• Subscribe
  • Anyone building your own AI sales call analyzer in-house?

    Tham (Sylvia) Nguyen
    6 replies
    AI sales call analyzers are not new, but most of them apply a generic analysis template to all businesses' sales calls. This doesn't work because each business is vastly different (niche, product/service, ICP, strategies) so the way we handle sales calls and the insights we look for afterward also differ as well. What I think works better is to build your own AI sales call analyzer. After all, only you truly understand what you need! I'm curious to learn if you guys are doing this and how. Share your experience below!

    Replies

    Briella
    Built me with a data scientist on the team. We focused on tracking emotional cues and engagement levels that typical analyzers don’t catch.
    Tham (Sylvia) Nguyen
    MindPal
    MindPal
    Launching soon!
    I recorded myself building my own custom AI sales call analyzer. You can watch it here if you think it may be helpful to you:
    Share
    Odette Celeste Montgomery
    Yeah, I'm using open-source Whisper for speech recognition and then GPT-3 to analyze the transcripts and pull out key insights. Working pretty well so far! The custom fine-tuning capabilities are really powerful.
    Share
    Tham (Sylvia) Nguyen
    MindPal
    MindPal
    Launching soon!
    @odettecelestemontgomery Nice! How long have you been using this? How do you integrate it into your business automation flow?
    Anthony Hughes
    I customized one by layering on top of an open source model. It took time to fine tune, but now it’s picking up on details that generic analyzers miss.
    Matthew Russell
    In the middle of building my own. Biggest challenge has been making sure it understands my unique customer personas, but the results are already better than anything we’ve tried before.