Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 28 Next »

Overview

The Sharethrough Reporting API offers publishers a programmatic way to access performance data. It's a HTTP API that follows a RESTful design pattern. You can retrieve results in a CSV or JSON format.

Note: If you are unable to access this data through an API connection, please contact your Account Manager who can give you a URL to access these fields in a UI. From there, you can download your own reports. 

If you have access to the current API, your token is valid for this V2 Supply API Endpoint. If you need a new token, please reach out to support@sharethrough.com or through your Sharethrough contact.

Authentication

The Sharethrough Reporting API uses API Keys to authenticate requests. Authentication to the API is performed via HTTP Basic Auth.

Be sure to not share your API key to keep your data secure. Also, all API requests must be made over HTTPS. Calls made over plain HTTP will fail. API requests without authentication will also fail.

API Endpoint

Method

POST: https://reporting-api.sharethrough.com/v2/programmatic/supply

Required Header

Authorization: Bearer $TOKEN_STRING

Optional Header

Accept: text/csv (if you want results in CSV format, for JSON just omit the header)

Request Body

Parameter

Required?

Description

Example

startDate

Required

Start date of the data

"2020-03-01"

endDate

Required

End date of the data

"2020-03-01"

publishers

Required

Publishers you wish to pull data for

["publisherKey1", "publisherKey2"]

fields

Required

Metric fields you wish to pull

["wins", "clicks"]

groupBy

Optional

Add in case you wish to group your data by any dimensions listed.

["publisher_name", "site_name"] 

filterBy

Optional.

The filterBy section is a place where you can specify filters (up to 3) for your API call. See the Filtering Results section for more information on how to use filters.

[ {field: "publisher_name", value: "Test Pub", operator: "EQUALS"}, {field: "site_name", value: "Some site name", operator: "NOT_EQUALS"} ]

timezone

Optional

Must be a valid IANA timezone. If not specified, results are returned by UTC timezone.

(list here)

"America/Chicago"

debug

Optional

Requests also return timestamp and a query ID that can be shared with Sharethrough for debugging purposes

"debug": true

Filter object 

If you decide to use filters for your API call, you first specify a filterBy section in your request payload, and within that, you create filters. Each filter should be shaped like this (table below). See the Filtering Results section further down for more information on how to use filtering with your API calls.

operator

Optional.

When not specified, only matching results are returned - that is, the operator is assumed to be “EQUALS”.

Possible values for this property are

  • “EQUALS”

  • “NOT_EQUALS”

  • “CONTAINS”

  • “IN”

  • “NOT_IN”

field

Required

This is the name of the dimension field on which you wish to filter.

value

Required

Evaluated against the field in manner determined by filter.

Results

  • Format: We support JSON and CSV Outputs

  • Date Range: Historical information is available up for 1 year

  • All numbers are reported in time zone UTC (+00:00), unless the 'timezone' attribute is specified in the request body

  • Data available in the API is processed hourly. For a full day's worth of data, we recommend pulling next day.

HTTP Response Codes

Sharethrough uses conventional HTTP response codes to indicate the success or failure of an API request.

HTTP Code

Definition

Conditions when used

200

General Success

Successful 

400

Bad Request

The API cannot process this request. This is usually because the request payload is malformed (i.e. missing fields, or improper/non-existent fields or filters specified)

401

Unauthorized

No valid API key provided

403

Forbidden

The API key doesn’t have permissions to perform the request

404 

Not found

The requested resource doesn’t exist

500, 502, 503, 504

Server errors

Something went wrong on Sharethrough’s end

Dimensions

The table below describes the types of dimensions that can be accessed through the API. You can group and/or filter by one or more (up to 6) of the following values:

Dimension

API Representation

Notes

Publisher Name

publisher_name

 Name of the publisher

DSP Name

dsp_name

DSP name.

Site

site_name

Collection of placements that often correspond to a website or application. Example: "Cats.com"

Primary Domain

primary_domain

The primary domain associated with the incoming web request.

Notable Domain

notable_domain

The notable domain associated with the incoming web request. Example: “www.cats.com

Country

country

 Two or three letter code abbreviation for the country. Example: "US"

Placement Name

placement_name

 Name given to the placement or ad slot.

Placement Key

placement_key

 Integration key associated with incoming request.  Example: "bQRKnWWz6KnFqsGmttQmCoGP"

Creative Type

creative_type

Possible values for filtering are:"article", "banner", "carousel", "clickout", "native-outstream", "native-video", "outstream", "scroller", "slideshow".

Device Type

device_type

 Possible values for filtering are:

  • "Connected Device"

  • "Connected TV"

  • "Mobile/Tablet"

  • “Personal Computer"

  • "Phone"

  • “Set Top Box”

  • “Tablet”

Date

date

Date which the impression was first requested.  Example: "2020-08-10"

Hour

hour

Hour an impression was requested. Possible values 1-24.

Inventory Type

inventory_type

Use this dimension to group or filter results by particular media formats.  Possible values for this dimension are

  • "display"

  • "video"

  • "native"

Supply Integration

supply_name

The name of the supply integration path to filter by. There are many possible values to supply as filters for this property. Below is a partial list of examples:

  • "Google OB"

  • "Index Exchange"

  • "Microsoft"

  • "Prebid Client-side"

  • "Prebid Server"

  • "Tag on Page"

  • "Verizon"

Type 

site_type

The type of the site (i.e. app or web). Possible values are

  • “AppProperty”

  • “WebProperty”

App Bundle ID

bundle_id

The bundle ID for a given app. This is a string whose format varies across providers (e.g. “com.something.else” for Android, “1234567” for iOS, etc.)

OS Type 

os_type

The operating system of the user, as determined by user agent. Possible values are

  • “Android”

  • “Chrome OS”

  • “MacOS”

  • “Windows”

  • “iPadOS”

  • “iPhoneOS”

Placement Size

placement_size

Size of the placement when set up by STR (display only placements)

Adelaide AU Group

adelaide_au_group

Use this field to break out results by Adelaide Attention Score. Attention Score can take the values of “High”, “Medium”, or “Low”. Note: not available when cpm or dsp_name are included as fields in the request payload.

Adverifai Score Bucket

adverifai_score_bucket

Use this field to break out results by Adverifai’s score bucket. Score buckets are basically score ranges that increment by 0.05 and go from 0 to 1. So, for example, one site might have a score of 0.15, while another might be 0.22. These two entries would be broken out into separate buckets.

Advertiser Domain

advertiser_domain

Use this field to break out results as Advertiser name coming from the DSP and set up in the DSP. Note: not available when cpm or dsp_name are included as fields in the request payload.

Jounce Directness

jounce_directness

Use this field to break out results by directness as defined by Jounce. Possible values are

  • Direct

  • Outsourced Yield Management

  • Proprietary Placement

  • Content Syndication

  • Rebroadcasting

  • Non-Transparent

  • Not Classified

Jounce SPO Segment

jounce_spo_segment

Use this field to break out results by SPO Segment as defined by Jounce. Possible values are

  • Premium Publisher-Controlled

  • Premium Intermediary-Controlled

  • Multi-Hop Reselling

  • Made For Advertising

  • Cheap Reach

The value is [not_in_jounce_table] when the combination of Publisher Key, Bundle ID App Only and Primary Domain is not classified by Jounce

Open vs Deal

open_vs_deal

Use this field to break out results by whether they are “open” or a “deal”. Open means that the advertiser served impression on STX without targeting a deal. Deal means that the advertiser served impression on STX by targeting a deal. Note: not available when cpm or dsp_name are included as fields in the request payload.

Fields (Metrics)

The table below describes the header bidding metrics that can be accessed through the API: At least one field is required in the request:

Field

API Representation

Description

Impression Requests

impression_requests

Total impression requests that the site sends to Sharethrough.

Server Impressions 

server_impressions

When a bid wins our internal auction. For the majority of our supply we need to send this winning bid to participate in the publisher's downstream Header Bidding auction against other exchanges.

Header Bidding Wins

wins

Sharethrough wins the impression in the publisher ad server/HB wrapper auction and a signal is sent to the page, the impression is loaded but the assets are not yet rendered. 

Rendered Impressions

rendered_impressions

The served ad creative is rendered on the page. This is the payable event for publishers.

Viewable Impressions

viewable_impressions

The ad comes at least 50% in view for at least 1 second.

Earnings

pub_earnings

Programmatic only. The estimated amount of revenue you’ve earned through the Sharethrough Exchange (STX)

CPM

cpm

The eCPM for the entries returned by the query.

Clicks

clicks

The total number of times users clicked on ads

CTR

ctr

Click-through rate (= clicks /rendered impressions)

 

The percentage of rendered impressions that a user clicks on 

Fill Rate

fill_rate

= (rendered impressions/impressions requests)*100

 The percentage of ad calls being filled with a creative for a given placement 

Viewability Rate

viewability_rate

= viewable impressions/ rendered impressions

The percentage of paid impressions that are viewable

Render Rate

render_rate

= (rendered impressions/ received impressions)*100

 Once an impression is won and loaded, the percentage of times the assets are actually appearing/rendered on a page.

Video Starts

video_starts

When 50% of the video placement is in view (no minimum time) and there is no error, the player starts

Video Start Rate

video_start_rate

(Video starts/Rendered impressions)*100

The percentage of videos that come into view and start once the final auction is won and the ad is rendered.

Video Views (3+ seconds)

video_views_3secs

How many users reached 3 seconds during the video

Video Views (10+ seconds)

video_views_10secs

How many users reached 10 seconds during the video

Video Views (15+ seconds)

video_views_15secs

How many users reached 15 seconds during the video

Video Views (30+ seconds)

video_views_30secs

How many users reached 30 seconds during the video

25% Content Completion

video_completion_25

How many users reached 25% of the video

50% Content Completion

video_completion_50

How many users reached 50% of the video

75% Content Completion

video_completion_75

How many users reached 75% of the video

95% Content Completion

video_completion_95

How many users reached 95% of the video

Filtering Results

In addition to specifying which fields you would like in your returned data, you can also further refine your results by filtering them.

You do this by adding one or more filters to a filterBy section in your request payload.  Filters and the filterBy section are completely optional.

Simple Example - filter data to only show results involving video inventory

In this example, we use a single filter to limit what we get to earnings and cpm that involve video inventory:

{
  "startDate":"2024-01-15",
  "endDate":"2024-01-15",    
    "fields": [
        "pub_earnings",
        "cpm"
    ],
    "groupBy": [
        "inventory_type"
    ],
    "filterBy": [{
        "field": "inventory_type",
        "value": "video",
        "operator": "EQUALS"
    }],
    "publishers": [
        "YOUR_PUB_KEY_HERE"
    ]
}

More Complex Example - multiple filters

We can further refine results by adding additional filters.  Here, we add a second one that further states we’re only interested in video-related results for a hypothetical site:

{
  "startDate":"2024-01-15",
  "endDate":"2024-01-15",    
    "fields": [
        "pub_earnings",
        "cpm"
    ],
    "groupBy": [
        "inventory_type"
    ],
    "filterBy": [{
        "field": "inventory_type",
        "value": "video",
        "operator": "EQUALS"
    },{
        "field": "site_name",
        "value": "some_hypothetical_site",
        "operator": "EQUALS"
    }],
    "publishers": [
        "YOUR_PUB_KEY_HERE"
    ]
}

Be sure to include all filters within the brackets ([ and ]) that follow filterBy.

Note that when we combine filters, we are “AND-ing” them together.  In other words: in the example above, the API call will only return results where BOTH filter conditions are true (result is video-related AND it’s for the “some_hypothetical_site” site).

Further ways of filtering

In the examples above, we created filters using the “EQUALS” value for the operator of each filter, which matches results to exactly whatever was specified for the value property of the filter. 

We can also filter in other ways.

NOT_EQUALS

Use this to limit results to anything that does NOT have exactly this value.  

Example - show us results for anything BUT “some_hypothetical_site”:

{
  “field”: “site_name”,
  “value”: “some_hypothetical_site”,
  “operator”: “NOT_EQUALS”
}

CONTAINS

Use this to limit results to anything that has the letters specified. 

Note that filtering with CONTAINS is case-sensitive.

Example - show us results where ‘foo’ is in the name of the site_name field:

{
  “field”: “site_name”,
  “value”: “foo”,
  “operator”: “CONTAINS”
}

Remember that because the CONTAINS filter operates on a case-sensitive basis, results will include sites with names like “foobar” and “barfoo”, but will NOT contain “Foobar” or “barFoo” (“foo” != “Foo”).

IN

Use this as a quick way to specify multiple values you want to specify for your results.  You do this by including all possible values within a pair of brackets ([ and ]).

Example - show us results for US, Canada and Australia:

{
  “field”: “country”,
  “value”: ["US","CA","AU"],
  “operator”: “IN”
}

NOT_IN works in much the same way, except that any values included in the brackets are excluded from results.

Example - show us results for anyplace BUT the US, Canada and Australia:

{
  “field”: “country”,
  “value”: ["US","CA","AU"],
  “operator”: “NOT_IN”
}

Token Management

Please reach out to support@sharethrough.com if you need a token activated or deactivated. 

Example Request & Response

Request

{
    "publishers": ["1234567"],
    "startDate": "2020-02-16",
    "endDate": "2020-02-18",
    "groupBy": ["date"],
    "fields": ["rendered_impressions", "viewable_impressions", "clicks", "pub_earnings"],
}

Response

{
    "results": [
        {
            "DATE": "2020-02-17",
            "RENDERED_IMPRESSIONS": 3265224,
            "VIEWABLE_IMPRESSIONS": 1553567,
            "CLICKS": 8334,
            "PUB_EARNINGS": 43966411
        },
        {
            "DATE": "2020-02-18",
            "RENDERED_IMPRESSIONS": 3064240,
            "VIEWABLE_IMPRESSIONS": 1519001,
            "CLICKS": 7723,
            "PUB_EARNINGS": 43069528
        },
        {
            "DATE": "2020-02-16",
            "RENDERED_IMPRESSIONS": 3146531,
            "VIEWABLE_IMPRESSIONS": 1412046,
            "CLICKS": 7106,
            "PUB_EARNINGS": 443284816
        }
    ]
}
  • No labels