TL;DR: There's no official API for the Google Ads Transparency Center. Google's free BigQuery dataset looks like one, but it only covers ads shown in the European Economic Area and Turkey, and it has no ad text or landing pages. I pulled 50 recent Shopify ads from the website with their country lists. About half were never shown in the EEA or Turkey, so they can't be in BigQuery at all. Code and numbers below.
"Is there an API?"
Someone asked exactly that on the Google Ads API forum on 15 July 2024. The Google Ads API team answered the same day: "there is no corresponding API for this yet". The Google Ads API itself only covers your own account. As of 18 November 2026 I found no newer announcement of a public API for the Transparency Center.
What Google does publish is a dataset.
What the BigQuery dataset has
bigquery-public-data.google_ads_transparency_center has two tables, creative_stats and removed_creative_stats. The README (last modified 18 March 2026) lists every column. The important line is the one for region_stats:
Collection of regions in which the ad served (in the European Economic Area and Turkey)
| You want | In creative_stats? |
|---|---|
| Advertiser ID, name, legal name, location, verification | Yes |
| Creative ID, format, link to the ad's page | Yes |
| First/last shown, impression ranges | Yes, EEA and Turkey only, impressions from 1 March 2023 |
| Audience targeting approach (demographics, geo, customer lists...) | Yes |
| Topic, Ad Grants flag, who paid | Yes |
| Ads shown only in the US, UK, India, Brazil... | No |
| Headline, description, landing page URL | No (no such columns) |
| Image or video of the ad | No (only the link to its page) |
Not "political ads only". That's a common mix-up: creative_stats holds commercial ads too. Political ads have their own download bundle, still online today.
What the gap looks like in practice
On 3 January 2025 a developer asked on Google's forum why BigQuery returned 3 creatives for an advertiser that shows about 200 on the website. The reply: the dataset covers the EEA or Turkey only, while the website covers everywhere.
I wanted to see how big that gap is for a normal global brand. On 18 November 2026 I pulled 50 Shopify ads shown in the last 30 days from the Transparency Center, with the list of countries each ad ran in, and checked each list against the 30 EEA countries plus Turkey.
| Run (18 Nov 2026) | Shown in the EEA or Turkey | Never shown there |
|---|---|---|
| 12:22 Kigali time | 25 | 25 |
| 13:16 Kigali time | 27 | 23 |
The two runs differ because Google returns a slightly different set of ads each time (it estimates 50,000 to 60,000 Shopify ads for this filter, and I took 50). Either way, about half.
Some details from the first run:
- The 50 ads ran in 140 countries between them.
- 5 ads ran only in the US, 2 only in Canada, 1 each only in India, Australia, Brazil, the Philippines and Singapore.
- Of the 9 ads shown in the US, 6 were never shown in Europe.
- For the 25 ads that did run in Europe, BigQuery would have a row, but only with the European regions. Their US or Indian numbers aren't in the schema.
One caveat: I didn't query BigQuery for these 50 creative IDs (that needs a Google Cloud project). The "can't be there" part follows from the README's definition and matches the forum example. If you have a project set up and want to check, the creative IDs are in the CSV.
The open-source route
The best-known Python package, GoogleAdsTransparencyScraper (42 stars), still gets 5,296 downloads a month on PyPI. Its last commit was on 23 July 2023. Open issues since then include "API endpoint not working" (August 2024), a domain search that returns nothing (February 2025) and "Error on Creative" (June 2025). It may still work for you, but expect to patch request formats yourself when the website changes.
A hosted scraper
Google Ads Transparency Scraper on Apify searches the website by advertiser name, advertiser ID or domain, with the country, format, platform and date filters the website has. With details on, each ad also gets the list of countries it ran in, any text Google shows, YouTube video IDs and image URLs.
Disclosure: I built this. It costs $0.001 per ad, plus $0.003 per ad for the details, on Apify's free plan (less on paid plans). Apify's free plan includes $5 of credit a month. The run below cost $0.20 at those prices.
With apify-client 3.x (pip install apify-client):
import os
from apify_client import ApifyClient
EEA_TR = set("AT BE BG HR CY CZ DK EE FI FR DE GR HU IE IT LV LT LU MT NL "
"PL PT RO SK SI ES SE IS LI NO TR".split())
client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("kuezi/google-ads-transparency-scraper").call(logger=None, run_input={
"searchTargets": ["shopify.com"],
"country": "anywhere",
"lastDays": 30,
"maxAdsPerTarget": 50,
"includeAdDetails": True, # adds shownInCountries, text, video IDs
})
inside, outside = 0, []
for ad in client.dataset(run.default_dataset_id).iterate_items():
countries = {c["country"] for c in ad["shownInCountries"]}
if countries & EEA_TR:
inside += 1
else:
outside.append((ad["format"], sorted(countries)))
print(f"shown in the EEA or Turkey: {inside}")
print(f"never shown there: {len(outside)}")
for fmt, countries in outside[:6]:
print(f" {fmt:5} {', '.join(countries[:8])}")
print(run.status_message)
Output of the 13:16 run (about 1.5 minutes):
shown in the EEA or Turkey: 27
never shown there: 23
TEXT PH
TEXT BD, CN, LK, PK, TH, VN
VIDEO CA
TEXT AR, BO, BR, CL, CO, CR, DO, EC
VIDEO CA
TEXT SA
Saved 50 ads from 1 target(s). shopify.com: 50 ads saved, Google estimates 50000-60000 for these filters; text 28, video 21, image 1; newest first shown 2026-11-03.
The last line is the run's status message, which is also what an AI agent sees when it calls the Actor through Apify's MCP server. To watch a competitor, set onlyNewAds, save the input as a task and schedule it daily. There's a ready-made task to compare several competitors.
Which one to use
| Need | Use |
|---|---|
| EU or Turkey ads, impression ranges, audience targeting, removed ads with reasons | BigQuery. It's official, and the scraper has no impression or targeting fields |
| Ads shown outside Europe | The website, by hand or with a scraper |
| Ad copy and creatives at scale | The website. Note the limit below |
| Political ads | Google's political ads bundle |
What the scraper doesn't give you
- Impressions and targeting. Those come from BigQuery, for Europe only.
- Clean ad text for every ad. Google shows many search ads as rendered images. In both Shopify runs only 2 of the 28 or 29 text ads came back with text, and none with a landing page. For the rest you get the image URLs and the link to the ad's page. Brands differ; check a small run first.
- Every ad. You get what the website lists for your filters, in the order Google returns it.
If Google ships a real API or widens the dataset, I'll update this post. If you've queried BigQuery for a non-EU brand and seen something different, post it in the comments.
AI disclosure: I wrote this with AI assistance. Every quote, number and output above was checked or run on 18 November 2026.
Top comments (0)