Story Sourcing Sanghralay

for Seekho · the scored story bank & the beat-level Composer
Internet-sourced stories. Source = whole web, or Reddit only (optionally a specific community, e.g. personalfinanceindia). Refresh from Web runs search (Tavily → Google) → enrich → score (≥4/5 auto-lands in the Library). Reload re-reads the bank from disk.
Pull recent Ringg transcripts, score them; ≥4/5 auto-lands in the Library.
Download CSV
Contact
Status
Duration
Cost
Agent
Listen · Read · Analyse · Rate · Ad
Loading calls…
Serving a PII-stripped snapshot of the User Research sheet — last 6 months, text > 280 chars, capped at 100 (52 stories). “Score sheet → Library” enriches & scores them into the Library (needs ANTHROPIC_API_KEY).
Click “Reload sheet”.
📤

Self-uploaded — coming soon

Upload your own call recordings or transcripts (audio/CSV/text) to score and add to the story bank. Wiring this up after the Web + User-calls pipeline is live.

📸

Instagram — coming soon

Pull candidate stories from Instagram (captions / reels transcripts) into the same enrich-and-score pipeline. Planned after Self-uploaded.

Build a composite story by picking the strongest contributor for each beat. The Tier-1 realism checker runs against the composed record before you ship.

Source beats (from library)

Effort & Tenure long effort + doubt
Specificity Anchor name, place, occupation, numbers
Fall & Stake concrete catastrophe + why it hurts
Wrong-Lever Diagnosis the reframe + right Seekho lever

Ad-side beats (you write)

Discovery how they found Seekho's right lever
Delayed Payoff specific result after trying it
Mirror CTA why you (viewer) should try too

Preview

Pick contributors and write the ad-side beats. Composed narrative appears here.

Story

✕

Story → Ad

✕

Transcript

✕

Custom Analysis

✕

Story-to-Ad Rating

✕

Story-to-Ad Script

✕