Seal Copilot

Get traffic broken down by device type (desktop/mobile/tablet), browser (Chrome, Safari, Firefox...), and operating system (Windows, macOS, iOS, Android...). Returns all three breakdowns in a single call.

Read one page of the official SealMetrics documentation as plain text, fetched live from docs.sealmetrics.com. Takes a documentation page path (e.g. 'getting-started/quick-start') or a full docs.sealmetrics.com URL.

Get the funnel table broken down by UTM source/medium/campaign/term: per row, entrances, page_views, and counts of each microconversion type and conversion type plus revenue per conversion type, with period totals and the list of types seen. The counts are independent aggregates, not a sequential funnel: a row can show more conversions than add_to_cart events, and ratios between columns are not per-session dropoff. Ratios computed from it are aggregate ratios. Rows are the top `limit` UTM combinations by entrances (default 100, max 500); `truncated: true` means more exist, so a small source may be missing: the utm_source / utm_medium / utm_campaign filters (exact match, case-insensitive) narrow the rows. `totals` always cover the whole filtered period.

Get landing page performance: entrances, bounce rate, conversions. Landing pages are the first page users see when entering the site. High bounce rates indicate poor landing page experience. Supports multi-value filters (country, device_type, browser, os, channel_group, UTMs) as arrays of strings — e.g. country=['ES','PT'] to filter to two countries. Pass include=['channel_group'] to also break down by channel group. The two are orthogonal — combine them as needed.

Get landing page metrics grouped by content grouping. Shows aggregate performance per content group (e.g. blog, product, category).

Returns the SealMetrics marketing analysis method as text: the steps, checks and benchmarks for a weekly, monthly or quarterly performance report and for diagnosing a channel or campaign. It contains no site data. Diagnose marketing performance for a SealMetrics site and explain *why* the numbers move. Generalist marketing analysis (acquisition, channels, landing pages, conversions, traffic quality) with strong bias toward causal explanation and concrete next actions for Founders, CEOs and CMOs.

Get detailed breakdown for a specific microconversion type. Shows metrics segmented by source, medium, campaign, country, device, browser, and OS.

Get microconversions (smaller engagement events) broken down by type: add_to_cart, newsletter_signup, pdf_download, etc. Shows count and trends per event type; aggregated, no per-event rows.

Returns raw microconversion rows from /stats/microconversions/raw (one row per event, with timestamp_utc and timestamp_local). Use for one-row-per-event detail of microconversions (add_to_cart, newsletter_signup, etc.). Custom `properties` are excluded by default — pass `include_properties=true` to receive them. Accepts ranges of up to 31 days (e.g. today, yesterday, 7d, 30d, last_week, last_month, or start_date/end_date); longer ones are rejected. `limit` defaults to 10 and is capped at 100 rows; `page` returns further rows.

Get traffic broken down by operating system: Windows, macOS, iOS, Android, Linux, etc. Shows entrances and pageviews per OS with pagination.

Get dashboard KPIs: pageviews, entrances, bounce rate, conversions, revenue. Includes time series data and optional period-over-period comparison. This is the best starting point for understanding a site's performance.

Get metrics per page URL path: pageviews and entrances. Useful for finding most popular pages and content performance. No bounce rate: bounce is only meaningful per entry page, not per page viewed. Supports multi-value filters (country, device_type, browser, os, channel_group, UTMs) as arrays of strings — e.g. device_type=['mobile'] for mobile only. include=['device'] adds a per-device breakdown; filters and include combine.

Get a complete property breakdown with pivot-table style data. Shows all values for a property key with their counts and revenue — ideal for analyzing product or category performance.

Get property values with counts, grouped by a UTM parameter. Paginated. Use to see which values a specific property key has and how they distribute across traffic sources.

Get detailed information about one site: name, domains, timezone, configuration and tracking status.

Get traffic and conversions broken down by UTM term (keyword). Shows which search keywords or ad terms drive the most traffic. Supports filtering by source, medium, and campaign.

Get the top campaigns (utm_campaign) ranked by entrances, as a compact list of the top N. No period comparison, sorting or pagination.

Get the top channels (Paid Search, Organic, Social, etc.) ranked by entrances, as a compact list of the top N. No pagination.

Get the top landing pages ranked by entrances, as a compact list of the top N. Filterable by content_grouping. No period comparison, sorting, pagination or device/browser/channel/path filters.

Get the top pages ranked by page views, as a compact list of the top N. No period comparison, sorting, pagination or device/browser/channel/path filters.

Get top referrer domains ranked by entrances. Shows which external sites send the most traffic.

Get the top traffic sources (utm_source) ranked by entrances, as a compact list of the top N. No period comparison, sorting or pagination. With landing_page, only the sessions that entered on that page are counted.

Get the top UTM terms (keywords) ranked by entrances, as a compact list of the top N. No period comparison, sorting or pagination.

Get the tracking pixel code for a site and the full JavaScript API reference for implementing pageviews, conversions, microconversions, and content grouping. Returns the site-specific <script> tag plus implementation guide with examples for e-commerce, SaaS, blog, and React/Next.js.

Get traffic broken down by medium (utm_medium): organic, cpc, email, referral, social, etc. Useful for understanding which marketing channels drive traffic.

Get traffic broken down by source (utm_source): google, facebook, twitter, direct, etc. Shows entrances, pageviews, conversions, and revenue per source.

Returns the SealMetrics troubleshooting guide as text; it contains no site data. Validated symptom→cause→fix answers for the most common SealMetrics setup and data questions - tags firing before the tracker loads, lost conversions/microconversions, duplicate pageviews in SPAs or Tag Manager setups, channel rules that never match (traffic stuck in Unassigned), payment-gateway and login/SSO domains showing up as traffic sources, numbers that do not match another analytics tool, filters that seem to be ignored (country names instead of ISO codes, landing-page filters on tools that do not take one), and live browser checks of whether the tag actually loads on a page.

List channel group rules configured for a site. Shows how traffic is classified into channels (Paid Search, Organic, Social, etc.) based on UTM parameters and referrer patterns.

List the microconversion type names that have events on a site in the period.

List the custom property keys present on conversions and/or microconversions in the period.

List the sites (web properties) this connection can read. Returns each site's ID, name and domains; the ID is the site_id the report tools take.

Search the available SealMetrics analytics reports for the connected site. Returns report references, each with an id that retrieves the report. A time range in the query (e.g. 'last 7 days') scopes the reports. No filters or period comparisons.

Search the official SealMetrics documentation (docs.sealmetrics.com), fetched live. Covers tracker installation (WordPress, WooCommerce, Shopify, Magento, PrestaShop, Google Tag Manager, React/Next.js/Vue SPAs), conversion and microconversion tracking, UTM/campaign tagging, report and metric definitions, API usage, account settings, and privacy/GDPR/consentless questions. Returns matching pages with title, path and description.

Classify a source/medium/campaign combination exactly like the tracking pixel would (same RE2 semantics, lowercase matching, custom rules before defaults). Returns the resulting channel and the rule that matched. With include_inactive=true, draft (not live) custom rules are evaluated too, so a draft can be checked before it is published in the dashboard.