Harvestr

Hi PH community 👋 Today, I’m beyond excited to announce the public release of our most advanced AI product yet. Introducing HAI 🤖—your AI Copilot for Feedback Analysis. Back in 2023, we launched our first AI features. Fast forward to today, and it feels like a lifetime ago, given how rapidly the AI landscape has evolved. At Harvestr, we’ve been experimenting, learning, and fine-tuning cutting-edge technologies to tackle our customers’ biggest challenge: capturing, analyzing, and acting on growing amounts of scattered product feedback. That’s where HAI comes in: 💡 It identifies the most valuable insights from all your feedback sources. 🗂 It categorizes feedback based on your unique product taxonomy, making it instantly actionable. 📈 It syncs with your CRM to combine feedback with revenue data, empowering you to make product decisions that drive growth. The results speak for themselves: HAI can process 100% of your incoming feedback with unparalleled accuracy, saving product teams countless hours on analysis. Try it it here: https://harvestr.io/product/ai-f... A massive thank-you to our incredible team and to the customers who’ve helped shape HAI through their feedback and insights. And this is just the beginning—we’re already hard at work on the next step. 🚀

Congratulations on launching HAI! I especially like the categorization functionality: when the stream of feedback gets overwhelming, AI is the only way to effectively cluster and make sense of it. I did notice a few things about the onboarding experience that might make it more efficient: 1. Starting with the Big Picture Currently, new users are encouraged to interact with individual components of the interface before they’ve seen the full system. While this is a common approach, in relatively complex systems it can backfire. For someone new to the platform, the system feels like a black box, and without context, the knowledge gained from early interaction might not stick. One idea is to let users input some real feedback during onboarding. For example, they could copy feedback directly from a table or even another system’s interface into a large text field. The AI could process it even with weird artifacts and create a project based on their actual data, which would be far more engaging than starting with a dummy project. You could even go a step further and suggest connecting feedback sources right from the start (though this might require some careful implementation). This approach also helps clarify concepts like components, which I found weren’t entirely intuitive even after the explanation provided in onboarding. 2. Interactive Onboarding vs. Highlighted Guides Interactive onboarding tools are popular, but I’ve found their effectiveness can vary greatly. Many users tend to skip through them. Alternatives like product tours or highlight guides that spotlight specific parts of the system might work better. That said, these guides should be thoughtfully integrated to avoid becoming overwhelming or annoying. Overall, I’m very impressed with HAI, it has a huge potential!