API reference · Platforms, status and usage

Featured ads with history and creative features for the website (public, cached 6 hours)

GET/public/showcase/featured

Up to 10 ads of the showcase advertisers that have archived media and text, varied across platforms, preferring ads with published reach or impressions. The selection changes once a day (UTC). Read from our archive only.

Example request

Shell

curl "https://api.adlibapi.com/public/showcase/featured" \
  -H "X-API-Key: $ADLIB_API_KEY"

Send your key in the X-API-Key header. Try it in the explorer.

Response

captured_atrequired

datetime

–

data.adrequired

object

–

advertiser

object

–

countries

string[]

ISO 3166-1 alpha-2

creatives

object[]

–

derived

object

–

eu_data

object

–

extra

map of object

Platform-specific fields

first_seen_by_us

datetime

–

first_shown

datetime

–

formats

string[]

–

id

string

Stable ID: <platform>:<platform_ad_id>

impressions_range

object

–

last_seen_by_us

datetime

–

last_shown

datetime

–

library_url

string

–

placements

string[]

–

platform

string

One of meta, google, tiktok, linkedin, pinterest, x, snapchat, apple, amazon, microsoft

platform_ad_id

string

–

political

boolean

Political or issue ad. Meta only; null where the platform doesn't label it

raw_ref

string

Pointer to the archived raw upstream record

removed_from_library_at

datetime

When the platform's library stopped showing this ad (taken down, deleted by the advertiser or past the library's retention). We keep it; null while the library still has it

spend_range

object

–

status

string

One of active, inactive, unknown

data.featuresrequired

object

As GET /v1/ads/{id}/features; null while not analysed yet

cta

string

–

format

string

From the measured media: vertical_video, horizontal_video, square_video, vertical_image, square_image, horizontal_image, carousel, text_only

headline

string

–

landing

object

–

media

object[]

–

method

string

rules = text rules with evidence, measurement = ffprobe/ffmpeg, ocr = tesseract; no AI model

ocr_text

string

–

text

object

–

version

string

–

data.features_analyzed_atrequired

datetime

–

data.historyrequired

object

–

reach

object[]

EU reach per day, oldest first (Meta ads delivered in the EU)

status_periods

object[]

–

version_list

object[]

–

versions

integer

Creative versions we recorded (GET /v1/ads/{id}/history → versions)

dayrequired

string

UTC day the selection belongs to