that had at least 30 key events in either week. 4. **Post the tile**: Total key events is the headline with 8 weeks of history, the glance shows key events per channel, and the opened card has the full table with a line on the biggest mismatch. ## Route B: the browser ### What you need - Signed in to Google with the Viewer role on the GA4 property; the Key events admin page is readable with Viewer access. 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. **See which events count**: Pick the property, open Admin > Data display > Key events and note the names marked as key events, for example `sign_up`, `generate_lead`, `purchase`. 2. **Open Traffic acquisition with the comparison**: Go to Reports > Acquisition > Traffic acquisition, set the date range to `Last 7 days`, switch on Compare with `Previous period (match day of week)` and Apply. The primary dimension should be `Session default channel group`. 3. **Read key events and the rate per channel**: The table already has Sessions, Key events and Session key event rate columns. The dropdown in the Key events column header switches from All events to one event, for example `sign_up`, and the rate column follows it. 4. **Compare shares and rate changes**: For each channel, set its share of total sessions next to its share of total key events (the Totals row gives both). Flag channels whose rate moved 20% or more and that had at least 30 key events in either week. 5. **Write it up**: Total key events and the change first, then the channel table, then the biggest mismatch and one suggestion on where to shift effort. Note: Session key event rate is the share of sessions with at least one key event, so it is always lower than key events divided by sessions. The comparison columns the report draws make this a good fit for the browser route. ## The report - **Key events** last week, with an 8-week trend and the change on the week before. - **By channel**: sessions, key events and session key event rate, with the change in rate. - **Mismatches**: channels with a much bigger share of sessions than of key events, or the reverse. - One line on where to shift effort, for example "Email converts at 7.2%, twice the site rate: send the launch sequence to more of the list". ## 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-key-events-by-channel/", "item_key": "google-analytics-key-events-by-channel", "project": "example.com", "source": "GA4 conversions by channel skill", "kind": "metric", "title": "Key events up 8% on last week", "metric": { "label": "Key events", "value": 1136, "better": "higher", "period_start": "", "period_end": "" }, "glance": [ { "type": "Grid", "props": { "columns": 2 }, "children": [ { "type": "Stat", "props": { "label": "Session key event rate", "value": 3.6, "unit": "%", "delta": 4 } }, { "type": "Stat", "props": { "label": "Best channel", "value": "Email, 7.2%" } } ] }, { "type": "Bars", "props": { "items": [ { "label": "Organic Search", "value": 402 }, { "label": "Paid Search", "value": 268 }, { "label": "Direct", "value": 236 }, { "label": "Email", "value": 151 }, { "label": "Referral", "value": 58 }, { "label": "Organic Social", "value": 21 } ] } } ], "widget": [ { "type": "Table", "props": { "columns": [ "Channel", "Sessions", "Key events", "Rate", "Rate change" ], "align": [ "left", "right", "right", "right", "right" ], "rows": [ [ "Organic Search", 12480, 402, "2.9%", "+4%" ], [ "Paid Search", 4320, 268, "5.6%", "-3%" ], [ "Direct", 6210, 236, "3.4%", "+2%" ], [ "Email", 1880, 151, "7.2%", "+18%" ], [ "Referral", 2150, 58, "2.4%", "-9%" ], [ "Organic Social", 1600, 21, "1.2%", "-25%" ], [ "All channels", 28640, 1136, "3.6%", "+4%" ] ] } }, { "type": "Callout", "props": { "text": "Email brings 7% of sessions and 13% of key events. Organic Social brings 6% of sessions and 2% of key events.", "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 9:30 (`30 9 * * 1`). From cron, with the agent signed in on that machine: ```bash 30 9 * * 1 claude -p "Run the google-analytics-key-events-by-channel 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-key-events-by-channel#schedule Source: https://briefkin.dev/skills/google-analytics-key-events-by-channel