The Cross Aggregated Metrics endpoint in LLM Mentions API allows you to directly compare how multiple products or brands are mentioned across AI searches. Besides delivering blended data on all mentions for your target parameters, it also offers separate results into distinct groups each labeled with an aggregation_key. This endpoint is designed for benchmarking competing products or brands in the same category side by side.
In this article, we’ll walk you through how to set the request parameters for the Cross Aggregated Metrics endpoint, explain what each parameter means, and how to interpret the API response data.
Targeting in Cross Aggregated Metrics
To begin with, the Cross Aggregated Metrics endpoint uses a targets array allowing you to specify several target sets of parameters for comparison. Overall, you can specify up to 10, but not less than 2 target sets (arrays), each containing up to 10 entities (objects) with a domain or keyword and optional additional settings.
Understanding the target array and aggregation_key
Each individual target set must be specified along with an aggregation_key, which groups results for comparison and serves as a label for the group, like “samsung_galaxy”. The target array itself defines what mentions your request should search for.
Example:
{
"aggregation_key": "samsung_galaxy",
"target": [
{
"domain": "www.samsung.com",
"search_filter": "exclude"
},
{
"keyword": "Galaxy S25",
"match_type": "word_match"
}
]
}
You can add up to 10 entities (domains and/or keywords with optional additional parameters) per target, but at least one is required. In short, a domain entity allows you to include or exclude mentions from a specific site, while a keyword entity allows you to include or exclude mentions of a keyword, with options for how it’s matched. You can learn more about these settings in this article.
Domain entity example
{
"domain": "www.apple.com",
"search_filter": "include",
"search_scope": ["sources"]
}
Keyword entity example
{
"keyword": "iPhone",
"search_filter": "include",
"search_scope": ["question"],
"match_type": "word_match"
}
Example request for side-by-side product comparison
As an example, let’s compare three flagship smartphones across Google’s AI Overviews:
- For target group “iPhone”, we’ll exclude Apple’s official site and capture mentions of “iPhone 17” as in “iPhone 17 Pro” or “iPhone 17 Pro Max.”
- For “Samsung Galaxy”, we exclude Samsung’s official site and track whole-word mentions of “Galaxy S25.”
- For “Google Pixel”, let’s exclude Google’s store and allow partial matches for “Pixel 9,” so variations like “Pixel 9a” are also counted.
- To keep results meaningful, we also filter out low-volume mentions
["ai_search_volume", ">", 1000], and limit results to the top 5 per group. See the full list of available filters here.
Full request with multiple target sets
[
{
"platform": "google",
"targets": [
{
"aggregation_key": "iphone",
"target": [
{
"domain": "www.apple.com",
"search_filter": "exclude"
},
{
"keyword": "iPhone 17",
"match_type": "word_match"
}
]
},
{
"aggregation_key": "samsung_galaxy",
"target": [
{
"domain": "www.samsung.com",
"search_filter": "exclude"
},
{
"keyword": "Galaxy S25",
"match_type": "word_match"
}
]
},
{
"aggregation_key": "google_pixel",
"target": [
{
"domain": "store.google.com",
"search_filter": "exclude"
},
{
"keyword": "Pixel 9",
"match_type": "partial_match"
}
]
}
],
"initial_dataset_filters": [
[
"ai_search_volume",
">",
1000
]
],
"internal_list_limit": 5
}
]
Interpreting the API response
When you receive the API response, you will see that the data is organized into a few main sections: total and items.
{
"version": "0.1.20251006",
"status_code": 20000,
"status_message": "Ok.",
"time": "1.2775 sec.",
"cost": 0.101,
"tasks_count": 1,
"tasks_error": 0,
"tasks": [
{
"id": "10091354-8284-0636-0000-015022d79a3d",
"status_code": 20000,
"status_message": "Ok.",
"time": "1.2295 sec.",
"cost": 0.101,
"result_count": 1,
"path": [
"v3",
"ai_optimization",
"llm_mentions",
"cross_aggregated_metrics",
"live"
],
"data": {
"api": "ai_optimization",
"function": "cross_aggregated_metrics",
"platform": "google",
"targets": [
{
"aggregation_key": "iphone",
"target": [
{
"domain": "www.apple.com",
"search_filter": "exclude"
},
{
"keyword": "iPhone 17",
"match_type": "word_match"
}
]
},
{
"aggregation_key": "samsung_galaxy",
"target": [
{
"domain": "www.samsung.com",
"search_filter": "exclude"
},
{
"keyword": "Galaxy S25",
"match_type": "word_match"
}
]
},
{
"aggregation_key": "google_pixel",
"target": [
{
"domain": "store.google.com",
"search_filter": "exclude"
},
{
"keyword": "Pixel 9",
"match_type": "partial_match"
}
]
}
],
"initial_dataset_filters": [
[
"ai_search_volume",
">",
1000
]
],
"internal_list_limit": 5
},
"result": [
{
"total": {
"location": [
{
"type": "group_element",
"key": "2840",
"mentions": 1796,
"ai_search_volume": 8807400,
"impressions": 15818090400
},
{
"type": "group_element",
"key": "2356",
"mentions": 323,
"ai_search_volume": 1578600,
"impressions": 509887800
},
{
"type": "group_element",
"key": "2392",
"mentions": 298,
"ai_search_volume": 1358600,
"impressions": 404862800
},
{
"type": "group_element",
"key": "2826",
"mentions": 297,
"ai_search_volume": 924600,
"impressions": 274606200
},
{
"type": "group_element",
"key": "2124",
"mentions": 109,
"ai_search_volume": 310300,
"impressions": 33822700
}
],
"language": [
{
"type": "group_element",
"key": "en",
"mentions": 2414,
"ai_search_volume": 10850000,
"impressions": 26191900000
},
{
"type": "group_element",
"key": "es",
"mentions": 311,
"ai_search_volume": 1492800,
"impressions": 464260800
},
{
"type": "group_element",
"key": "ja",
"mentions": 298,
"ai_search_volume": 1358600,
"impressions": 404862800
},
{
"type": "group_element",
"key": "hi",
"mentions": 75,
"ai_search_volume": 300500,
"impressions": 22537500
},
{
"type": "group_element",
"key": "de",
"mentions": 62,
"ai_search_volume": 168300,
"impressions": 10434600
}
],
"platform": [
{
"type": "group_element",
"key": "google",
"mentions": 3378,
"ai_search_volume": 14982400,
"impressions": 50610547200
}
],
"sources_domain": [
{
"type": "group_element",
"key": "www.youtube.com",
"mentions": 7204,
"ai_search_volume": 10660600,
"impressions": 76798962400
},
{
"type": "group_element",
"key": "support.apple.com",
"mentions": 3789,
"ai_search_volume": 7025200,
"impressions": 26618482800
},
{
"type": "group_element",
"key": "discussions.apple.com",
"mentions": 969,
"ai_search_volume": 2460900,
"impressions": 2384612100
},
{
"type": "group_element",
"key": "www.reddit.com",
"mentions": 840,
"ai_search_volume": 2575700,
"impressions": 2163588000
},
{
"type": "group_element",
"key": "support.google.com",
"mentions": 580,
"ai_search_volume": 1443100,
"impressions": 836998000
}
],
"search_results_domain": null
},
"items": [
{
"key": "iphone",
"location": [
{
"type": "group_element",
"key": "2840",
"mentions": 1640,
"ai_search_volume": 7861900,
"impressions": 12893516000
},
{
"type": "group_element",
"key": "2356",
"mentions": 209,
"ai_search_volume": 1000800,
"impressions": 209167200
},
{
"type": "group_element",
"key": "2392",
"mentions": 270,
"ai_search_volume": 1257800,
"impressions": 339606000
},
{
"type": "group_element",
"key": "2826",
"mentions": 288,
"ai_search_volume": 836200,
"impressions": 240825600
},
{
"type": "group_element",
"key": "2124",
"mentions": 104,
"ai_search_volume": 301100,
"impressions": 31314400
}
],
"language": [
{
"type": "group_element",
"key": "en",
"mentions": 2101,
"ai_search_volume": 9123500,
"impressions": 19168473500
},
{
"type": "group_element",
"key": "es",
"mentions": 265,
"ai_search_volume": 1320800,
"impressions": 350012000
},
{
"type": "group_element",
"key": "ja",
"mentions": 270,
"ai_search_volume": 1257800,
"impressions": 339606000
},
{
"type": "group_element",
"key": "hi",
"mentions": 55,
"ai_search_volume": 228100,
"impressions": 12545500
},
{
"type": "group_element",
"key": "de",
"mentions": 52,
"ai_search_volume": 128200,
"impressions": 6666400
}
],
"platform": [
{
"type": "group_element",
"key": "google",
"mentions": 2909,
"ai_search_volume": 12588200,
"impressions": 36619073800
}
],
"sources_domain": [
{
"type": "group_element",
"key": "www.youtube.com",
"mentions": 2287,
"ai_search_volume": 9647100,
"impressions": 22062917700
},
{
"type": "group_element",
"key": "support.apple.com",
"mentions": 1642,
"ai_search_volume": 7013700,
"impressions": 11516495400
},
{
"type": "group_element",
"key": "discussions.apple.com",
"mentions": 596,
"ai_search_volume": 2455900,
"impressions": 1463716400
},
{
"type": "group_element",
"key": "www.reddit.com",
"mentions": 530,
"ai_search_volume": 2233800,
"impressions": 1183914000
},
{
"type": "group_element",
"key": "support.google.com",
"mentions": 327,
"ai_search_volume": 1394500,
"impressions": 456001500
}
],
"search_results_domain": null
},
{
"key": "samsung_galaxy",
"location": [
{
"type": "group_element",
"key": "2840",
"mentions": 92,
"ai_search_volume": 565100,
"impressions": 51989200
},
{
"type": "group_element",
"key": "2356",
"mentions": 89,
"ai_search_volume": 501100,
"impressions": 44597900
},
{
"type": "group_element",
"key": "2392",
"mentions": 14,
"ai_search_volume": 61100,
"impressions": 855400
},
{
"type": "group_element",
"key": "2826",
"mentions": 6,
"ai_search_volume": 64600,
"impressions": 387600
},
{
"type": "group_element",
"key": "2124",
"mentions": 2,
"ai_search_volume": 3700,
"impressions": 7400
}
],
"language": [
{
"type": "group_element",
"key": "en",
"mentions": 222,
"ai_search_volume": 1292100,
"impressions": 286846200
},
{
"type": "group_element",
"key": "es",
"mentions": 15,
"ai_search_volume": 46300,
"impressions": 694500
},
{
"type": "group_element",
"key": "ja",
"mentions": 14,
"ai_search_volume": 61100,
"impressions": 855400
},
{
"type": "group_element",
"key": "hi",
"mentions": 17,
"ai_search_volume": 85000,
"impressions": 1445000
},
{
"type": "group_element",
"key": "de",
"mentions": 9,
"ai_search_volume": 34700,
"impressions": 312300
}
],
"platform": [
{
"type": "group_element",
"key": "google",
"mentions": 321,
"ai_search_volume": 1774500,
"impressions": 569614500
}
],
"sources_domain": [
{
"type": "group_element",
"key": "www.youtube.com",
"mentions": 155,
"ai_search_volume": 757400,
"impressions": 117397000
},
{
"type": "group_element",
"key": "support.apple.com",
"mentions": 3,
"ai_search_volume": 9600,
"impressions": 28800
},
{
"type": "group_element",
"key": "discussions.apple.com",
"mentions": 2,
"ai_search_volume": 2600,
"impressions": 5200
},
{
"type": "group_element",
"key": "www.reddit.com",
"mentions": 42,
"ai_search_volume": 145400,
"impressions": 6106800
},
{
"type": "group_element",
"key": "support.google.com",
"mentions": 5,
"ai_search_volume": 7700,
"impressions": 38500
}
],
"search_results_domain": null
},
{
"key": "google_pixel",
"location": [
{
"type": "group_element",
"key": "2840",
"mentions": 93,
"ai_search_volume": 508500,
"impressions": 47290500
},
{
"type": "group_element",
"key": "2356",
"mentions": 56,
"ai_search_volume": 277500,
"impressions": 15540000
},
{
"type": "group_element",
"key": "2392",
"mentions": 25,
"ai_search_volume": 63900,
"impressions": 1597500
},
{
"type": "group_element",
"key": "2826",
"mentions": 4,
"ai_search_volume": 25400,
"impressions": 101600
},
{
"type": "group_element",
"key": "2124",
"mentions": 3,
"ai_search_volume": 5500,
"impressions": 16500
}
],
"language": [
{
"type": "group_element",
"key": "en",
"mentions": 159,
"ai_search_volume": 804200,
"impressions": 127867800
},
{
"type": "group_element",
"key": "es",
"mentions": 39,
"ai_search_volume": 145900,
"impressions": 5690100
},
{
"type": "group_element",
"key": "ja",
"mentions": 25,
"ai_search_volume": 63900,
"impressions": 1597500
},
{
"type": "group_element",
"key": "hi",
"mentions": 10,
"ai_search_volume": 29500,
"impressions": 295000
},
{
"type": "group_element",
"key": "de",
"mentions": 5,
"ai_search_volume": 23000,
"impressions": 115000
}
],
"platform": [
{
"type": "group_element",
"key": "google",
"mentions": 257,
"ai_search_volume": 1128400,
"impressions": 289998800
}
],
"sources_domain": [
{
"type": "group_element",
"key": "www.youtube.com",
"mentions": 109,
"ai_search_volume": 469800,
"impressions": 51208200
},
{
"type": "group_element",
"key": "support.apple.com",
"mentions": 1,
"ai_search_volume": 1900,
"impressions": 1900
},
{
"type": "group_element",
"key": "discussions.apple.com",
"mentions": 1,
"ai_search_volume": 2400,
"impressions": 2400
},
{
"type": "group_element",
"key": "www.reddit.com",
"mentions": 43,
"ai_search_volume": 223500,
"impressions": 9610500
},
{
"type": "group_element",
"key": "support.google.com",
"mentions": 11,
"ai_search_volume": 40900,
"impressions": 449900
}
],
"search_results_domain": null
}
]
}
]
}
]
}
Use “total” for a macro view, such as “Which countries/languages/platforms or domains dominate overall?”
The total object offers aggregated data across all targets combined (i.e., not split by aggregation_key). Inside it, you’ll see total mentions by location, language, platform, sources_domain, and search_results_domain. The sources_domain field shows top cited domains contributing to target mentions, while the search_results_domain (when available) reveals top domains from the model’s background retrieval.
The data on total aggregated mentions is useful for understanding the overall scale and top sources before comparing your stats and exploring each group individually.
Use “items” to compare targets side-by-side and dive deeper
The items array in the API response returns aggregated data for each target group labeled with your aggregation_key (iphone, samsung_galaxy, google_pixel). Each item has the same sections as in the total object, but scoped to a specific target.
For example, for the “aggregation_key”: "iphone", we have the location, language, platform, sources_domain, search_results_domain arrays with the same structure as under total, but only for that target. Each element inside these arrays is a group_element with:
key – the grouping value (e.g., 2840 for the US location, en for English language, google for Google AI Overview, www.youtube.com for top domain).
mentions, ai_search_volume, impressions – metrics for your target and that group (i.e. per location, language, domain, etc.).
Wrap up
The Cross Aggregated Metrics endpoint helps you move beyond individual analysis to a broader, comparative view of AI search visibility. It reveals how each brand or product performs relative to others, allowing you to see who dominates in mentions, impressions, and authority across AI search platforms, all with a single API request. You can use this endpoint to track competitive positioning, benchmark brand strength, and uncover emerging leaders in your market. To start configuring a request with your parameters, visit the endpoint docs.