Stats API v2 reference
Plausible Stats API v2 is a powerful single endpoint HTTP interface to view historical and real-time stats. In a nutshell, the endpoint /api/v2/query
accepts both simple and complex stats queries in the POST request body and returns the metrics as JSON.
Take a look at our Events API Reference if you want to record pageviews or custom events for your sites, or Sites API Reference if you want to manage your sites over the API.
Authentication
You can obtain an API key for your account by going to your user settings page plausible.io/settings.
After creating a token, you can authenticate your request by sending the token in the Authorization header of your request.
Example curl request
In the following request, replace YOUR-TOKEN with a reference to your token and site_id value with your domain.
curl \
--request POST \
--header 'Authorization: Bearer YOUR-TOKEN' \
--header 'Content-Type: application/json' \
--url 'https://plausible.io/api/v2/query' \
--data '{ "site_id": "dummy.site", "metrics": ["visitors"], "date_range": "7d" }'
API keys have a rate limit of 600 requests per hour by default. If you have special needs for more requests, please contact us to request more capacity.
Request structure
/api/v2/query
endpoint accepts a query
object. Example:
{
"site_id": "dummy.site",
"metrics": ["visitors", "pageviews", "bounce_rate"],
"date_range": "7d",
"filters": [
["is_not", "visit:country_name", [""]]
],
"dimensions": ["visit:country_name", "visit:city_name"]
}
Query can contain the following keys:
site_id REQUIRED
Domain of your site on Plausible to be queried.
date_range REQUIRED
Date range to be queried.
Option | Description |
---|---|
["2024-01-01", "2024-07-01"] | Custom date range (ISO8601) |
["2024-01-01T12:00:00+02:00", "2024-01-01T15:59:59+02:00"] | Custom date-time range (ISO8601) |
"day" | Current day (e.g. 2024-07-01) |
"7d" | Last 7 days relative to today |
"30d" | Last 30 days relative to today |
"month" | Since the start of the current month |
"6mo" | Last 6 months relative to start of this month |
"12mo" | Last 12 months relative to start of this month |
"year" | Since the start of this year |
"all" | Since the start of stats in Plausible |
metrics REQUIRED
Metrics represent values to be calculated with the query.
Valid metrics are:
Metric name | Description |
---|---|
visitors | The number of unique visitors |
visits | The number of visits/sessions |
pageviews | The number of pageview events |
views_per_visit | The number of pageviews divided by the number of visits. Returns a floating point number. |
bounce_rate | Bounce rate percentage |
visit_duration | Visit duration in seconds |
events | The number of events (pageviews + custom events). When filtering by a goal, this metric corresponds to "Total Conversions" in the dashboard. |
percentage | The percentage of visitors of total who fall into this category: Requires: dimension list |
conversion_rate | The percentage of visitors who completed the goal. Requires: dimension list passed, an event:goal filter or event:goal dimension |
group_conversion_rate | The percentage of visitors who completed the goal with the same dimension. Requires: dimension list passed, an event:goal filter or event:goal dimension |
dimensions optional
Default: []
List of dimensions to group by. See example
Dimensions are attributes of your dataset. Using them in queries enables analyzing and compare multiple groups against each other.
Think of them as GROUP BY
in SQL.
Event dimensions
Valid dimensions include:
Dimension | Example | Description |
---|---|---|
event:goal | Register | A custom action that you want your users to take. To use this property, you first need to configure some goals in the site settings, or via the Sites API. The value is the goal's display_name . Learn more about goals here. |
event:page | /blog/remove-google-analytics | Pathname of the page where the event is triggered. You can also use an asterisk to group multiple pages (/blog* ) |
event:hostname | example.com | Hostname of the event. |
Mixing session metrics bounce_rate
, views_per_visit
and visit_duration
with event dimensions is not allowed.
Visit dimensions
Values of these dimensions are determined by the first pageview in a session.
Dimension | Example | Description |
---|---|---|
visit:entry_page | /home | Page on which the visit session started (landing page). |
visit:exit_page | /home | Page on which the visit session ended (last page viewed). |
visit:source | Visit source, populated from an url query parameter tag (utm_source , source or ref ) or the Referer HTTP header. | |
visit:referrer | t.co/fzWTE9OTPt | Raw Referer header without http:// , http:// or www. . |
visit:utm_medium | social | Raw value of the utm_medium query param on the entry page. |
visit:utm_source | Raw value of the utm_source query param on the entry page. | |
visit:utm_campaign | profile | Raw value of the utm_campaign query param on the entry page. |
visit:utm_content | banner | Raw value of the utm_content query param on the entry page. |
visit:utm_term | keyword | Raw value of the utm_term query param on the entry page. |
visit:device | Desktop | Device type. Possible values are Desktop , Laptop , Tablet and Mobile . |
visit:browser | Chrome | Name of the browser vendor. Most popular ones are Chrome , Safari and Firefox . |
visit:browser_version | 88.0.4324.146 | Version number of the browser used by the visitor. |
visit:os | Mac | Name of the operating system. Most popular ones are Mac , Windows , iOS and Android . Linux distributions are reported separately. |
visit:os_version | 10.6 | Version number of the operating system used by the visitor. |
visit:country | US | ISO 3166-1 alpha-2 code of the visitor country. |
visit:region | US-MD | ISO 3166-2 code of the visitor region. |
visit:city | 4347778 | GeoName ID of the visitor |
visit:country_name | United States | Name of the visitor country. |
visit:region_name | California | Name of the visitor region. |
visit:city_name | San Francisco | Name of the visitor city. |
Time dimensions
It's useful to be able to group data by time, which can be done via the following dimensions.
Dimension | Example | Description |
---|---|---|
time | 2024-01-01 | Time or date to group by. Automatically figures out the appropriate time:bucket value from date range. Response is a valid ISO8601 date or timestamp |
time:hour | 2021-01-27T23:43:10Z | Time grouped by hour. ISO8601 timestamp |
time:day | 2021-01-27 | Time grouped by date. ISO8601 date |
time:week | 2021-01-04 | Time grouped by start of the week. ISO8601 date |
time:month | 2021-01-01 | Time grouped by start of month. ISO8601 date |
Note that:
time
dimensions are not usable in filters. Setdate_range
instead.- If no data falls into a given time bucket, no values are returned. See
include.time_labels
option for a workaround.
Custom properties
Custom properties can also be used as dimensions with the form event:props:<custom_prop_name>
. See example
filters optional
Default: []
Filters allow limiting the data analyzed in a query. See example.
Simple filters
Each simple filter takes the form of [operator, dimension, clauses]
.
operators
The following operators are currently supported:
Operator | Example | Explanation |
---|---|---|
is | ["is", "visit:country_name", ["Germany", "Poland"]] | Sessions originating from Germany or Poland. |
is_not | ["is_not", "event:page", ["/pricing"]] | Events that did not visit /pricing page |
contains | ["contains", "event:page", ["/login"]] | Events visited any page containing /login |
contains_not | ["contains_not", "event:page", ["docs", "pricing"]] | Events that did not visit any page containing docs or pricing |
matches | ["matches", "event:page", ["^/user/\d+$"]] | Events where page matches regular expression ^/user/\d+$ . Uses re2 syntax |
matches_not | ["matches", "event:page", ["^/user/\d+$"]] | Events where page does not match regular expression ^/user/\d+$ . Uses re2 syntax |
dimension
Event and visit dimensions are valid for filters.
Note that only is
operator is valid for event:goal
dimension.
clauses
List of values to match against. A data point matches filter if any of the clauses matches.
Logical operations
Filters can be combined using and
, or
and not
operators.
Operator | Example | Explanation |
---|---|---|
and | ["and", [["is", "visit:country_name", ["Germany"]]], ["is", "visit:city_name", ["Berlin"]]]] | Sessions originating from Berlin, Germany |
or | ["and", [["is", "visit:country_name", ["Germany"]]], ["is", "visit:city_name", ["Tallinn"]]]] | Sessions originating from Germany or city of Tallinn |
not | ["not", ["is", "visit:country_name", ["Germany"]]] | Sessions not originating from Germany |
Note that top level filters is wrapped in an implicit and
.
order_by optional
Allows for custom ordering of query results.
List of tuples [dimension_or_metric, direction]
, where:
dimension_or_metric
needs to be listed in querymetrics
ordimensions
respectively.direction
can be one of"asc"
or"desc"
For example:
[["visitors", "desc"], ["visit:country_name", "asc"]]
When not specified, the default ordering is:
- If a time dimensions is present,
[time_dimension, "asc"]
- By the first metric specified, descending.
include optional
Default: {}
Additional options for the query as to what data to include.
include.imports
Default: false
If true, tries to include imported data in the result. See imported stats for more details, query example.
Read more on limitations of including imported data
Using custom property dimensions (event:props:*
) are only supported for 2 properties: url
and path
. Additionally, these breakdowns will only work in combination with a certain subset of event:goal
filters.
Filtering imported stats
Filtering by imported data is limited. The general rule is that you cannot filter by two different properties at the same time.
For example, event:page==/;visit:source==Twitter
is not able to return any imported results. The same happens when you try to filter by one dimension and set another as a dimension.
There are some exceptions though. The following dimensions are aggregated and grouped together and can be combined in a query:
- Countries, regions, cities
- Operating systems and their versions
- Hostnames and pages
- Specific custom events and their properties
Outbound Link: Click
andFile Download
goals with theurl
property404
goals with thepath
property
For example, you can set a country
dimension and filter by both city
and region
.
If set, meta.imports_included
field will be set as a boolean.
If the applied combination of filters and dimensions is not supported for imported stats, the results are still returned based only on native stats. Additionally, meta.imports_skip_reason
and meta.imports_warning
response fields will contain more information on why including imported data failed. See example
include.time_labels
Default: false
Requires a time
dimension being set. If true, sets meta.time_labels
in response containing all
time labels valid for date_range
.
include.total_rows
Default: false
Should be used for pagination. If true, sets meta.total_rows
in response containing the total number of
rows for this query.
pagination optional
Default: { "limit": 10000, "offset: 0 }
Response structure
Example response:
{
"results": [
{"metrics": [99, 98, 94], "dimensions": ["Estonia", "Tallinn"]},
{"metrics": [98, 82, 97], "dimensions": ["Brazil", "São Paulo"]},
{"metrics": [97, 77, 98], "dimensions": ["Germany", "Berlin"]},
{"metrics": [94, 86, 93], "dimensions": ["Italy", "Rome"]},
{"metrics": [89, 77, 96], "dimensions": ["United States", "San Francisco"]},
{"metrics": [82, 78, 92], "dimensions": ["Poland", "Warsaw"]}
],
"meta": {},
"query": {
"site_id": "dummy.site",
"metrics": ["visitors", "pageviews", "bounce_rate"],
"date_range": ["2024-09-04T00:00:00+00:00", "2024-09-10T23:59:59+00:00"],
"filters": [["is_not", "visit:country_name", [""]]],
"dimensions": ["visit:country_name", "visit:city_name"],
"order_by": [["visitors", "desc"]],
"include": {}
}
}
results
Results is an ordered list query results.
Each result row contains:
dimensions
- values for eachdimension
listed in query. In the same order as querydimensions
, empty if no dimensions requested.metrics
- List of metric values, in the same order as querymetrics
meta
Meta information about this query. Related: include.imports and include.time_labels.
query
The query that was executed, after manipulations performed on the backend.