APIAnalytics

Google Analytics 4 API. Channels, pages, audience and conversions for your own dashboard.

You connect each site's Google Analytics 4 property in MarketingTracer, and we keep the connection running. With one API key, you pull site totals, daily trends, channels, pages, audience, conversions and referrers for every client site into your own dashboard, report or AI assistant. The REST API and the MCP server are in every plan at no extra cost.

Start free trial Endpoint reference

Totals and daily

Site totals with period comparison, plus daily sessions, engaged sessions, users and conversions.

Report breakdowns

Channels, sources, landing pages, viewed pages, audience, events, e-commerce products and external referrers.

Shared filters

Pass channel, source, landing page, folder, country and device across reports to drill down step by step.

In every plan

One personal API key for REST and MCP, with 60 calls a minute across all your sites.

Google Analytics 4 endpoints in REST and the MCP server

Send a POST with a JSON body or a GET with query parameters to /rest/v1/. Your personal API key goes in the header as a Bearer token. Pass the site_id from list_sites or copy it from the site's address in the app.

Every endpoint takes a date range as from and to or as days (default 28). The range ends yesterday because today's Google Analytics 4 figures are incomplete. Set compare to previous_period or previous_year (52 weeks back, so weekdays match). Filter by any combination of channel, source_medium, landing_page, path_prefix, country and device.

Endpoint	MCP tool	What it returns
`analytics.summary`	`get_analytics_summary`	Site totals for sessions, engaged sessions, engagement rate, users, new users, views, views per session, average session duration, conversions and revenue with period comparison, plus a daily series for both windows.
`analytics.channels`	`get_analytics_channels`	Sessions, engagement, users, views, conversions and revenue per GA4 channel group (Organic Search, Direct, Referral, AI Assistant and more), share of sessions and previous-period figures.
`analytics.sources`	`get_analytics_sources`	The summary metrics per source and medium (such as google / organic or chatgpt.com / referral), share of sessions and previous-period figures.
`analytics.landing_pages`	`get_analytics_landing_pages`	The summary metrics per entry page path and full URL, share of sessions and previous-period figures.
`analytics.pages`	`get_analytics_pages`	Page views, sessions, users, engagement rate and average session duration per viewed page path and URL, plus share of views and previous-period figures.
`analytics.audience`	`get_analytics_audience`	The summary metrics by country, city, device, browser, language or new_vs_returning, with share of sessions and previous-period figures.
`analytics.events`	`get_analytics_events`	Event count, key event conversions, users and revenue per GA4 event name, alongside the previous period.
`analytics.products`	`get_analytics_products`	E-commerce item revenue, items purchased and items viewed per product name, alongside the previous period.
`analytics.referrers`	`get_analytics_referrers`	External referring URLs and domains paired with the landing page on your site, sessions, engaged sessions, views, previous-period figures and a domain_only flag.

In Claude Code or Cursor, the get_analytics_* tools return the same reports. The MCP server also includes the free-form get_analytics tool (MCP only). Use it to request any GA4 dimension and metric, filter paths by regular expression with path_regex, or roll pages up into URL folders with group_by_section.

What a call to analytics.summary returns

Pass site_id and optional period or filter arguments. By default, days is 28 and compare is previous_period.

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.summary \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","days":28}'

{
  "data": {
    "compare": "previous_period",
    "sessions": {"value": 1150, "previous": 1031, "change": 119, "change_pct": 11.5},
    "engaged_sessions": {"value": 752, "previous": 703, "change": 49, "change_pct": 7},
    "engagement_rate_pct": {"value": 65.4, "previous": 68.2, "change": -2.8, "change_pct": -4.1},
    "users": {"value": 727, "previous": 674, "change": 53, "change_pct": 7.9},
    "new_users": {"value": 669, "previous": 622, "change": 47, "change_pct": 7.6},
    "views": {"value": 2668, "previous": 2379, "change": 289, "change_pct": 12.1},
    "views_per_session": {"value": 2.32, "previous": 2.31, "change": 0.01, "change_pct": 0.4},
    "avg_session_duration_seconds": {"value": 166.2, "previous": 178.4, "change": -12.2, "change_pct": -6.8},
    "conversions": {"value": 291, "previous": 205, "change": 86, "change_pct": 42},
    "revenue_eur": {"value": 0, "previous": 0, "change": 0, "change_pct": null},
    "daily": [
      {
        "date": "2026-10-06",
        "sessions": 37,
        "engaged_sessions": 26,
        "users": 33,
        "conversions": 6,
        "previous_date": "2026-09-08",
        "previous_sessions": 40,
        "previous_engaged_sessions": 27,
        "previous_users": 34,
        "previous_conversions": 7
      },
      {
        "date": "2026-10-07",
        "sessions": 52,
        "engaged_sessions": 32,
        "users": 37,
        "conversions": 14,
        "previous_date": "2026-09-09",
        "previous_sessions": 37,
        "previous_engaged_sessions": 29,
        "previous_users": 33,
        "previous_conversions": 12
      },
      …
    ]
  },
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Every total carries value, previous, change and change_pct (for engagement_rate_pct, change is in percentage points). Each item in daily pairs a date from the current window with previous_date from the comparison window, so a two-line chart plots straight from the array. When sessions is 0, engagement_rate_pct, views_per_session and avg_session_duration_seconds come back as null. A previous value of 0 also gives change_pct: null, as on revenue_eur above. If Google Analytics 4 is not connected for the site, the call returns HTTP 409 with error code not_connected.

Drill from channels to sources to landing pages with the same filters

A Monday report shows conversions up in Organic Search while mobile engagement dipped. Pass channel: "Organic Search" to analytics.sources to separate google / organic from other search engines. Next, call analytics.landing_pages with source_medium: "google / organic", path_prefix: "/rental/" and device: "mobile".

The session filters (channel, source_medium, landing_page, path_prefix, country and device) work together across analytics.summary and the session breakdowns. Every row includes its previous object (for the preceding window or previous_year on seasonal shops), so your dashboard renders the comparison from one request.

Which mobile landing pages under /rental/ grew in google / organic this month?

get_analytics_landing_pages

source_mediumpath_prefixdevicesessionsconversionsprevious

See which external pages and AI assistants send visitors, and where they land

When a client gets mentioned on a forum, a partner blog or in an AI assistant, analytics.referrers pairs each referrer_url and referrer_domain with the landing_page on the client's site, plus sessions, engaged_sessions and views for both periods. Search engines and self-referrals are already filtered out.

Combine it with source_medium: "chatgpt.com / referral" to inspect a single source. Many browsers trim the referrer header to the origin. When that happens, referrer_url holds https://domain and domain_only is true, so your report can tell a domain-level referrer apart from a full article URL.

Put audience, key events and e-commerce product revenue on one client scorecard

Call analytics.audience with by set to country, city, device, browser, language or new_vs_returning. Each breakdown shows who visits the site and how their engagement compares with the previous period.

Next to audience, analytics.events lists every GA4 event name with total count, conversions, users and revenue_eur. For webshops that send item data to GA4, analytics.products adds revenue_eur, purchased and viewed per product.

Endpoints, arguments and field descriptions

Every endpoint is a POST with a JSON body; reading endpoints also take a GET with the same arguments in the query string. Send your API key as Authorization: Bearer YOUR_API_KEY. Every answer is {"data": …, "meta": {…}}: meta.site_id always, meta.next_cursor and meta.total for a list, meta.period where the endpoint covers a period. An error is {"error": {"code", "message"}} with the HTTP status: 400 invalid_argument, unknown_argument or missing_argument; 401 not_authenticated; 403 no_access or tool_off; 404 not_found; 409 not_connected (a connection the endpoint needs, such as Google Analytics 4, is missing); 429 rate_limited; 502 google_error (Google returned an error). The machine-readable version is openapi.json.

analytics.summary

POST /rest/v1/analytics.summary · MCP tool get_analytics_summary

Returns the Google Analytics 4 totals of one site for the period against the previous period: sessions, engaged sessions, engagement rate, users, new users, page views, views per session, average session duration, conversions (key events) and revenue, each as value, previous, change and change in percent; plus per day the sessions, engaged sessions, users and conversions of both periods. The period runs up to yesterday at the latest. The filters narrow everything down (all at once); meta.period gives the period and the previous period. Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`channel`	string	Only sessions from this channel, as Google Analytics names it (Organic Search, Direct, Referral, Organic Social, Paid Search, Email, AI Assistant, …).
`source_medium`	string	Only sessions from this source / medium, exactly as in analytics.sources (google / organic).
`landing_page`	string	Only sessions that started on this page: its path, exactly as in analytics.landing_pages (/blog/post).
`path_prefix`	string	Only sessions that started on a page whose path begins with this (/blog/).
`country`	string	Only sessions from this country, in English as in analytics.audience (Netherlands).
`device`	one of `desktop`, `mobile`, `tablet`, `smart tv`	Only sessions on this kind of device.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.summary \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","days":28}'

Example response · A real answer, anonymised; daily shortened to two days (a 28-day period has 28).

{
  "data": {
    "compare": "previous_period",
    "sessions": {"value": 1150, "previous": 1031, "change": 119, "change_pct": 11.5},
    "engaged_sessions": {"value": 752, "previous": 703, "change": 49, "change_pct": 7},
    "engagement_rate_pct": {"value": 65.4, "previous": 68.2, "change": -2.8, "change_pct": -4.1},
    "users": {"value": 727, "previous": 674, "change": 53, "change_pct": 7.9},
    "new_users": {"value": 669, "previous": 622, "change": 47, "change_pct": 7.6},
    "views": {"value": 2668, "previous": 2379, "change": 289, "change_pct": 12.1},
    "views_per_session": {"value": 2.32, "previous": 2.31, "change": 0.01, "change_pct": 0.4},
    "avg_session_duration_seconds": {"value": 166.2, "previous": 178.4, "change": -12.2, "change_pct": -6.8},
    "conversions": {"value": 291, "previous": 205, "change": 86, "change_pct": 42},
    "revenue_eur": {"value": 0, "previous": 0, "change": 0, "change_pct": null},
    "daily": [
      {
        "date": "2026-10-06",
        "sessions": 37,
        "engaged_sessions": 26,
        "users": 33,
        "conversions": 6,
        "previous_date": "2026-09-08",
        "previous_sessions": 40,
        "previous_engaged_sessions": 27,
        "previous_users": 34,
        "previous_conversions": 7
      },
      {
        "date": "2026-10-07",
        "sessions": 52,
        "engaged_sessions": 32,
        "users": 37,
        "conversions": 14,
        "previous_date": "2026-09-09",
        "previous_sessions": 37,
        "previous_engaged_sessions": 29,
        "previous_users": 33,
        "previous_conversions": 12
      }
    ]
  },
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · in data:

Field	Type	What it is
`compare`	string	previous_period or previous_year: what the previous figures are.
`sessions`	object	Sessions in the period, against the previous period.
`sessions.value`	number or null	This period; null when it does not apply (no sessions).
`sessions.previous`	number or null	The previous period; null when it does not apply.
`sessions.change`	number or null	value minus previous.
`sessions.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`engaged_sessions`	object	Engaged sessions in the period, against the previous period.
`engaged_sessions.value`	number or null	This period; null when it does not apply (no sessions).
`engaged_sessions.previous`	number or null	The previous period; null when it does not apply.
`engaged_sessions.change`	number or null	value minus previous.
`engaged_sessions.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`engagement_rate_pct`	object	Engagement rate, 0-100, in the period, against the previous period.
`engagement_rate_pct.value`	number or null	This period; null when it does not apply (no sessions).
`engagement_rate_pct.previous`	number or null	The previous period; null when it does not apply.
`engagement_rate_pct.change`	number or null	value minus previous, in percentage points.
`engagement_rate_pct.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`users`	object	Users in the period, against the previous period.
`users.value`	number or null	This period; null when it does not apply (no sessions).
`users.previous`	number or null	The previous period; null when it does not apply.
`users.change`	number or null	value minus previous.
`users.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`new_users`	object	New users in the period, against the previous period.
`new_users.value`	number or null	This period; null when it does not apply (no sessions).
`new_users.previous`	number or null	The previous period; null when it does not apply.
`new_users.change`	number or null	value minus previous.
`new_users.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`views`	object	Page views in the period, against the previous period.
`views.value`	number or null	This period; null when it does not apply (no sessions).
`views.previous`	number or null	The previous period; null when it does not apply.
`views.change`	number or null	value minus previous.
`views.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`views_per_session`	object	Page views per session in the period, against the previous period.
`views_per_session.value`	number or null	This period; null when it does not apply (no sessions).
`views_per_session.previous`	number or null	The previous period; null when it does not apply.
`views_per_session.change`	number or null	value minus previous.
`views_per_session.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`avg_session_duration_seconds`	object	Average session duration in seconds in the period, against the previous period.
`avg_session_duration_seconds.value`	number or null	This period; null when it does not apply (no sessions).
`avg_session_duration_seconds.previous`	number or null	The previous period; null when it does not apply.
`avg_session_duration_seconds.change`	number or null	value minus previous.
`avg_session_duration_seconds.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`conversions`	object	Conversions (key events) in the period, against the previous period.
`conversions.value`	number or null	This period; null when it does not apply (no sessions).
`conversions.previous`	number or null	The previous period; null when it does not apply.
`conversions.change`	number or null	value minus previous.
`conversions.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`revenue_eur`	object	Revenue in the period, against the previous period.
`revenue_eur.value`	number or null	This period; null when it does not apply (no sessions).
`revenue_eur.previous`	number or null	The previous period; null when it does not apply.
`revenue_eur.change`	number or null	value minus previous.
`revenue_eur.change_pct`	number or null	The change in percent of previous; null when previous is 0 or null.
`daily`	array	Per day of the period, oldest first, next to the matching day of the previous period.
`daily[].date`	string	The day, YYYY-MM-DD.
`daily[].sessions`	integer	Sessions.
`daily[].engaged_sessions`	integer	Engaged sessions.
`daily[].users`	integer	Users.
`daily[].conversions`	integer	Conversions (key events).
`daily[].previous_date`	string	The matching day of the previous period, YYYY-MM-DD.
`daily[].previous_sessions`	integer	Sessions on previous_date.
`daily[].previous_engaged_sessions`	integer	Engaged sessions on previous_date.
`daily[].previous_users`	integer	Users on previous_date.
`daily[].previous_conversions`	integer	Conversions on previous_date.

analytics.channels

POST /rest/v1/analytics.channels · MCP tool get_analytics_channels

Returns the sessions of one site per channel (Organic Search, Direct, Referral, AI Assistant, …) in the period, most sessions first: all metrics, the share of all sessions and the same figures in the previous period. Always lists every channel; the channel filter does not apply here. The filters narrow everything down (all at once); meta.period gives the period and the previous period. Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`channel`	string	Only sessions from this channel, as Google Analytics names it (Organic Search, Direct, Referral, Organic Social, Paid Search, Email, AI Assistant, …).
`source_medium`	string	Only sessions from this source / medium, exactly as in analytics.sources (google / organic).
`landing_page`	string	Only sessions that started on this page: its path, exactly as in analytics.landing_pages (/blog/post).
`path_prefix`	string	Only sessions that started on a page whose path begins with this (/blog/).
`country`	string	Only sessions from this country, in English as in analytics.audience (Netherlands).
`device`	one of `desktop`, `mobile`, `tablet`, `smart tv`	Only sessions on this kind of device.
`limit`	integer, 1 to 500, default 100	Items per call.
`cursor`	string	meta.next_cursor of the previous call, for the next page.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.channels \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","limit":2}'

Example response · A real answer, anonymised and shortened to two items per list.

{
  "data": [
    {
      "channel": "Organic Search",
      "sessions": 742,
      "engaged_sessions": 520,
      "engagement_rate_pct": 70.1,
      "users": 471,
      "new_users": 417,
      "views": 1534,
      "views_per_session": 2.07,
      "avg_session_duration_seconds": 183.7,
      "conversions": 147,
      "revenue_eur": 0,
      "share_pct": 64.4,
      "previous": {
        "sessions": 696,
        "engaged_sessions": 500,
        "engagement_rate_pct": 71.8,
        "users": 466,
        "new_users": 421,
        "views": 1508,
        "views_per_session": 2.17,
        "avg_session_duration_seconds": 180.7,
        "conversions": 121,
        "revenue_eur": 0
      }
    },
    {
      "channel": "Direct",
      "sessions": 378,
      "engaged_sessions": 216,
      "engagement_rate_pct": 57.1,
      "users": 246,
      "new_users": 237,
      "views": 1042,
      "views_per_session": 2.76,
      "avg_session_duration_seconds": 138.5,
      "conversions": 140,
      "revenue_eur": 0,
      "share_pct": 32.8,
      "previous": {
        "sessions": 316,
        "engaged_sessions": 188,
        "engagement_rate_pct": 59.5,
        "users": 197,
        "new_users": 186,
        "views": 838,
        "views_per_session": 2.65,
        "avg_session_duration_seconds": 170.7,
        "conversions": 83,
        "revenue_eur": 0
      }
    }
  ],
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "next_cursor": "2",
    "total": 7,
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · data is a list; each item has:

Field	Type	What it is
`channel`	string	The channel, as Google Analytics names it.
`sessions`	integer	Sessions.
`engaged_sessions`	integer	Engaged sessions: longer than 10 seconds, two or more page views, or a conversion.
`engagement_rate_pct`	number or null	Engaged sessions as a share of sessions, 0-100; null without sessions.
`users`	integer	Users (active users, as Google Analytics counts them).
`new_users`	integer	Users who came for the first time.
`views`	integer	Page views.
`views_per_session`	number or null	Page views per session; null without sessions.
`avg_session_duration_seconds`	number or null	Average session duration in seconds; null without sessions.
`conversions`	integer	Key events (conversions as set up in Google Analytics).
`revenue_eur`	number	Revenue, in the currency of the Google Analytics property (usually euro).
`share_pct`	number or null	Share of the sessions of the whole list, 0-100.
`previous`	object or null	The same figures in the previous period; null when it did not occur then.
`previous.sessions`	integer	Sessions.
`previous.engaged_sessions`	integer	Engaged sessions: longer than 10 seconds, two or more page views, or a conversion.
`previous.engagement_rate_pct`	number or null	Engaged sessions as a share of sessions, 0-100; null without sessions.
`previous.users`	integer	Users (active users, as Google Analytics counts them).
`previous.new_users`	integer	Users who came for the first time.
`previous.views`	integer	Page views.
`previous.views_per_session`	number or null	Page views per session; null without sessions.
`previous.avg_session_duration_seconds`	number or null	Average session duration in seconds; null without sessions.
`previous.conversions`	integer	Key events (conversions as set up in Google Analytics).
`previous.revenue_eur`	number	Revenue, in the currency of the Google Analytics property (usually euro).

analytics.sources

POST /rest/v1/analytics.sources · MCP tool get_analytics_sources

Returns the sessions of one site per source / medium (google / organic, chatgpt.com / referral, (direct) / (none)) in the period, most sessions first: all metrics, the share of the sessions in the list and the same figures in the previous period. Use channel to see the sources of one channel. The filters narrow everything down (all at once); meta.period gives the period and the previous period. Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`channel`	string	Only sessions from this channel, as Google Analytics names it (Organic Search, Direct, Referral, Organic Social, Paid Search, Email, AI Assistant, …).
`source_medium`	string	Only sessions from this source / medium, exactly as in analytics.sources (google / organic).
`landing_page`	string	Only sessions that started on this page: its path, exactly as in analytics.landing_pages (/blog/post).
`path_prefix`	string	Only sessions that started on a page whose path begins with this (/blog/).
`country`	string	Only sessions from this country, in English as in analytics.audience (Netherlands).
`device`	one of `desktop`, `mobile`, `tablet`, `smart tv`	Only sessions on this kind of device.
`limit`	integer, 1 to 500, default 100	Items per call.
`cursor`	string	meta.next_cursor of the previous call, for the next page.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.sources \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","limit":2}'

Example response · A real answer, anonymised and shortened to two items per list.

{
  "data": [
    {
      "source_medium": "google / organic",
      "source": "google",
      "medium": "organic",
      "sessions": 711,
      "engaged_sessions": 496,
      "engagement_rate_pct": 69.8,
      "users": 448,
      "new_users": 396,
      "views": 1478,
      "views_per_session": 2.08,
      "avg_session_duration_seconds": 181.3,
      "conversions": 145,
      "revenue_eur": 0,
      "share_pct": 61.7,
      "previous": {
        "sessions": 665,
        "engaged_sessions": 479,
        "engagement_rate_pct": 72,
        "users": 443,
        "new_users": 399,
        "views": 1424,
        "views_per_session": 2.14,
        "avg_session_duration_seconds": 180.2,
        "conversions": 111,
        "revenue_eur": 0
      }
    },
    {
      "source_medium": "(direct) / (none)",
      "source": "(direct)",
      "medium": "(none)",
      "sessions": 378,
      "engaged_sessions": 216,
      "engagement_rate_pct": 57.1,
      "users": 246,
      "new_users": 237,
      "views": 1042,
      "views_per_session": 2.76,
      "avg_session_duration_seconds": 138.5,
      "conversions": 140,
      "revenue_eur": 0,
      "share_pct": 32.8,
      "previous": {
        "sessions": 316,
        "engaged_sessions": 188,
        "engagement_rate_pct": 59.5,
        "users": 197,
        "new_users": 186,
        "views": 838,
        "views_per_session": 2.65,
        "avg_session_duration_seconds": 170.7,
        "conversions": 83,
        "revenue_eur": 0
      }
    }
  ],
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "next_cursor": "2",
    "total": 16,
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · data is a list; each item has:

Field	Type	What it is
`source_medium`	string	Source / medium, as Google Analytics gives it; use it as the source_medium filter.
`source`	string	The source (google, chatgpt.com, (direct)).
`medium`	string or null	The medium (organic, referral, cpc, (none)); null when Google Analytics gives none.
`sessions`	integer	Sessions.
`engaged_sessions`	integer	Engaged sessions: longer than 10 seconds, two or more page views, or a conversion.
`engagement_rate_pct`	number or null	Engaged sessions as a share of sessions, 0-100; null without sessions.
`users`	integer	Users (active users, as Google Analytics counts them).
`new_users`	integer	Users who came for the first time.
`views`	integer	Page views.
`views_per_session`	number or null	Page views per session; null without sessions.
`avg_session_duration_seconds`	number or null	Average session duration in seconds; null without sessions.
`conversions`	integer	Key events (conversions as set up in Google Analytics).
`revenue_eur`	number	Revenue, in the currency of the Google Analytics property (usually euro).
`share_pct`	number or null	Share of the sessions of the whole list, 0-100.
`previous`	object or null	The same figures in the previous period; null when it did not occur then.
`previous.sessions`	integer	Sessions.
`previous.engaged_sessions`	integer	Engaged sessions: longer than 10 seconds, two or more page views, or a conversion.
`previous.engagement_rate_pct`	number or null	Engaged sessions as a share of sessions, 0-100; null without sessions.
`previous.users`	integer	Users (active users, as Google Analytics counts them).
`previous.new_users`	integer	Users who came for the first time.
`previous.views`	integer	Page views.
`previous.views_per_session`	number or null	Page views per session; null without sessions.
`previous.avg_session_duration_seconds`	number or null	Average session duration in seconds; null without sessions.
`previous.conversions`	integer	Key events (conversions as set up in Google Analytics).
`previous.revenue_eur`	number	Revenue, in the currency of the Google Analytics property (usually euro).

analytics.landing_pages

POST /rest/v1/analytics.landing_pages · MCP tool get_analytics_landing_pages

Returns the sessions of one site per landing page (the first page of a session) in the period, most sessions first: all metrics, the share of the sessions in the list and the same figures in the previous period. Combine with channel or source_medium to see where a channel lands. The filters narrow everything down (all at once); meta.period gives the period and the previous period. Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`channel`	string	Only sessions from this channel, as Google Analytics names it (Organic Search, Direct, Referral, Organic Social, Paid Search, Email, AI Assistant, …).
`source_medium`	string	Only sessions from this source / medium, exactly as in analytics.sources (google / organic).
`landing_page`	string	Only sessions that started on this page: its path, exactly as in analytics.landing_pages (/blog/post).
`path_prefix`	string	Only sessions that started on a page whose path begins with this (/blog/).
`country`	string	Only sessions from this country, in English as in analytics.audience (Netherlands).
`device`	one of `desktop`, `mobile`, `tablet`, `smart tv`	Only sessions on this kind of device.
`limit`	integer, 1 to 500, default 100	Items per call.
`cursor`	string	meta.next_cursor of the previous call, for the next page.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.landing_pages \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","limit":2}'

Example response · A real answer, anonymised and shortened to two items per list.

{
  "data": [
    {
      "path": "/rental/tens-device",
      "url": "https://www.example-shop.com/rental/tens-device",
      "sessions": 414,
      "engaged_sessions": 307,
      "engagement_rate_pct": 74.2,
      "users": 357,
      "new_users": 328,
      "views": 997,
      "views_per_session": 2.41,
      "avg_session_duration_seconds": 230.7,
      "conversions": 75,
      "revenue_eur": 0,
      "share_pct": 35.9,
      "previous": {
        "sessions": 464,
        "engaged_sessions": 322,
        "engagement_rate_pct": 69.4,
        "users": 394,
        "new_users": 372,
        "views": 1011,
        "views_per_session": 2.18,
        "avg_session_duration_seconds": 206.2,
        "conversions": 68,
        "revenue_eur": 0
      }
    },
    {
      "path": "/",
      "url": "https://www.example-shop.com/",
      "sessions": 193,
      "engaged_sessions": 105,
      "engagement_rate_pct": 54.4,
      "users": 152,
      "new_users": 110,
      "views": 495,
      "views_per_session": 2.56,
      "avg_session_duration_seconds": 156.4,
      "conversions": 12,
      "revenue_eur": 0,
      "share_pct": 16.8,
      "previous": {
        "sessions": 150,
        "engaged_sessions": 97,
        "engagement_rate_pct": 64.7,
        "users": 125,
        "new_users": 87,
        "views": 394,
        "views_per_session": 2.63,
        "avg_session_duration_seconds": 153.2,
        "conversions": 8,
        "revenue_eur": 0
      }
    }
  ],
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "next_cursor": "2",
    "total": 172,
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · data is a list; each item has:

Field	Type	What it is
`path`	string	The path of the landing page, as Google Analytics gives it; (not set) when unknown.
`url`	string or null	The full address of the page; null for (not set).
`sessions`	integer	Sessions.
`engaged_sessions`	integer	Engaged sessions: longer than 10 seconds, two or more page views, or a conversion.
`engagement_rate_pct`	number or null	Engaged sessions as a share of sessions, 0-100; null without sessions.
`users`	integer	Users (active users, as Google Analytics counts them).
`new_users`	integer	Users who came for the first time.
`views`	integer	Page views.
`views_per_session`	number or null	Page views per session; null without sessions.
`avg_session_duration_seconds`	number or null	Average session duration in seconds; null without sessions.
`conversions`	integer	Key events (conversions as set up in Google Analytics).
`revenue_eur`	number	Revenue, in the currency of the Google Analytics property (usually euro).
`share_pct`	number or null	Share of the sessions of the whole list, 0-100.
`previous`	object or null	The same figures in the previous period; null when it did not occur then.
`previous.sessions`	integer	Sessions.
`previous.engaged_sessions`	integer	Engaged sessions: longer than 10 seconds, two or more page views, or a conversion.
`previous.engagement_rate_pct`	number or null	Engaged sessions as a share of sessions, 0-100; null without sessions.
`previous.users`	integer	Users (active users, as Google Analytics counts them).
`previous.new_users`	integer	Users who came for the first time.
`previous.views`	integer	Page views.
`previous.views_per_session`	number or null	Page views per session; null without sessions.
`previous.avg_session_duration_seconds`	number or null	Average session duration in seconds; null without sessions.
`previous.conversions`	integer	Key events (conversions as set up in Google Analytics).
`previous.revenue_eur`	number	Revenue, in the currency of the Google Analytics property (usually euro).

analytics.pages

POST /rest/v1/analytics.pages · MCP tool get_analytics_pages

Returns the page views of one site per page (every viewed page, not only landing pages) in the period, most views first: views, the sessions and users that viewed the page, engagement and duration of those sessions, the share of all views and the same figures in the previous period. The filters narrow everything down (all at once); meta.period gives the period and the previous period. Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`channel`	string	Only sessions from this channel, as Google Analytics names it (Organic Search, Direct, Referral, Organic Social, Paid Search, Email, AI Assistant, …).
`source_medium`	string	Only sessions from this source / medium, exactly as in analytics.sources (google / organic).
`landing_page`	string	Only sessions that started on this page: its path, exactly as in analytics.landing_pages (/blog/post).
`path_prefix`	string	Only sessions that started on a page whose path begins with this (/blog/).
`country`	string	Only sessions from this country, in English as in analytics.audience (Netherlands).
`device`	one of `desktop`, `mobile`, `tablet`, `smart tv`	Only sessions on this kind of device.
`limit`	integer, 1 to 500, default 100	Items per call.
`cursor`	string	meta.next_cursor of the previous call, for the next page.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.pages \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","limit":2}'

Example response · A real answer, anonymised and shortened to two items per list.

{
  "data": [
    {
      "path": "/rental/tens-device",
      "url": "https://www.example-shop.com/rental/tens-device",
      "views": 597,
      "sessions": 486,
      "users": 396,
      "engagement_rate_pct": 76.3,
      "avg_session_duration_seconds": 139.4,
      "share_pct": 22.4,
      "previous": {
        "views": 632,
        "sessions": 528,
        "users": 431,
        "engagement_rate_pct": 71.2,
        "avg_session_duration_seconds": 131
      }
    },
    {
      "path": "/cart/list",
      "url": "https://www.example-shop.com/cart/list",
      "views": 391,
      "sessions": 78,
      "users": 4,
      "engagement_rate_pct": 80.8,
      "avg_session_duration_seconds": 227.7,
      "share_pct": 14.7,
      "previous": {
        "views": 372,
        "sessions": 74,
        "users": 3,
        "engagement_rate_pct": 89.2,
        "avg_session_duration_seconds": 277.5
      }
    }
  ],
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "next_cursor": "2",
    "total": 250,
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · data is a list; each item has:

Field	Type	What it is
`path`	string	The path of the page, as Google Analytics gives it.
`url`	string or null	The full address of the page; null for (not set).
`views`	integer	Page views.
`sessions`	integer	Sessions in which the page was viewed.
`users`	integer	Users who viewed the page.
`engagement_rate_pct`	number or null	Engagement rate of those sessions, 0-100; null without sessions.
`avg_session_duration_seconds`	number or null	Average duration of those sessions in seconds; null without sessions.
`share_pct`	number or null	Share of the views of the whole list, 0-100.
`previous`	object or null	The same figures in the previous period; null when the page had no views then.
`previous.views`	integer	Page views.
`previous.sessions`	integer	Sessions in which the page was viewed.
`previous.users`	integer	Users who viewed the page.
`previous.engagement_rate_pct`	number or null	Engagement rate, 0-100.
`previous.avg_session_duration_seconds`	number or null	Average session duration in seconds.

analytics.audience

POST /rest/v1/analytics.audience · MCP tool get_analytics_audience

Returns the sessions of one site per country, city, device, browser, language or new versus returning (by) in the period, most sessions first: all metrics, the share of the sessions in the list and the same figures in the previous period. The filters narrow everything down (all at once); meta.period gives the period and the previous period. Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`by`	one of `country`, `city`, `device`, `browser`, `language`, `new_vs_returning`	country, city, device (desktop, mobile, tablet), browser, language or new_vs_returning (new, returning). Defaults to country.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`channel`	string	Only sessions from this channel, as Google Analytics names it (Organic Search, Direct, Referral, Organic Social, Paid Search, Email, AI Assistant, …).
`source_medium`	string	Only sessions from this source / medium, exactly as in analytics.sources (google / organic).
`landing_page`	string	Only sessions that started on this page: its path, exactly as in analytics.landing_pages (/blog/post).
`path_prefix`	string	Only sessions that started on a page whose path begins with this (/blog/).
`country`	string	Only sessions from this country, in English as in analytics.audience (Netherlands).
`device`	one of `desktop`, `mobile`, `tablet`, `smart tv`	Only sessions on this kind of device.
`limit`	integer, 1 to 500, default 100	Items per call.
`cursor`	string	meta.next_cursor of the previous call, for the next page.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.audience \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","by":"country","limit":2}'

Example response · A real answer, anonymised and shortened to two items per list.

{
  "data": [
    {
      "key": "Netherlands",
      "sessions": 1023,
      "engaged_sessions": 716,
      "engagement_rate_pct": 70,
      "users": 610,
      "new_users": 552,
      "views": 2488,
      "views_per_session": 2.43,
      "avg_session_duration_seconds": 182.9,
      "conversions": 278,
      "revenue_eur": 0,
      "share_pct": 88.9,
      "previous": {
        "sessions": 964,
        "engaged_sessions": 664,
        "engagement_rate_pct": 68.9,
        "users": 610,
        "new_users": 558,
        "views": 2268,
        "views_per_session": 2.35,
        "avg_session_duration_seconds": 188.8,
        "conversions": 197,
        "revenue_eur": 0
      }
    },
    {
      "key": "United States",
      "sessions": 73,
      "engaged_sessions": 15,
      "engagement_rate_pct": 20.5,
      "users": 73,
      "new_users": 72,
      "views": 109,
      "views_per_session": 1.49,
      "avg_session_duration_seconds": 7.2,
      "conversions": 8,
      "revenue_eur": 0,
      "share_pct": 6.3,
      "previous": {
        "sessions": 23,
        "engaged_sessions": 18,
        "engagement_rate_pct": 78.3,
        "users": 23,
        "new_users": 23,
        "views": 59,
        "views_per_session": 2.57,
        "avg_session_duration_seconds": 16.6,
        "conversions": 8,
        "revenue_eur": 0
      }
    }
  ],
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "next_cursor": "2",
    "total": 23,
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · data is a list; each item has:

Field	Type	What it is
`key`	string	The country, city, device, browser, language or new / returning, as Google Analytics gives it (English names).
`sessions`	integer	Sessions.
`engaged_sessions`	integer	Engaged sessions: longer than 10 seconds, two or more page views, or a conversion.
`engagement_rate_pct`	number or null	Engaged sessions as a share of sessions, 0-100; null without sessions.
`users`	integer	Users (active users, as Google Analytics counts them).
`new_users`	integer	Users who came for the first time.
`views`	integer	Page views.
`views_per_session`	number or null	Page views per session; null without sessions.
`avg_session_duration_seconds`	number or null	Average session duration in seconds; null without sessions.
`conversions`	integer	Key events (conversions as set up in Google Analytics).
`revenue_eur`	number	Revenue, in the currency of the Google Analytics property (usually euro).
`share_pct`	number or null	Share of the sessions of the whole list, 0-100.
`previous`	object or null	The same figures in the previous period; null when it did not occur then.
`previous.sessions`	integer	Sessions.
`previous.engaged_sessions`	integer	Engaged sessions: longer than 10 seconds, two or more page views, or a conversion.
`previous.engagement_rate_pct`	number or null	Engaged sessions as a share of sessions, 0-100; null without sessions.
`previous.users`	integer	Users (active users, as Google Analytics counts them).
`previous.new_users`	integer	Users who came for the first time.
`previous.views`	integer	Page views.
`previous.views_per_session`	number or null	Page views per session; null without sessions.
`previous.avg_session_duration_seconds`	number or null	Average session duration in seconds; null without sessions.
`previous.conversions`	integer	Key events (conversions as set up in Google Analytics).
`previous.revenue_eur`	number	Revenue, in the currency of the Google Analytics property (usually euro).

analytics.events

POST /rest/v1/analytics.events · MCP tool get_analytics_events

Returns the events of one site per event name (page_view, scroll, purchase, form_submit, …) in the period, most first: how often, how many of them count as conversions (key events), the users and the revenue, and the same in the previous period. The filters narrow everything down (all at once); meta.period gives the period and the previous period. Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`channel`	string	Only sessions from this channel, as Google Analytics names it (Organic Search, Direct, Referral, Organic Social, Paid Search, Email, AI Assistant, …).
`source_medium`	string	Only sessions from this source / medium, exactly as in analytics.sources (google / organic).
`landing_page`	string	Only sessions that started on this page: its path, exactly as in analytics.landing_pages (/blog/post).
`path_prefix`	string	Only sessions that started on a page whose path begins with this (/blog/).
`country`	string	Only sessions from this country, in English as in analytics.audience (Netherlands).
`device`	one of `desktop`, `mobile`, `tablet`, `smart tv`	Only sessions on this kind of device.
`limit`	integer, 1 to 500, default 100	Items per call.
`cursor`	string	meta.next_cursor of the previous call, for the next page.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.events \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","limit":2}'

Example response · A real answer, anonymised and shortened to two items per list.

{
  "data": [
    {
      "event": "page_view",
      "count": 2668,
      "conversions": 0,
      "users": 725,
      "revenue_eur": 0,
      "previous": {"count": 2379, "conversions": 0, "users": 672, "revenue_eur": 0}
    },
    {
      "event": "user_engagement",
      "count": 1991,
      "conversions": 0,
      "users": 509,
      "revenue_eur": 0,
      "previous": {"count": 1802, "conversions": 0, "users": 481, "revenue_eur": 0}
    }
  ],
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "next_cursor": "2",
    "total": 10,
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · data is a list; each item has:

Field	Type	What it is
`event`	string	The event name.
`count`	integer	How often the event happened.
`conversions`	integer	How many of them are key events (0 when the event is not a key event).
`users`	integer	Users with the event.
`revenue_eur`	number	Revenue with the event, in the currency of the property.
`previous`	object or null	The same in the previous period; null when the event did not occur then.
`previous.count`	integer	How often.
`previous.conversions`	integer	Key events.
`previous.users`	integer	Users.
`previous.revenue_eur`	number	Revenue.

analytics.products

POST /rest/v1/analytics.products · MCP tool get_analytics_products

Returns the e-commerce products of one site in the period, most revenue first: revenue, items bought and items viewed, and the same in the previous period. Empty when the site sends no e-commerce data to Google Analytics. The filters do not apply here (product data has no sessions). Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`limit`	integer, 1 to 500, default 100	Items per call.
`cursor`	string	meta.next_cursor of the previous call, for the next page.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.products \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","limit":2}'

Example response · Illustrative: the account behind the other examples sends no e-commerce data, so these products are made up in the exact shape of the answer.

{
  "data": [
    {
      "product": "TENS device, 4 weeks",
      "revenue_eur": 1188,
      "purchased": 33,
      "viewed": 412,
      "previous": {"revenue_eur": 972, "purchased": 27, "viewed": 380}
    },
    {
      "product": "Electrode pads (4 pack)",
      "revenue_eur": 89.7,
      "purchased": 10,
      "viewed": 61,
      "previous": null
    }
  ],
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "next_cursor": "2",
    "total": 14,
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · data is a list; each item has:

Field	Type	What it is
`product`	string	The product name (item name).
`revenue_eur`	number	Revenue from the product, in the currency of the property.
`purchased`	integer	Items bought.
`viewed`	integer	Times the item was viewed.
`previous`	object or null	The same in the previous period; null when the product did not occur then.
`previous.revenue_eur`	number	Revenue.
`previous.purchased`	integer	Items bought.
`previous.viewed`	integer	Times viewed.

analytics.referrers

POST /rest/v1/analytics.referrers · MCP tool get_analytics_referrers

Returns the pages on other sites that sent visitors to one site in the period, per referring page and landing page, most sessions first: sessions, engaged sessions and page views, and the same in the previous period. Search engines and the site itself are left out. Browsers often pass only the domain of the referring site, not the page; then referrer_url is the domain and domain_only is true. Combine with source_medium (chatgpt.com / referral) for one source. The filters narrow everything down (all at once); meta.period gives the period and the previous period. Fails with 409 not_connected when Google Analytics 4 is not connected for the site.

Arguments

Argument	Type	What it does
`site_id` (required)	string	The site: its id from list_sites, or the id in the app’s address.
`from`	string, `YYYY-MM-DD`	First day of the period, YYYY-MM-DD; give it together with to.
`to`	string, `YYYY-MM-DD`	Last day of the period, YYYY-MM-DD (yesterday at the latest); give it together with from. At most 366 days.
`days`	integer, 1 to 366, default 28	Without from and to: the last this many days up to yesterday. Defaults to 28.
`compare`	one of `previous_period`, `previous_year`	What the previous figures are: previous_period (the same number of days just before) or previous_year (the same days 52 weeks earlier, same weekdays; for periods up to 364 days). Defaults to previous_period.
`channel`	string	Only sessions from this channel, as Google Analytics names it (Organic Search, Direct, Referral, Organic Social, Paid Search, Email, AI Assistant, …).
`source_medium`	string	Only sessions from this source / medium, exactly as in analytics.sources (google / organic).
`landing_page`	string	Only sessions that started on this page: its path, exactly as in analytics.landing_pages (/blog/post).
`path_prefix`	string	Only sessions that started on a page whose path begins with this (/blog/).
`country`	string	Only sessions from this country, in English as in analytics.audience (Netherlands).
`device`	one of `desktop`, `mobile`, `tablet`, `smart tv`	Only sessions on this kind of device.
`limit`	integer, 1 to 500, default 100	Items per call.
`cursor`	string	meta.next_cursor of the previous call, for the next page.

Example request

curl -X POST https://app.marketingtracer.com/rest/v1/analytics.referrers \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"site_id":"YOUR_SITE_ID","limit":2}'

Example response · A real answer, anonymised and shortened to two items per list.

{
  "data": [
    {
      "referrer_url": "https://www.example-forum.com",
      "referrer_domain": "example-forum.com",
      "domain_only": true,
      "landing_page": "/rental/tens-device",
      "landing_url": "https://www.example-shop.com/rental/tens-device",
      "sessions": 7,
      "engaged_sessions": 2,
      "views": 8,
      "previous": {"sessions": 0, "engaged_sessions": 0, "views": 0}
    },
    {
      "referrer_url": "https://l.wl.co",
      "referrer_domain": "l.wl.co",
      "domain_only": true,
      "landing_page": "/rental/tens-device",
      "landing_url": "https://www.example-shop.com/rental/tens-device",
      "sessions": 3,
      "engaged_sessions": 1,
      "views": 4,
      "previous": {"sessions": 0, "engaged_sessions": 0, "views": 0}
    }
  ],
  "meta": {
    "site_id": "YOUR_SITE_ID",
    "next_cursor": "2",
    "total": 5,
    "period": {
      "from": "2026-10-06",
      "to": "2026-11-02",
      "compare_from": "2026-09-08",
      "compare_to": "2026-10-05"
    }
  }
}

Response fields · data is a list; each item has:

Field	Type	What it is
`referrer_url`	string	The referring page without query string; only https://domain when the browser passed no more.
`referrer_domain`	string	The domain of the referring site, without www.
`domain_only`	boolean	Whether only the domain of the referring page is known.
`landing_page`	string	The path of the page the visitor arrived on.
`landing_url`	string or null	The full address of that page; null when unknown.
`sessions`	integer	Sessions.
`engaged_sessions`	integer	Engaged sessions.
`views`	integer	Page views.
`previous`	object or null	The same in the previous period; null when it did not occur then.
`previous.sessions`	integer	Sessions.
`previous.engaged_sessions`	integer	Engaged sessions.
`previous.views`	integer	Page views.

White label / agency

White-label Google Analytics 4 API. On your own domain, under your name.

When your white-label dashboard runs on your own domain, the REST API and the MCP server answer at that address too. One personal API key works across every client site you can access. You decide whether your clients also get the API and the MCP.

API hubSEO API overview MCPGoogle Search Console MCP AI assistant trafficAI visibility API MCPSEO MCP server

Free 14-day trial

MarketingTracer smart A.I. tool shows you what you need to rank. Then we'll help you get it done.

No credit card required
No installation needed
No strings attached

Start your 14-day trial

FAQ

Something else? Ask us.

Is the Google Analytics 4 API included in every plan?

Yes. The Google Analytics 4 endpoints and MCP tools are included in every plan at no extra cost. One personal API key from the MCP page in the app works for both the REST API and the MCP server across all your sites. See the pricing page for plan details.

Which date ranges and comparisons can I request?

Pass from and to (YYYY-MM-DD) together, or set days (default 28) for a rolling window ending yesterday. Today is excluded because today's Google Analytics 4 data is not complete. Set compare to previous_period for the preceding window of equal length, or previous_year for the same days 52 weeks earlier so weekdays match.

How do the shared filters work across endpoints?

Pass any combination of channel, source_medium, landing_page, path_prefix, country and device on analytics.summary and the session-based lists. All filters you pass apply together. Two endpoints work differently: analytics.channels always lists every channel and ignores channel, and analytics.products takes no session filters because GA4 product rows are item-scoped.

How do I connect Claude Code or Cursor to the Google Analytics 4 MCP tools?

Point your editor at /rest/mcp/server on app.marketingtracer.com or on your own white-label domain. Pass your personal API key from the MCP page in the app as a Bearer token. Ready-to-paste setup commands for Claude Code and Cursor are on the SEO MCP page.

What is the rate limit, and can guest users read analytics data?

You get 60 calls per minute per key across the REST API and the MCP server together. Every endpoint and MCP tool is read-only, and users with guest access on a site can read the Google Analytics 4 endpoints too.

What happens when Google Analytics 4 is not connected, or when a row has zero sessions?

If Google Analytics 4 is not connected for a site, the endpoint returns HTTP 409 with error code not_connected. When a period or row has 0 sessions, ratios and averages (engagement_rate_pct, views_per_session and avg_session_duration_seconds) return null. On breakdown rows, previous is also null if that row did not appear in the comparison window.

Why is domain_only true on some rows in analytics.referrers?

Many browsers strip the full path from the referrer header on cross-site links and send only the origin. When Google Analytics 4 receives only the domain, referrer_url is https://domain and domain_only is true. If the browser passed the full page address, domain_only is false.