Run segmentation query

Enterprise

This endpoint is part of Countly Enterprise. To get access, contact sales or compare versions. Existing customers can reach the support portal with questions.

Endpoint

/o?method=segmentation

Overview

Executes a Drill segmentation query with filters, period, bucket, and optional breakdown fields.

The response is not a single fixed schema. Drill returns different payload shapes depending on whether projectionKey is omitted, present, or combined with list=true.

Authentication

Pass api_key or auth_token as a query parameter, or send countly-token as a header. See Authentication.

Permissions

Requires drill Read permission.

Request Parameters

Parameter	Type	Required	Description
method	String	Yes	Must be segmentation.
app_id	String	Yes	Target app ID.
event	String	Yes	Event key to query.
queryObject	JSON String (Object)	Yes	Mongo-style Drill filter object. Use {} for no additional filters. Drill adds app, event, and time filters internally from app_id, event, and period.
period	String or Array	Yes	Period value accepted by Countly period parser, for example 30days, day, month, or a date range array.
bucket	String or JSON String (Array)	Yes	Time bucket or buckets. Supported values are hourly, daily, weekly, and monthly.
projectionKey	String or JSON String (Array)	No	Group-by field or fields. Examples: up.p, up.cc, sg.plan, ["up.p","sg.plan"].
list	Boolean/String	No	When truthy and projectionKey is provided, returns table/list-style segment rows with pagination metadata instead of graph-style time-series buckets.
limit	Number	No	Maximum segment rows returned for projected graph/table modes. Defaults to the Drill projection limit.
skip	Number	No	Offset for list=true table results.
sort	JSON String (Object)	No	Sort object for list=true table results, for example {"u":-1} or {"s0":1}.
sortGraphBy	String	No	Graph mode segment ordering. Use u to sort by users; default ordering is by total count t.
db_override	String	No	mongodb, clickhouse, or compare.
api_key	String	Conditional	Required if auth_token is not provided.
auth_token	String	Conditional	Required if api_key is not provided.

Examples

Basic session query

/o?method=segmentation&
  app_id=64f5c0d8f4f7ac0012ab3456&
  event=[CLY]_session&
  queryObject={}&
  period=30days&
  bucket=daily

Filter by user profile fields

/o?method=segmentation&
  app_id=64f5c0d8f4f7ac0012ab3456&
  event=[CLY]_session&
  queryObject={"up.cc":{"$in":["US","CA"]},"up.p":{"$in":["iOS"]}}&
  period=30days&
  bucket=daily

Filter by event segmentation

/o?method=segmentation&
  app_id=64f5c0d8f4f7ac0012ab3456&
  event=purchase&
  queryObject={"sg.plan":{"$in":["pro","enterprise"]},"s":{"$gte":20}}&
  period=30days&
  bucket=daily

Use logical operators in queryObject

/o?method=segmentation&
  app_id=64f5c0d8f4f7ac0012ab3456&
  event=purchase&
  queryObject={"$or":[{"up.cc":{"$in":["US"]}},{"sg.campaign":{"$in":["spring-sale"]}}]}&
  period=30days&
  bucket=daily

Query grouped by platform

/o?method=segmentation&
  app_id=64f5c0d8f4f7ac0012ab3456&
  event=[CLY]_session&
  queryObject={"up.cc":"US"}&
  projectionKey=up.p&
  period=30days&
  bucket=daily

Query grouped by multiple fields

/o?method=segmentation&
  app_id=64f5c0d8f4f7ac0012ab3456&
  event=purchase&
  queryObject={"sg.plan":{"$exists":true}}&
  projectionKey=["up.p","sg.plan"]&
  period=30days&
  bucket=daily&
  limit=20

Table/list mode for projected segments

/o?method=segmentation&
  app_id=64f5c0d8f4f7ac0012ab3456&
  event=purchase&
  queryObject={"sg.plan":{"$exists":true}}&
  projectionKey=sg.plan&
  list=true&
  period=30days&
  bucket=daily&
  limit=10&
  skip=0&
  sort={"u":-1}

Response

Success Response Shapes

Without projectionKey

Returns totals and bucketed time-series data. Each bucket contains period keys such as daily dates, weekly keys, or monthly keys.

Example shape

{
  "totals": {
    "u": 42,
    "t": 128,
    "s": 312.5,
    "dur": 980
  },
  "data": {
    "daily": {
      "2026.4.16": {
        "u": 12,
        "t": 30,
        "s": 75,
        "dur": 240
      }
    }
  },
  "buckets": ["daily"],
  "app_id": "64f5c0d8f4f7ac0012ab3456",
  "lu": "2026-04-17T09:10:11.000Z",
  "query_time": 37
}

With projectionKey

Returns totals, segment metadata, and bucketed data grouped by segment value. Segment labels are built from the projected values. Multiple projection keys are joined with |.

Example shape

{
  "totals": {
    "u": 42,
    "t": 128,
    "s": 312.5,
    "dur": 980
  },
  "segments": {
    "iOS": {
      "u": 22,
      "t": 70,
      "s": 180,
      "dur": 500,
      "s0": "iOS",
      "keys": {
        "up.p": "iOS"
      },
      "segment": "iOS"
    }
  },
  "data": {
    "daily": {
      "2026.4.16": {
        "iOS": {
          "u": 10,
          "t": 28,
          "s": 70,
          "dur": 210,
          "keys": {
            "up.p": "iOS"
          }
        }
      }
    }
  },
  "buckets": ["daily"],
  "page_data": {
    "total": 3,
    "limit": 10,
    "start": 1,
    "end": 3,
    "pages": 1,
    "curPage": 1
  }
}

With projectionKey and list=true

Returns table/list-style segment rows. This mode is used for segment tables and pagination. Time-series data can be empty or contain only initialized bucket objects, depending on backend and data availability.

Example shape

{
  "totals": {
    "u": 42,
    "t": 128,
    "s": 312.5,
    "dur": 980
  },
  "segments": {
    "iOS": {
      "u": 22,
      "t": 70,
      "s": 180,
      "dur": 500,
      "keys": {
        "up.p": "iOS"
      },
      "segment": "iOS"
    }
  },
  "page_data": {
    "total": 3,
    "limit": 10,
    "start": 1,
    "end": 3,
    "pages": 1,
    "curPage": 1
  },
  "data": {
    "daily": {}
  },
  "buckets": ["daily"]
}

Long-running query

If the same long-running query is already running, or if the request is pushed into the long-task path, the endpoint can return a task reference instead of final data.

{
  "task_id": "65f1f7b2ad5b9b001f12ab34"
}

Cached query

When a cached result is reused, the payload has the same result shape as the original query and includes:

{
  "cached": true
}

Response Fields

Field	Type	Description
totals	Object	Overall totals for the query. Common metric keys are u (unique users), t (total count), s (sum), and dur (duration).
data	Object	Bucketed graph data. Keys are requested bucket names such as daily, weekly, or monthly.
data.<bucket>.<period>	Object	Unprojected bucket totals for one period, or an object keyed by segment label when projectionKey is used.
segments	Object	Segment totals keyed by segment label. Present when projectionKey is used.
segments.<segment>.keys	Object	Maps each requested projectionKey to the segment value represented by that row.
page_data	Object	Pagination metadata for projected segment rows. Commonly present for projected/list results.
buckets	Array	Bucket values used by the query after parsing.
app_id	String	App id copied into newly computed responses.
lu	String/Date	Last update timestamp for newly computed or cached responses.
query_time	Number	Query execution time in milliseconds for newly computed responses.
cached	Boolean	Present and true when returning cached data.
task_id	String	Present instead of result data when the query is represented by a long task.

Error Responses

• 400

{
  "result": "Missing request parameter: event"
}

• 400

{
  "result": "Missing request parameter: queryObject"
}

• 400

{
  "result": "Bad request parameter: queryObject"
}

• 400

{
  "result": "Missing request parameter: period"
}

• 400

{
  "result": "Bad request parameter: period"
}

• 400

{
  "result": "Invalid request parameter: period"
}

• 400

{
  "result": "Missing request parameter: bucket"
}

• 400

{
  "result": "Bad request parameter: bucket"
}

• 500

{
  "result": "Error in processing: <details>"
}

Behavior

• Validates required query parameters.
• Parses and normalizes filter and period parameters.
• Parses queryObject from JSON string into an object.
• Adds ts period bounds internally.
• Adds app/event constraints internally before the backend query runs:
  • a is set from app_id
  • e is set from event
  • custom events are queried as e="[CLY]_custom" with n=<event>
• Runs segmentation query on selected backend.
• If projectionKey is provided, fetches segmentation field types and returns grouped segment results.
• If list=true is provided with projectionKey, uses table/list aggregation and pagination.
• With db_override=compare, runs both MongoDB and ClickHouse and returns comparison-selected output.

Query Object

queryObject is a Drill filter object encoded as JSON. The object uses stored Drill field names:

Prefix/field	Meaning	Examples
up.*	User profile fields stored with the event row.	up.p platform, up.cc country code, up.av app version
sg.*	Event segmentation fields.	sg.plan, sg.campaign, sg.ItemID
uid	Countly user/device identifier.	{"uid":{"$in":["device-1","device-2"]}}
c	Event count.	{"c":{"$gte":1}}
s	Event sum.	{"s":{"$gt":20}}
dur	Event duration.	{"dur":{"$gte":60}}
chr	Cohort filter, preprocessed by the Cohorts plugin when available.	{"chr":"65f1f7b2ad5b9b001f12ab34"}

Supported operators are Mongo-style operators accepted by the Drill backend, such as $in, $nin, $gt, $gte, $lt, $lte, $ne, $exists, $and, $or, and $nor.

Do not normally include a, e, n, or ts in queryObject; the endpoint derives those from app_id, event, and period.

Related Endpoints

• Query Metadata - Read
• Bookmarks - Read

Implementation details

Configuration Impact

Setting	Default	Affects	User-visible impact
api.*	Server API defaults	Shared API execution controls (for example processing thresholds/limits).	Changes to API-level controls can affect runtime behavior, limits, or response timing for this endpoint.
drill.*	Drill feature defaults	Drill query result shaping and list-size behavior.	Changes to drill settings can affect result size and query output details.

Database Collections

Collection	Used for	Data touched by this endpoint
countly_drill.drill_meta	Drill metadata model	Stores event/segment/property metadata dictionaries used by this endpoint.
countly_drill.drill_events	Drill event records	Stores granular event rows queried or updated by this endpoint.
countly_drill.drill_cache	Endpoint data source	Stores endpoint-related records this endpoint reads or modifies.
countly_drill.drill_snapshots	Endpoint data source	Stores endpoint-related records this endpoint reads or modifies.