Viktor.comAn AI coworker that actually does the work
Promoted
Sentor
Entity-level sentiment analysis, built for AI assistants
9 followers
Entity-level sentiment analysis, built for AI assistants
9 followers
Sentor scores sentiment against specific things (a feature, a competitor, shipping), not just one overall positive/negative label per review. Connect 20+ sources like Trustpilot, Amazon, and the App Store in the dashboard, or use our API, JS/Python SDKs, and MCP server to run entity sentiment, clustering, and topic naming in your own product or directly from Claude, Cursor, or Windsurf. Free plan, no card required.
Would love to see a way to compare sentiment trends side by side across competitors, like a chart that shows how my product stacks up against two or three others on the same features over time. That would make the data way more actionable for positioning.
@yldzxkpp Yıldız, this is genuinely the exact feature we already have marked "coming soon" in the UI: side-by-side sentiment trends across competitors on the same features. Good instinct; that's precisely the use case we built it around. Comparison is what actually makes the data useful for positioning instead of just a number in isolation.
The reason it's not live yet is a real tradeoff, not just a roadmap placeholder: a genuinely useful comparison needs real, comparable data about each competitor, not just re-scoring your own reviews against a different entity, so it's a bigger data problem than it first looks like. With the launch happening now, we made the call to get the core experience solid first and come back to this right after, rather than ship it half-baked under launch pressure.
Appreciate you flagging it specifically; good signal we're prioritizing the right thing next.
Honestly the aspect-based sentiment is what got me, like pulling out what people actually feel about shipping versus the product itself is super useful for our team. Took like five minutes to wire up the API too.
@fahrikscj Really glad the entity-based split landed for you; that's exactly the gap we built this to close. "4.2 stars" tells you nothing about whether it's shipping or the product itself dragging the number down. And genuinely good to hear the API took five minutes; that's the actual bar we hold ourselves to. If it takes longer than that, we consider it a bug in the docs, not just the code.
Curious what you're building on top of it; happy to help if there's anything specific you're trying to pull out beyond shipping vs. product.