Original Summary

I'm 22, building an AI startup called AskVerb, and we're barely 2 months in (including development). For the past few weeks, most of my time has gone into building features, fixing bugs, and making the product better. But I kept avoiding one question: Who the hell am I actually building this for? For context, AskVerb lets SaaS companies embed AI agents into their products so users can perform actual tasks instead of just chatting with a support bot. Initially, I thought I'd just scrape a few thousand SaaS companies, find their founders' emails, and start cold outreach. But the more I thought about it, the more I realized: A list of 5,000 random founders isn't 5,000 potential customers. So I spent almost 6 hours going back and forth with ChatGPT, refining my ICP and figuring out how to use Cowork to do the heavy lifting. Here's what I ended up with. 1. Narrowed my ICP down to three categories (to start with) Niche CRM and operations SaaS Scheduling and booking platforms Project and client management software I'm specifically looking for smaller, established SaaS companies with real users and complicated workflows. Not startups that launched yesterday with zero customers. 2. Created a qualification system out of 100 points Every company gets scored on: Active customers (20) Workflow complexity (20) Evidence of user pain (20) Decision-maker accessibility (15) Existing AI capabilities (10) Ability to pay (10) Integration feasibility (5) Then I rank them from S to D tier. The idea is to prioritize companies that have a genuine reason to need what I'm building, rather than just companies that happen to sell software. 3. Made Cowork research actual decision-makers One thing I was very particular about: I don't want generic company emails or random sales representatives. I want the actual founder, CEO, or CTO. Their publicly available professional emails, LinkedIn, X profiles, and business phone numbers where possible. If something isn't verifiable, it stays blank. No AI-generated guesses. 4. The most important column isn't the email address. It's a column called "Why AskVerb?" For every lead, Cowork needs to identify a specific workflow that our product could improve. Instead of sending: "Hey, we built an AI agent. Interested?" I want to send something like: "Hey, I noticed your users have to go through multiple steps to manage bookings. What if they could complete those workflows just by asking?" Obviously, the observation needs to be backed by actual research. 5. Everything goes into a structured Excel workbook Five sheets covering: Qualified companies and their scores Founders and decision-makers Evidence and source URLs Outreach tracking Qualification rules The initial goal is 100 qualified SaaS companies across the US, UK, Australia, and Canada. The Cowork research is underway, so I don't have results to share yet. And I have no idea whether this will actually convert

  • 情报分类：商业与市场研究
  • 分类依据：内容涉及商业、投资或市场动态
  • 信息来源：Reddit · SaaS
  • 发布时间：2026/10/9 21:26:04