API

Sources

The Sources API returns the AI-cited domains and URLs analyzed throughout a project. It is part of citation analysis of SE Visible where prompts and AI results tell you what AI engines said, and sources tell you which web pages they pulled the information from. Use it to identify the domains and pages that drive your AI visibility, see how often they appear in AI answers, and track whether they mention your brand or your competitors’ brands.

Using the SE Visible API, you can:

list the domains (or subdomains) AI models cite for a project, with appearance counts, coverage, prompt reach, and domain trust, backlink, and traffic data
drill into the individual URLs behind those domains, with page title, category, HTTP status, traffic, and backlink type
filter sources by model, location, topic, prompt, sentiment, brand mentions, competitor presence, backlink type, and domain type
compare a selected metric across two different date ranges to analyze performance changes
get aggregated source metrics grouped by domain, URL, page type, domain type, AI model, or check date for tables and charts

All endpoints are relative to the base URL https://api.seranking.com/v1/se-visible.

Get project sources

GET https://api.seranking.com/v1/se-visible/projects/{project_id}/sources

Retrieves the sources (domains, subdomains, or URLs) that AI models cited across a project, with extensive filtering. The shape of each row in the response depends on source_type: domain returns one row per registrable domain (its subdomains are aggregated into the row), subdomain returns one row per host exactly as collected, and url returns one row per cited page.

Note:

source_type switches the response between domain rows (domain and subdomain) and URL rows. It also changes which source_sort_field and columns values are valid: passing a value that belongs to the other type returns 422.
Most numeric fields are wrapped in a metric object with a values array. A single period returns one entry in values. When you supply date_range[1] (compare mode), the field returns two entries plus a delta with absolute and percentage.

Request parameters

Path parameters

Parameter	Type	Required	Description
project_id	String (UUID)	Yes	UUID of the project.

Query parameters

Parameter	Type	Required	Default	Description
date_range	Array of Object	Yes	N/A	1 to 2 entries: base period at index `0`, optional compare period at index `1`. Each has `from` and `to` in `YYYY-MM-DD`. See Date ranges and compare mode.
source_type	String	No	`domain`	Row shape of the response: `domain` (one row per registrable domain, subdomains aggregated into it), `subdomain` (one row per host as collected), or `url` (one row per page). Also determines the valid set of `source_sort_field` and `columns` values.
model_types	Array of String	No	N/A	Filter by AI model types (e.g. `chatgpt`).
country_codes	Array of String	No	N/A	Deprecated — use `location_ids`. Filter by ISO 3166-1 alpha-2 country codes. Resolved to the project’s live locations in those countries, so a deleted location is not matched, a country the project no longer tracks returns an empty result, and two locations in one country cannot be told apart. It keeps working for now; migrate to `location_ids` before it is withdrawn.
location_ids	Array of String (UUID)	No	N/A	Filter by the project’s tracked locations, using location IDs from List project locations. Multiple values combine as OR. An ID from another project or of a deleted location returns `422`. Combined with `country_codes`, only locations matching both are read. A row is attributed to the location its check collected it under, so a location added later has no earlier history.
topic_ids	Array of Integer	No	N/A	Filter by topic IDs.
prompt_ids	Array of Integer	No	N/A	Filter by prompt IDs.
req_num	Integer	No	N/A	Limit to the answers with this request number within a check.
sentiments	Array of String	No	N/A	Filter by sentiment: `positive`, `neutral`, `negative`.
exclude_branded	Boolean	No	N/A	When `true`, every number in the response is computed from non-branded prompts only, and branded prompts drop out. A prompt not yet classified counts as non-branded. Off when omitted.
search_query	String	No	N/A	Filter sources by a substring of the domain or URL.
domain_id	Integer	No	N/A	Restrict results to a single domain.
types	Array of String	No	N/A	Filter by URL category type (e.g. `blog`, `news`, `social`). See the Reference for the full list.
is_my_brand_mentions	Integer	No	N/A	Filter sources by whether the project’s primary brand is mentioned: `1` mentioned, `0` not mentioned, `-1` mention status not available.
has_backlink	Array of String	No	N/A	Filter source pages by `backlink_type`: `dofollow` (links to a brand domain and passes authority), `nofollow` (links to a brand domain without passing authority), `none` (checked, no link to a brand domain). Several values combine as OR. Pages not yet checked or that could not be read match none of the values; `-1` is not accepted. URL rows only: with `source_type=domain` (the default) it returns `422`.
competitors_count_from	Integer	No	N/A	Minimum number of competitor brands on the source (inclusive, min `0`).
competitors_count_to	Integer	No	N/A	Maximum number of competitor brands on the source (inclusive, min `0`).
brand_ids	Array of Integer	No	N/A	Keep sources that mention these brand IDs (see `brand_match_mode`).
brand_match_mode	String	No	`any`	When filtering by `brand_ids`, match `any` of the brands or `all` of them.
domain_types	Array of String	No	N/A	Filter rows by their domain type; several values combine as OR. Matches any element of the row’s `domain_type` array. Values: `you` (the primary brand’s own domain), `competitors` (a domain of another tracked brand), `corporate`, `ugc`, `editorial`, `reference`, `institutional` (stored base classification), `other` (neither a role nor a classification). `domain` and `subdomain` rows only: with `source_type=url` it returns `422`.
columns	Array of String	No	All columns	Column ids to return. When omitted, every column is returned. When provided, each row contains only these columns plus the always-present locked first column (`domain` or `url`), `domain_id`, which is used to drill down from a domain row to its URLs, and `url_id` on URL rows. Key order in the response is not guaranteed. The valid set depends on `source_type`; see the note below.
source_sort_field	String	No	`ai_answers_count`	Field to sort by. The valid set depends on `source_type`; see the note below.
sort_order	String	No	`desc`	Sort direction: `asc` or `desc`.
limit	Integer	No	`1000`	Maximum number of sources to return (min `0`). `0` returns every row and ignores `offset`.
offset	Integer	No	`0`	Number of sources to skip from the start of the list (min `0`).

Valid source_sort_field values

source_type=domain and source_type=subdomain — ai_answers_count, pages, coverage, prompts_count, mention_rate, my_brand_mentions_count, competitors_count, first_seen, last_seen, dt, backlinks, referring_domains, domain_traffic, and the delta variants ai_answers_count_delta, pages_delta, coverage_delta, prompts_count_delta, mention_rate_delta, my_brand_mentions_count_delta, competitors_count_delta, responses_count_delta.
source_type=url — ai_answers_count, coverage, prompts_count, competitors_count, my_brand_mentions, gap_score, first_seen, last_seen, last_check_date, dt, traffic, domain_traffic, and the delta variants ai_answers_count_delta, prompts_count_delta, coverage_delta, responses_count_delta.
The delta variants sort by the absolute change between periods (period 0 minus period 1) and require date_range[1], otherwise the request is rejected with 422. Rows with a null delta — for example when a strict-null metric such as coverage is missing in one period — are pushed to the end of the result in both ascending and descending order.
Rows with no value for dt, traffic, domain_traffic, backlinks, referring_domains, gap_score, first_seen, last_seen, or last_check_date also go to the end in both directions. Period metrics computed per request, such as coverage, keep the default order: nulls first ascending, last descending.

Valid columns ids

source_type=domain — domain, domain_type, subdomains_count, pages, coverage, ai_answers_count, prompts_count, my_brand_mentions_count, mention_rate, brand_ids, domain_traffic, dt, backlinks, referring_domains, first_seen, last_seen, competitors_count, responses_count.
source_type=subdomain — same as domain, minus subdomains_count.
source_type=url — url, title, type, coverage, ai_answers_count, prompts_count, my_brand_mentions, gap_score, brand_ids, domain_traffic, dt, traffic, backlink_type, last_check_date, first_seen, last_seen, status, competitors_count, responses_count.
The locked column (domain or url) is accepted but redundant — it is always returned regardless of the requested ids. An id outside the set allowed for the current source_type returns 422.

Note: tracked_brand_ids is no longer part of this endpoint’s contract. It was previously accepted but never applied, so removing it changes no response. If your integration passed it here, the data you received was unfiltered — please re-check it. To filter sources by brand, use brand_ids together with brand_match_mode.

Request example

Copy

curl -X GET --globoff 'https://api.seranking.com/v1/se-visible/projects/550e8400-e29b-41d4-a716-446655440000/sources?source_type=domain&date_range[0][from]=2026-05-01&date_range[0][to]=2026-05-31&source_sort_field=ai_answers_count&sort_order=desc' \

  -H 'Authorization: Token YOUR_API_KEY'

To return only selected columns, pass columns[]:

Copy

curl -X GET --globoff 'https://api.seranking.com/v1/se-visible/projects/550e8400-e29b-41d4-a716-446655440000/sources?source_type=url&date_range[0][from]=2026-05-01&date_range[0][to]=2026-05-31&columns[]=title&columns[]=coverage&columns[]=ai_answers_count' \

  -H 'Authorization: Token YOUR_API_KEY'

Response parameters

If successful, the server returns a sources object with the row list and a total count.

Parameter	Type	Description
items	Array	Source rows. Each row is a domain object (when `source_type=domain`) or a URL object (when `source_type=url`).
total	Integer	Total number of sources available for the filters.

Most numeric fields are returned as a metric object rather than a plain number:

Field	Type	Description
values	Array of Number	Metric value per period. One entry for a single period; two entries in compare mode (base, then compare). An entry can be `null` when a period has no data.
delta	Object	Present only in compare mode. Difference between the two periods.
delta.absolute	Number or null	Absolute change (base minus compare). `null` when a period has no data.
delta.percentage	Number or null	Percentage change. `null` when a period has no data.

Domain rows (source_type=domain or source_type=subdomain)

Parameter	Type	Description
domain	String	Domain name: the registrable domain for `source_type=domain`, the host exactly as collected for `source_type=subdomain`.
domain_id	Integer	Unique domain identifier.
domain_type	Array of String	The row’s domain types (1 or 2 values): the project role first — `you` for the primary brand’s own domain, otherwise `competitors` for a domain owned by another tracked brand — then the stored base classification (`corporate`, `ugc`, `editorial`, `reference`, `institutional`). `["other"]` when the row has neither, so a competitor that is also an editorial site reads `["competitors", "editorial"]`.
brand_ids	Array of Integer	Brand IDs mentioned on this domain.
first_seen	Datetime	First time this domain appeared in results.
last_seen	Datetime	Last time this domain appeared in results.
ai_answers_count	Metric object	Distinct answers in the period that cite at least one page of this source.
pages	Metric object	Distinct pages of this domain cited in the period’s answers.
subdomains_count	Metric object	Distinct subdomains of the domain that appeared as a source in the period. `source_type=domain` only; absent for `subdomain` rows.
coverage	Metric object	`ai_answers_count` as a percentage of `responses_count`, one decimal. `null` when `responses_count` is `0`.
competitors_count	Metric object	Distinct competitor brands attached to this domain’s source URLs in the period.
prompts_count	Metric object	Distinct prompts whose answers cited this source in the period.
my_brand_mentions_count	Metric object	URLs on this domain that mention the primary brand.
mention_rate	Metric object	Percentage of this domain’s pages in the period whose content mentions the primary brand (`my_brand_mentions_count / pages * 100`), whole number. `null` in a period where the domain has no pages.
responses_count	Metric object	Answers in the period that name at least one brand, after the request’s filters. The same for every row; `0` in a period the source was not cited.
dt	Integer or null	Domain trust (`0`–`100` inlink rank). Domain-level and project-agnostic, so a plain number rather than a metric object. A non-negative number is a measurement, `-1` means the data source had nothing for this row, and `null` means it has not been requested yet.
backlinks	Integer or null	Total backlinks to the domain. Plain number; same `null` rule as `dt`.
referring_domains	Integer or null	Total referring domains. Plain number; same `null` rule as `dt`.
domain_traffic	Integer or null	Estimated monthly organic search traffic of the whole domain. Plain number; `-1` when no data is available, `null` when not requested yet.

URL rows (source_type=url)

Parameter	Type	Description
url	String	Full URL.
url_id	Integer	URL identifier. Always returned, like `domain_id`.
domain_id	Integer	Identifier of the domain this URL belongs to.
brand_ids	Array of Integer or null	Brand IDs mentioned on this URL. An empty list means the page was read and no tracked brand matched. `-1` means the page could not be read; `null` means it has not been parsed yet. Always consistent with `my_brand_mentions`.
title	String	Page title extracted from the HTML.
type	String	URL category type (e.g. `blog`, `news`, `social`). See the Reference for the full list.
status	Integer or null	HTTP status code of the URL. `null` if not yet checked.
first_seen	Datetime	First time this URL appeared in results.
last_seen	Datetime	Last time this URL appeared in results.
last_check_date	Datetime or null	When this URL was last parsed for brand mentions. `null` if not yet parsed.
competitors_count	Integer	Competitor brand count for this URL. Project-scoped rather than period-scoped, so it is a plain number rather than a metric object. `my_brand_mentions` and `gap_score` behave the same way.
my_brand_mentions	Integer or null	`1` if the URL currently mentions the primary brand, `0` if the page was read and the primary brand was not found, `-1` if the page could not be read, `null` if not parsed yet.
gap_score	Integer or null	Gap score for ranking: base-period `ai_answers_count` times the number of tracked brands found on the page, so a page citing competitors in many answers scores high. `0` when the page mentions the primary brand, `-1` when no tracked brand was found on the page or the page could not be read; `null` while the page is waiting for its first parse.
ai_answers_count	Metric object	Distinct answers in the period that cite at least one page of this source.
coverage	Metric object	`ai_answers_count` as a percentage of `responses_count`, one decimal. `null` when `responses_count` is `0`.
prompts_count	Metric object	Distinct prompts whose answers cited this source in the period.
responses_count	Metric object	Answers in the period that name at least one brand, after the request’s filters. The same for every row; `0` in a period the source was not cited.
dt	Integer or null	Domain trust (`0`–`100` inlink rank) of the URL’s domain. Plain number. A non-negative number is a measurement, `-1` means the data source had nothing for this row, and `null` means it has not been requested yet.
traffic	Integer or null	Estimated monthly organic search traffic of this page, worldwide (not scoped to a search region). Plain number; `-1` when no data is available, `null` when not requested yet.
domain_traffic	Integer or null	Estimated monthly organic search traffic of the page’s whole domain — the same value the domain row shows, distinct from `traffic`. Same `-1` and `null` rules as `traffic`.
backlink_type	String or null	Whether this URL links to a domain of a tracked brand: `dofollow` (the link passes authority), `nofollow` (it does not), `none` (checked, no link to a brand domain), `-1` (a parse was attempted and no link type was determined), `null` (not parsed yet).

Response example

Domain sources, single period

Copy

{

  "items": [

    {

      "domain": "example.com",

      "domain_id": 123,

      "ai_answers_count": {

        "values": [45]

      },

      "pages": {

        "values": [12]

      },

      "coverage": {

        "values": [75.5]

      },

      "brand_ids": [1, 2],

      "competitors_count": {

        "values": [2]

      },

      "prompts_count": {

        "values": [10]

      },

      "my_brand_mentions_count": {

        "values": [3]

      },

      "mention_rate": {

        "values": [25]

      },

      "responses_count": {

        "values": [60]

      },

      "first_seen": "2024-01-01 00:00:00",

      "last_seen": "2024-01-15 00:00:00",

      "domain_type": ["competitors", "editorial"],

      "subdomains_count": {

        "values": [2]

      },

      "dt": 62,

      "backlinks": 1520,

      "referring_domains": 84,

      "domain_traffic": 41000

    }

  ],

  "total": 1

}

URL sources, compare mode

Copy

{

  "items": [

    {

      "url": "https://example.com/article1",

      "url_id": 4567,

      "domain_id": 123,

      "ai_answers_count": {

        "values": [15, 8],

        "delta": {

          "absolute": 7,

          "percentage": 87.5

        }

      },

      "coverage": {

        "values": [25.5, 16],

        "delta": {

          "absolute": 9.5,

          "percentage": 59.38

        }

      },

      "type": "blog",

      "title": "Article 1",

      "brand_ids": [1, 2],

      "competitors_count": 2,

      "my_brand_mentions": 1,

      "gap_score": null,

      "prompts_count": {

        "values": [5, 3],

        "delta": {

          "absolute": 2,

          "percentage": 66.67

        }

      },

      "responses_count": {

        "values": [60, 50],

        "delta": {

          "absolute": 10,

          "percentage": 20

        }

      },

      "first_seen": "2024-01-01 00:00:00",

      "last_seen": "2024-02-15 00:00:00",

      "last_check_date": "2024-02-15 00:00:00",

      "status": 200,

      "dt": 62,

      "backlink_type": "dofollow",

      "domain_traffic": null,

      "traffic": 4321

    }

  ],

  "total": 1

}

Get aggregated source metrics

GET https://api.seranking.com/v1/se-visible/projects/{project_id}/sources/metrics

Returns aggregated source metrics for dynamic table and chart rendering — the sources equivalent of Get aggregated brand metrics. You choose how the data is grouped (dimensions[]) and which metrics to return (metrics[]).

Note:

You pass 1 to 3 dimensions: at most one entity or type grain (url, domain, subdomain, page_type, domain_type), optionally model_type (the AI engine of the citing answers) beside it, and optionally one date dimension — check_id or check_date for one row per check, check_week or check_month for one row per calendar period (its checks aggregated, labelled by the period start). model_type on its own is also a grain.
With 1 dimension, each row holds that dimension’s own columns plus one metric object per requested metric.
With 2 or 3 dimensions, the response is a pivot: every dimension but the last keys the rows (in request order), and the last one becomes the columns. meta.columns.dynamic lists the remaining row dimension names, then the pivot column IDs, labelled by meta.columns.labels. Each cell is an object mapping every requested metric to its metric object.
Pivot rows are gap-filled with 0 cells: a date dimension returns every base check or period, model_type every engine the project runs; entities and types appear only when cited. A filled cell still carries the responses_count of its scope. limit and offset page the filled rows.
A pivot over an entity grain (url, domain, subdomain) requires url_ids[] or domain_ids[] and plots exactly those IDs — fetch the top rows first, then the pivot for those IDs. A pivot does not accept filters or sort_field, and a date dimension covers a single period: either returns 422.
A date dimension on its own returns, per check, the period’s distinct AI answers that carry at least one brand mention, as rows with a single ai_answers_count cell. These are totals across all sources (per-source counts never sum to them, since one answer cites several sources). Only ai_answers_count is supported there, and filters require a grain.
Every metric value is an object, not a plain number: { "values": [v0] } in single-period mode. Passing date_range[1] enables compare mode (requests without a date dimension only), which adds the second period and a delta object: { "values": [v0, v1], "delta": { "absolute": 15, "percentage": 50 } }. Deltas are returned only in compare mode.
subdomains_count is valid only when the grain is domain, the only grain that folds subdomains into a domain row. With any other grain it returns 422.

Request parameters

Path parameters

Parameter	Type	Required	Description
project_id	String (UUID)	Yes	UUID of the project.

Query parameters

Parameter	Type	Required	Default	Description
dimensions[]	Array of String	Yes	N/A	1 to 3 dimensions, no duplicates. Two or more pivot by the last one, in request order. Values: `check_id`, `check_date`, `check_week`, `check_month`, `url`, `domain`, `subdomain`, `page_type`, `domain_type`, `model_type`.
week_start	Integer	No	`1`	First weekday for a `check_week` dimension, from 0 to 6 (`0` Sunday, `6` Saturday).
metrics[]	Array of String	Yes	N/A	At least one metric, each returned as a metric object. Values: `ai_answers_count`, `citations_share`, `prompts_count`, `my_brand_mentions_count`, `competitors_count`, `responses_count`, `subdomains_count`. `responses_count` is the number of answers in the row’s scope (its period or check, and its engine under a `model_type` split) that carry at least one brand mention, cited or not — the `citations_share` denominator. It is the same for every entity or type in that scope and is not additive across them.
date_range	Array of Object	Yes	N/A	1 to 2 entries: base period at index `0`, optional compare period at index `1`. Each has `from` and `to` in `YYYY-MM-DD`. Each period aggregates all checks within its dates. See Date ranges and compare mode.
model_types	Array of String	No	N/A	Filter by AI model types (e.g. `chatgpt`).
country_codes	Array of String	No	N/A	Deprecated — use `location_ids`. Filter by ISO 3166-1 alpha-2 country codes. Resolved to the project’s live locations in those countries, so a deleted location is not matched, a country the project no longer tracks returns an empty result, and two locations in one country cannot be told apart. It keeps working for now; migrate to `location_ids` before it is withdrawn.
location_ids	Array of String (UUID)	No	N/A	Filter by the project’s tracked locations, using location IDs from List project locations. Multiple values combine as OR. An ID from another project or of a deleted location returns `422`. Combined with `country_codes`, only locations matching both are read.
topic_ids	Array of Integer	No	N/A	Filter by topic IDs.
sentiments	Array of String	No	N/A	Filter by sentiment: `positive`, `neutral`, `negative`.
exclude_branded	Boolean	No	N/A	When `true`, every number in the response is computed from non-branded prompts only. A prompt not yet classified counts as non-branded. Off when omitted.
prompt_ids	Array of Integer	No	N/A	Filter by prompt IDs.
req_num	Integer	No	N/A	Limit to the answers with this request number within a check.
search_query	String	No	N/A	Search query for filtering results.
filters	Array of Object	No	N/A	Up to 10 post-aggregation metric conditions, combined with AND. Rows failing any condition are dropped, and `meta.pagination.total` reflects that. Each condition has `metric` (a `metrics[]` value), `target` (`0` base period value, `1` compare period value, or `delta` absolute change; `1` and `delta` require `date_range[1]`), `operator` (`gt`, `gte`, `eq`, `lt`, `lte`), and `value` (a number). All four are required. Example: `filters[0][metric]=ai_answers_count&filters[0][target]=delta&filters[0][operator]=gt&filters[0][value]=0`.
url_ids[]	Array of Integer	No	N/A	Up to 50 IDs. Narrow the sources the rows aggregate to these `url_id` values, whatever the grain. Required for a pivot over an entity grain, which plots exactly these IDs.
domain_ids[]	Array of Integer	No	N/A	Up to 50 IDs. Narrow the sources to these `domain_id` values (as returned in the rows), whatever the grain. In the `subdomain` grain an ID selects exactly that subdomain; in the other grains a domain ID also matches every URL and subdomain under it.
types[]	Array of String	No	N/A	Up to 50 values. Narrow the sources to these page types (e.g. `listicle`, `blog`, `news`), whatever the grain. See the Reference for the full list.
domain_types[]	Array of String	No	N/A	Up to 50 values. Narrow the sources to these domain types (matches any element of a row’s `domain_type` array), whatever the grain: `you`, `competitors`, `corporate`, `ugc`, `editorial`, `reference`, `institutional`, `other`.
sort_field	String	No	First metric, descending	A requested metric name; sorts on period 0 values. Append `_delta` to sort by the absolute delta (e.g. `ai_answers_count_delta`), which requires `date_range[1]`, otherwise the request returns `422`.
sort_order	String	No	`desc`	Sort direction: `asc` or `desc`.
limit	Integer	No	`50`	Maximum rows to return, from 0 to 1000. `0` returns every row and requires two or more dimensions (`422` otherwise).
offset	Integer	No	`0`	Number of rows to skip (min `0`).

Request example

Copy

curl -X GET --globoff 'https://api.seranking.com/v1/se-visible/projects/550e8400-e29b-41d4-a716-446655440000/sources/metrics?dimensions[]=domain&metrics[]=ai_answers_count&metrics[]=citations_share&date_range[0][from]=2026-05-01&date_range[0][to]=2026-05-31' \

  -H 'Authorization: Token YOUR_API_KEY'

To chart selected domains per check date, pivot by domain and pass the domain IDs:

Copy

curl -X GET --globoff 'https://api.seranking.com/v1/se-visible/projects/550e8400-e29b-41d4-a716-446655440000/sources/metrics?dimensions[]=check_date&dimensions[]=domain&domain_ids[]=123&domain_ids[]=456&metrics[]=ai_answers_count&metrics[]=citations_share&date_range[0][from]=2026-01-01&date_range[0][to]=2026-01-31' \

  -H 'Authorization: Token YOUR_API_KEY'

Response parameters

If successful, the server returns a metrics object with pagination, column layout, and the data rows.

Parameter	Type	Description
meta	Object	Pagination and column layout.
meta.pagination.total	Integer	Total rows: the grain values, or the filled rows of a pivot.
meta.columns.primary	String	First dimension the rows are keyed by (one of the `dimensions[]` values).
meta.columns.dynamic	Array of String	With 1 dimension, the requested metric names. In a pivot, the remaining row dimension names, then the pivot column IDs (entity IDs, type values, or date labels, as strings).
meta.columns.labels	Object	Pivot only. Maps each pivot column ID to its display label (URL or domain name, or the value itself).
data	Array	Rows. With 1 dimension, each row is keyed by its own columns — `domain_id` and `domain` for the `domain` and `subdomain` grains; `url_id`, `url`, and `domain_id` for `url`; `page_type`, `domain_type`, or `model_type` for the type grains; the date label for a date dimension — plus one metric object per requested metric. In a pivot, each row holds the row dimensions’ columns, then one cell per pivot column keyed by its ID, mapping every requested metric to its metric object, with `0` cells where nothing was cited.

Response example

Domain grain, single period

Copy

{

  "meta": {

    "pagination": { "total": 2 },

    "columns": { "primary": "domain", "dynamic": ["ai_answers_count", "citations_share"] }

  },

  "data": [

    {

      "domain_id": 123,

      "domain": "example.com",

      "ai_answers_count": { "values": [45] },

      "citations_share": { "values": [75] }

    },

    {

      "domain_id": 456,

      "domain": "other.com",

      "ai_answers_count": { "values": [27] },

      "citations_share": { "values": [45] }

    }

  ]

}

Domain grain, compare mode

Copy

{

  "meta": {

    "pagination": { "total": 1 },

    "columns": { "primary": "domain", "dynamic": ["ai_answers_count"] }

  },

  "data": [

    {

      "domain_id": 123,

      "domain": "example.com",

      "ai_answers_count": {

        "values": [45, 30],

        "delta": { "absolute": 15, "percentage": 50 }

      }

    }

  ]

}

Pivot: check date by domain, gap-filled

Copy

{

  "meta": {

    "pagination": { "total": 2 },

    "columns": {

      "primary": "check_date",

      "dynamic": ["123", "456"],

      "labels": { "123": "example.com", "456": "other.com" }

    }

  },

  "data": [

    {

      "check_date": "2026-01-01",

      "123": {

        "ai_answers_count": { "values": [45] },

        "citations_share": { "values": [12.5] }

      },

      "456": {

        "ai_answers_count": { "values": [0] },

        "citations_share": { "values": [0] }

      }

    },

    {

      "check_date": "2026-01-08",

      "123": {

        "ai_answers_count": { "values": [41] },

        "citations_share": { "values": [11.9] }

      },

      "456": {

        "ai_answers_count": { "values": [3] },

        "citations_share": { "values": [0.9] }

      }

    }

  ]

}

Sources

Get project sources

Request parameters

Request example

Response parameters

Response example

Get aggregated source metrics

Request parameters

Request example

Response parameters

Response example

Learn how SE Ranking’s API can boost your SEO!

Request a free demo to see our tools and integrations in action