Hey all š I'm Peter, founder of Parsley. Saw discussion here about capturing signals from cold outreach. Building something that tackles this head-on. The problem: Cold prospects don't reply, but some do check you out. You have no idea what they cared about, what objections they're sitting on, or how ready they are to buy. What we're building: https://www.parsley.id ā a digital business card with an inbuilt AI chatbot that captures pre-call intelligence:
Intent scoring ā prospects auto-qualify themselves based on what they ask (pricing + competitor questions = hot lead, flagged for follow-up)
Pain point mapping ā see exactly which topics they care about: pricing, technical, integrations, comparisons
Knowledge gaps ā what questions couldn't be answered = objections you'll face on the call
Prospect enrichment ā extracts company, role, timeline, and budget context from the conversation itself
Basically: Gong tells you what happened on the call. Parsley tells you what they cared about before it. Looking for early users ā whether you're carrying a quota, managing a team or building the stack. Would love feedback from anyone willing to try it or poke holes in the idea. Live at parsley.id ā or DM me here :)
Hi Peter, curious what data or signals does the bot use to answer questions and infer intent. and how customizable is that per client? and how do you avoid hallucinated enrichment?
Hi Lana, it's a conversational chatbot and it uses semantic analysis to infer meaning. Essentially, you upload a bunch of documents and it behaves like a normal chatbot, but it's looking for pain points, buying signals, and other intent-level data. Why do you ask? What's on your mind?
Im familiar with semantic search and conversational retrieval. i was considering your tool for myself and trying to understand more clearly is how you handle hallucinations, especially when it comes to inference and enrichment from soft signals. Im also curious about the actual conversion impact from the chatbot in practice. Iām evaluating this for my business with horizontal processes rather than vertical, where intent patterns and documentation can be much less consistent. Does the product still perform well there, or is it primarily optimized for repeatable, vertical-specific business cases?
Thanks for the interest, but I'm heads-down on launch right now and don't have bandwidth for a deeper dive. Best of luck with your evaluation!
