Insights API

Provides a single, consistent interface to retrieve ad statistics.

Breakdowns - Group results
Action Breakdowns - Understanding the response from action breakdowns.
Async Jobs - For requests with large results, use asynchronous jobs
Limits and Best Practices - Call limits, filtering and best practices.

Before you can get data on your ad's performance, you should set up your ads to track the metrics you are interested in. For that, you can use URL Tags, Meta Pixel, and the Conversions API.

Before you begin

You will need:

The ads_read permission.
An app. See Meta App Development for more information.

Campaign Statistics

To get the statistics of a campaign's last 7 day performance:

curl -G \
  -d "date_preset=last_7d" \
  -d "access_token=ACCESS_TOKEN" \
  "https://graph.facebook.com/API_VERSION/AD_CAMPAIGN_ID/insights"

To learn more, see the Ad Insights Reference.

Making Calls

The Insights API is available as an edge on any ads object.

API Method
`act_<AD_ACCOUNT_ID>/insights`
`<CAMPAIGN_ID>/insights`
`<ADSET_ID>/insights`
`<AD_ID>/insights`

Request

You can request specific fields with a comma-separated list in the fields parameters. For example:

curl -G \
-d "fields=impressions" \
-d "access_token=ACCESS_TOKEN" \
"https://graph.facebook.com/v22.0/<AD_ID>/insights"

Response

{
  "data": [
    {
      "impressions": "2466376",
      "date_start": "2009-03-28",
      "date_stop": "2016-04-01"
    }
  ],
  "paging": {
    "cursors": {
      "before": "MAZDZD",
      "after": "MAZDZD"
    }
  }
}

Levels

Aggregate results at a defined object level. This automatically deduplicates data.

Request

For example, get a campaign's insights on ad level.

curl -G \
-d "level=ad" \
-d "fields=impressions,ad_id" \
-d "access_token=ACCESS_TOKEN" \
"https://graph.facebook.com/v22.0/CAMPAIGN_ID/insights"

Response

{
  "data": [
    {
      "impressions": "9708",
      "ad_id": "6142546123068",
      "date_start": "2009-03-28",
      "date_stop": "2016-04-01"
    },
    {
      "impressions": "18841",
      "ad_id": "6142546117828",
      "date_start": "2009-03-28",
      "date_stop": "2016-04-01"
    }
  ],
  "paging": {
    "cursors": {
      "before": "MAZDZD",
      "after": "MQZDZD"
    }
  }
}

If you don't have access to all ad objects at the requested level, the insights call returns no data. For example, while requesting insights with level set to ad, if you don't have access to one or more ad objects under the ad account, this API call will return a permission error.

Attribution windows

The conversion attribution window provides timeframes that define when we attribute an event to an ad on a Meta app. For background information, see Meta Business Help Center, About attribution windows. We measure the actions that occur when a conversion event occurs and look back in time 1-day and 7-days. To view actions attributed to different attribution windows, make a request to /{ad-account-id}/insights. If you do not provide action_attribution_windows we use 7d_click and provide it under value.

For example specify action_attribution_windows and 'value' is fixed at 7d_click attribution window. Make a request to act_10151816772662695/insights?action_attribution_windows=['1d_click','1d_view'] and get this result:

"spend": 2352.45,
"actions": [
{
"action_type": "link_click",
"value": 6608,
"1d_view": 86,
"1d_click": 6510
},
"cost_per_action_type": [
{
"action_type": "link_click",
"value": 0.35600030266344,
"1d_view": 27.354069767442,
"1d_click": 0.36135944700461
},

// if attribution window is _not_ specified in query. And note that the number under 'value' key is the same even if attribution window is specified.
// act_10151816772662695/insights
"spend": 2352.45,
"actions": [
{
"action_type": "link_click",
"value": 6608
},
"cost_per_action_type": [
{
"action_type": "link_click",
"value": 0.35600030266344
},

Field Expansion

Request fields at the node level and by fields specified in field expansion.

Request

curl -G \
-d "fields=insights{impressions}" \
-d "access_token=ACCESS_TOKEN" \
"https://graph.facebook.com/v22.0/AD_ID"

Response

{
  "id": "6042542123268",
  "name": "My Website Clicks Ad",
  "insights": {
    "data": [
      {
        "impressions": "9708",
        "date_start": "2016-03-06",
        "date_stop": "2016-04-01"
      }
    ],
    "paging": {
      "cursors": {
        "before": "MAZDZD",
        "after": "MAZDZD"
      }
    }
  }
}

Sorting

Sort results by providing the sort parameter with {fieldname}_descending or {fieldname}_ascending:

Request

curl -G \
-d "sort=reach_descending" \
-d "level=ad" \
-d "fields=reach" \
-d "access_token=ACCESS_TOKEN" \
"https://graph.facebook.com/v22.0/AD_SET_ID/insights"

Response


{
  "data": [
    {
      "reach": 10742,
      "date_start": "2009-03-28",
      "date_stop": "2016-04-01"
    },
    {
      "reach": 5630,
      "date_start": "2009-03-28",
      "date_stop": "2016-04-03"
    },
    {
      "reach": 3231,
      "date_start": "2009-03-28",
      "date_stop": "2016-04-02"
    },
    {
      "reach": 936,
      "date_start": "2009-03-29",
      "date_stop": "2016-04-02"
    }
  ],
  "paging": {
    "cursors": {
      "before": "MAZDZD",
      "after": "MQZDZD"
    }
  }
}

Ads Labels

Stats for all labels whose names are identical. Aggregated into a single value at an ad object level. See the Ads Labels Reference for more information.

Request

curl -G \  
-d "fields=id,name,insights{unique_clicks,cpm,total_actions}" \
-d "level=ad" \
-d 'filtering=[{"field":"ad.adlabels","operator":"ANY", "value":["Label Name"]}]'  \
-d 'time_range={"since":"2015-03-01","until":"2015-03-31"}' \
-d "access_token=ACCESS_TOKEN" \
"https://graph.facebook.com/v22.0/AD_OBJECT_ID/insights"

Response

{
  "data": [
    {
      "unique_clicks": 74,
      "cpm": 0.81081081081081,
      "total_actions": 49,
      "date_start": "2015-03-01",
      "date_stop": "2015-03-31",
    },
  ], 
  "paging": {
    "cursors": {
      "before": "MA==",
      "after": "MA==",
    }
  }
}

Clicks definition

To better understand the click metrics that Meta offers today, please read the definitions and usage of each below:

Link Clicks, actions:link_click - The number of clicks on ad links to select destinations or experiences, on or off Meta-owned properties. See Ads Help Center, Link Clicks
Clicks (All), clicks - The metric counts multiple types of clicks on your ad, including certain types of interactions with the ad container, links to other destinations, and links to expanded ad experiences. See Ads Help Center, Clicks(All)

Deleted and Archived Objects

Ad units may be DELETED or ARCHIVED. The stats of deleted or archived objects appear when you query their parents. This means if you query impressions at the ad set level, results include impressions from all ads in the set it, regardless of whether the the ads are in a deleted or archived state. See also, Storing and Retrieving Ad Objects Best Practice.

However, if you query using filtering, status filtering will be applied by default to return only Active objects. As a result, the total stats of the parent node may be greater than the stats of its children.

You can get the stats of ARCHIVED objects from their parent nodes though, by providing an extra filtering parameter.

Request

To get the stats of all ARCHIVED ads in an ad account listed one by one:

curl -G \
  -d "level=ad" \
  -d "filtering=[{'field':'ad.effective_status','operator':'IN','value':['ARCHIVED']}]" \
  -d "access_token=<ACCESS_TOKEN>" \
  "https://graph.facebook.com/v22.0/act_<AD_ACCOUNT_ID>/insights/"

Response

Note that only archived objects are returned in this response.

{
  "data": [
    {
      "impressions": "1741",
      "date_start": "2016-03-11",
      "date_stop": "2016-03-12"
    }
  ],
  "paging": {
    "cursors": {
      "before": "MAZDZD",
      "after": "MAZDZD"
    }
  }
}

Deleted Objects Insights

You can query insights on deleted objects if you have their IDs or by using the ad.effective_status filter.

Request

For example, if you have the ad set ID:

curl -G \
-d "fields=id,name,status,insights{impressions}" \
-d "access_token=ACCESS_TOKEN" \
"https://graph.facebook.com/v22.0/AD_SET_ID"

In this example, we query with ad.effective_status:

POST https://graph.facebook.com/<VERSION>/act_ID/insights?access_token=token&appsecret_proof=proof&fields=ad_id,impressions&date_preset=lifetime&level=ad&filtering=[{"field":"ad.effective_status","operator":"IN","value":["DELETED"]}]

Response

{
  "id": "6042147342661",
  "name": "My Like Campaign",
  "status": "DELETED",
  "insights": {
    "data": [
      {
        "impressions": "1741",
        "date_start": "2016-03-11",
        "date_stop": "2016-03-12"
      }
    ],
    "paging": {
      "cursors": {
        "before": "MAZDZD",
        "after": "MAZDZD"
      }
    }
  }
}

Troubleshooting

Timeouts

The most common issues causing failure at this endpoint are too many requests and time outs:

On /GET or synchronous requests, you can get out-of-memory or timeout errors.
On /POST or asynchronous requests, you can possibly get timeout errors. For asynchronous requests, it can take up to an hour to complete a request including retry attempts. For example if you make a query that tries to fetch large volume of data for many ad level objects.

Recommendations

There is no explicit limit for when a query will fail. When it times out, try to break down the query into smaller queries by putting in filters like date range.
Unique metrics are time consuming to compute. Try to query unique metrics in a separate call to improve performance of non-unique metrics.

Rate Limiting

The Meta Insights API utilizes rate limiting to ensure an optimal reporting experience for all of our partners. For more information and suggestions, see our Insights API Limits & Best Practices.

Discrepancy with Ads Manager

Beginning June 10, 2025, to reduce discrepancies with Meta Ads Manager, use_unified_attribution_setting and action_report_time parameters will be disregarded and API responses will mimic Ads Manager settings:

Attributed values will be based on Ad-Set-level attribution settings (similar to use_unified_attribution_setting=true), and inline/on-ad actions will be included in 1d_click or 1d_view attribution window data.
Actions will be reported using action_report_time=mixed: on-Meta actions (like Link Clicks) will use impression-based reporting time; whereas off-Meta actions (like Web Purchases) will leverage conversion-based reporting time.

The default behavior of the API is different from the default behavior in Ads Manager. If you would like to observe the same behavior as in Ads Manager, please set the field use_unified_attribution_setting to true.

Learn More

Any endpoints not in the above list are not covered in this API. If you plan to include reports from Meta in your solution, see Meta Platform Terms and Developer Policies for Marketing API.