Free Shadow Reach Analyzer - Know what people say in untagged Instagram & Tiktok videos
byβ’
Enter a brand, topic, or celebrity and our AI watches Instagram and TikTok videos to surface every mention that never made it into a caption or tag. See the reach you're actually getting, the conversations no metadata-based report will ever show.
Free run and no sign-up required.


Replies
Oriane
Pythagora
@yuri_mihaileanu1Β Such a solid idea and really nicely presented. Wishing you a great launch day! π
Oriane
@leon_ostrezΒ thank you man, expect another tool from what's happening in videos on monday!
I'm launching 10 free tools in 10 weeks. This one was #2.
cool concept honestly, the structured intelligence layer sounds super useful. one thing that would help though is letting teams set custom alerts for specific hooks or niches they care about, like basically a watchlist with notifications when something matches. would save a lot of manual checking.
Oriane
@phyllis_pearsonΒ Extremely interesting. It's definitely something worth considering.
You don't need to rely heavily on this tool though. Everything is based on @Oriane's API. You could get even more tailored results for your business if you decide to run the queries directly. This might instantly meet your needs, even the watchlist system you can think about on top of the data. Because once you have the data, you're free to build whatever fits you best.
Oriane
Hey PH! Thibaut here, CTO of Oriane.
@yuri_mihaileanu1 covered the "why". I'll add a bit of the "how," for anyone curious about what's actually running under the hood.
The Shadow Reach Analyzer is a thin app sitting on top of our core API. What it does concretely: you give it a brand, topic, or celebrity, and it pulls the relevant videos from TikTok and Instagram, runs them through our video intelligence search engine (transcription + visual analysis, not just captions or hashtags), and scores each mention with reasoning and a direct link back to the source clip as evidence.
The interesting part technically is that we're not doing metadata search. Most tools stop at what's written around a video (captions, tags, hashtags). We're processing the actual audio and visual content at scale, which is a very different (and much harder) problem: transcription, object/brand recognition, semantic matching, all running across millions of videos continuously.
Happy to go deep on any tech topic. Ask away π