meta_ads_get_insights
Get Meta Ads Insights
Description
Retrieve performance metrics from Meta Ads with date presets/custom ranges, time grouping, entity-level aggregation, dimensional breakdowns, and filtering. Always fetches fresh data. Provide date_preset or time_range. For standard full-funnel analysis, prefer meta_ads_get_performance_report; use this for custom fields, hourly/creative-asset breakdowns, video metrics, or the \
Usage
{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "meta_ads_get_insights",
"arguments": {
"account_id": "act_123456789",
"date_preset": "last_7d",
"action_attribution_windows": [
"7d_click"
],
"reason": "Isolate 7-day click attribution conversions for accurate ROAS measurement"
}
}
}
hopkin meta insights get
| Flag | Type | Required | Description |
|---|---|---|---|
--account | string | Required | Meta ad account ID |
--date-preset | string | Optional | Predefined date range: today, yesterday, last_3d, last_7d, last_14d, last_28d, last_30d, last_90d, this_month, last_month, this_quarter, last_quarter, this_year, last_year, lifetime, maximum |
--time-range | object | Optional | Custom date range {since, until} in YYYY-MM-DD |
--time-increment | object | Optional | Time grouping: 1=daily, 7=weekly, or "monthly" |
--level | string | Optional | Aggregation level: account, campaign, adset, ad |
--fields | array | Optional | Metrics to retrieve (defaults to standard set) |
--breakdowns | array | Optional | Dimensions to segment data by. Pass multiple values in a single call to get cross-tabulated rows (e.g. ["age","gender"] → one row per "35-44 / female" segment). Do NOT make separate calls for each dimension. Available: age, gender, country, region, dma, device_platform, publisher_platform, platform_position, impression_device, frequency_value, place_page_id, product_id, ad_format_asset, body_asset, call_to_action_asset, description_asset, image_asset, link_url_asset, title_asset, video_asset, user_segment_key. Note: `frequency_value` is compatible only with `fields: ["reach"]` and cannot be combined with other breakdowns or with action_breakdowns; the tool rejects incompatible calls. Note: `user_segment_key` is the Advantage+ Shopping Campaigns "Audience Segments" breakdown. Returned values: prospecting, engaged, existing, unknown (lowercase). Cross-tabulation with demographic breakdowns like age is rejected by Meta's API — combine with country if needed; use separate calls for demographic splits. |
--action-breakdowns | array | Optional | Action breakdown dimensions: action_type, action_target_id, action_destination, action_reaction, action_video_sound, action_video_type, action_carousel_card_id, action_carousel_card_name |
--filtering | array | Optional | Filters as [{field, operator, value}] |
--action-attribution-windows | array | Optional | Attribution windows for action/conversion metrics. Controls which click/view windows are reported. Common: ["7d_click"] for 7-day click only. Default (when omitted): ["7d_click","1d_view"]. Available: 1d_click, 7d_click, 28d_click, 1d_view, 7d_view, 28d_view, 1d_ev, default |
--limit | integer | Optional | Maximum number of rows per page (default: 100, max: 500) |
--cursor | string | Optional | Pagination cursor from previous response |
{
"mcpServers": {
"meta-ads": {
"url": "https://meta.mcp.hopkin.ai",
"transport": "sse"
}
}
}
- 7-day click only conversions
- Age and gender breakdown
- Publisher platform breakdown
- Video asset metrics
- Frequency-bucket reach
- ASC audience-segment split (prospecting vs engaged vs existing vs unknown)
Parameters
| Name | Type | Required | Description |
|---|---|---|---|
account_id |
string |
Required | Meta ad account IDminLength: 1 |
reason |
string |
Required | Why this tool call is neededminLength: 1, maxLength: 500 |
Optional parameters (11)
| Name | Type | Required | Description |
|---|---|---|---|
date_preset |
string |
Optional | Predefined date range: today, yesterday, last_3d, last_7d, last_14d, last_28d, last_30d, last_90d, this_month, last_month, this_quarter, last_quarter, this_year, last_year, lifetime, maximumtoday yesterday this_week_mon_today this_week_sun_today last_week_mon_sun last_week_sun_sat this_month last_month this_quarter last_quarter this_year last_year last_3d last_7d last_14d last_28d last_30d last_90d lifetime maximum |
time_range |
object |
Optional | Custom date range {since, until} in YYYY-MM-DD |
time_increment |
object |
Optional | Time grouping: 1=daily, 7=weekly, or "monthly" |
level |
string |
Optional | Aggregation level: account, campaign, adset, adaccount campaign adset ad |
fields |
array |
Optional | Metrics to retrieve (defaults to standard set) |
breakdowns |
array |
Optional | Dimensions to segment data by. Pass multiple values in a single call to get cross-tabulated rows (e.g. ["age","gender"] → one row per "35-44 / female" segment). Do NOT make separate calls for each dimension. Available: age, gender, country, region, dma, device_platform, publisher_platform, platform_position, impression_device, frequency_value, place_page_id, product_id, ad_format_asset, body_asset, call_to_action_asset, description_asset, image_asset, link_url_asset, title_asset, video_asset, user_segment_key. Note: `frequency_value` is compatible only with `fields: ["reach"]` and cannot be combined with other breakdowns or with action_breakdowns; the tool rejects incompatible calls. Note: `user_segment_key` is the Advantage+ Shopping Campaigns "Audience Segments" breakdown. Returned values: prospecting, engaged, existing, unknown (lowercase). Cross-tabulation with demographic breakdowns like age is rejected by Meta's API — combine with country if needed; use separate calls for demographic splits. |
action_breakdowns |
array |
Optional | Action breakdown dimensions: action_type, action_target_id, action_destination, action_reaction, action_video_sound, action_video_type, action_carousel_card_id, action_carousel_card_name |
filtering |
array |
Optional | Filters as [{field, operator, value}] |
action_attribution_windows |
array |
Optional | Attribution windows for action/conversion metrics. Controls which click/view windows are reported. Common: ["7d_click"] for 7-day click only. Default (when omitted): ["7d_click","1d_view"]. Available: 1d_click, 7d_click, 28d_click, 1d_view, 7d_view, 28d_view, 1d_ev, default |
limit |
integer |
Optional | Maximum number of rows per page (default: 100, max: 500)min: 1, max: 500 |
cursor |
string |
Optional | Pagination cursor from previous responseminLength: 1 |
Examples
7-day click only conversions
{
"account_id": "act_123456789",
"date_preset": "last_7d",
"action_attribution_windows": [
"7d_click"
],
"reason": "Isolate 7-day click attribution conversions for accurate ROAS measurement"
}
hopkin meta insights get --account act_123456789 --date-preset last_7d --action-attribution-windows 7d_click
Age and gender breakdown
{
"account_id": "act_123",
"date_preset": "last_7d",
"breakdowns": [
"age",
"gender"
],
"reason": "Demographic breakdown not available in performance report"
}
hopkin meta insights get --account act_123 --date-preset last_7d --breakdowns age,gender
Publisher platform breakdown
{
"account_id": "act_123",
"date_preset": "last_7d",
"level": "adset",
"breakdowns": [
"publisher_platform",
"platform_position"
],
"reason": "Placement breakdown by publisher and position"
}
hopkin meta insights get --account act_123 --date-preset last_7d --level adset --breakdowns publisher_platform,platform_position
Video asset metrics
{
"account_id": "act_123",
"date_preset": "last_7d",
"level": "ad",
"fields": [
"video_p25_watched_actions",
"video_p50_watched_actions",
"video_p75_watched_actions",
"video_p100_watched_actions",
"video_avg_time_watched_actions"
],
"reason": "Video completion funnel for creative analysis"
}
hopkin meta insights get --account act_123 --date-preset last_7d --level ad --fields video_p25_watched_actions,video_p50_watched_actions,video_p75_watched_actions,video_p100_watched_actions,video_avg_time_watched_actions
Frequency-bucket reach
{
"account_id": "act_123",
"date_preset": "last_30d",
"breakdowns": [
"frequency_value"
],
"fields": [
"reach"
],
"reason": "Distribution of unique users by exposure frequency (1×, 2×, 3+); frequency_value is reach-only per Meta"
}
hopkin meta insights get --account act_123 --date-preset last_30d --breakdowns frequency_value --fields reach
ASC audience-segment split (prospecting vs engaged vs existing vs unknown)
{
"account_id": "act_123",
"date_preset": "last_7d",
"level": "campaign",
"breakdowns": [
"user_segment_key"
],
"filtering": [
{
"field": "campaign.objective",
"operator": "EQUAL",
"value": "OUTCOME_SALES"
}
],
"reason": "Weekly WoW report for Advantage+ Shopping Campaign — split spend/conversions by acquisition (prospecting) vs retargeting (engaged) vs existing customers"
}
hopkin meta insights get --account act_123 --date-preset last_7d --level campaign --breakdowns user_segment_key --filtering [object Object]