What datasets does Telegram Ads Spy publish?

Telegram Ads Spy publishes 23 CC-BY-4.0 datasets covering Telegram advertising: ad creatives, channel directory, advertiser registry, niche taxonomy, country breakdowns, cashier creatives (Stars / invoice / giveaway), mini-app leaderboard, glossary, and ecosystem aggregates. JSON, CSV and RSS distributions. Updated continuously — datasets reflect the live archive, not a stale snapshot.

Can I use these datasets in academic research or AI training?

Yes — CC-BY-4.0 license (Creative Commons Attribution 4.0 International) explicitly permits commercial use, derivative works, model training and academic citation. Required attribution: tgadsspy.com (or the specific dataset URL). Stable permalinks: dataset URLs do not move once published. Schema.org Dataset + DataDownload markup is exposed so Google Dataset Search indexes the catalogue.

What's the right citation format?

For academic papers: "tgadsspy.com (2026). . https://tgadsspy.com . CC-BY-4.0." For journalism: link directly to the dataset URL and mention "data from tgadsspy.com, CC-BY-4.0." For models / derivative datasets: include tgadsspy.com in the data card and link to /datasets. Stable slugs guarantee that citations stay live indefinitely.

How fresh is the data?

Live — every download reflects the current state of the archive. New creatives typically appear within 30 minutes of being served on Telegram. Channels are refreshed nightly (top-1000) or weekly (long tail). The /api/v1/stats endpoint exposes lastCreativeAt as a pipeline-freshness probe, and dataset endpoints inherit that cadence.

Do you offer bulk downloads or only individual datasets?

Individual datasets are downloadable as JSON / CSV / RSS directly from each dataset URL (rate-limited at the Free tier, unrestricted on Pro / Enterprise). Bulk downloads of the entire archive (multi-GB Parquet exports, full historical creative gallery) are available on the Enterprise tier — contact hello@tgadsspy.com.

Open Datasets · CC-BY-4.0

Telegram Ads public data — free to download, cite and reuse.

Every dataset is a live slice of the open ad archive. Attribution required (CC-BY-4.0), no registration needed. Stable URLs — safe to cite in academic papers or reports.

84,005

Ad creatives

45,014

Advertisers

600,996

Channels tracked

Cashier creatives

4,289

Mini-app bots

Ad Creatives — CSVCSVCC-BY-4.0Updated: Live
All non-hidden ad creatives: creative ID, title, text snippet, niche, geo, advertiser domain, payment source, first/last seen dates, CTA URL.
→
Ad Creatives — JSONJSONCC-BY-4.0Updated: Live
Same dataset as JSON. Includes media hash, accent color, channel slug and full advertiser object for cross-referencing.
→
Telegram Channels — CSVCSVCC-BY-4.0Updated: Live
Public Telegram channels with audience and trust signals: username, title, subscriber count, language, verified / bot / scam flags, saved Stars-gifts count, channel URL. Filter by language, minimum members or verified-only; sort by audience or recency. The only open, citeable dataset of the Telegram channel landscape.
→

How to cite

When using this data in publications, please attribute as follows:

tgadsspy.com (2024–). Telegram Ads Open Archive. Retrieved from https://adsguru.org/datasets. CC-BY-4.0.

For questions about bulk access, API keys, or academic data agreements — contact via @tgadsspybot.

Available datasets

How to cite