Clay Icon
  • Avatar of Arturo O.
    Arturo O.
    Β·
    Β·

    I was blown away given how many workarounds there are to automate this! Quick question, I know serper runs a standard Google search, do you see any incorrect profile matches at all, compared vs asking Claygent to run the same search but double checking if it is? I ask because I usually recommend Claygent when we only have these kinds of inputs cause of the AI but have not checked/measured accuracy. Thanks! :D

  • Avatar of Hamid A.
    Hamid A.
    Β·
    Β·

    Thanks for sharing Andreas! How can we prevent AI and/or Google search from getting profiles of similar names (closer in semantic search)? There is a good amount of people that have private profiles or used personal emails when they sign up with LinkedIn.

  • Avatar of Andreas W.
    Andreas W.
    Β·
    Β·

    The workflow only looks for profiles from business emails. For personal emails, if that’s the only data point you have, use the waterfall. You’re right at bigger companies that can happen, so maybe a conditional formula that for companies >1000 employees also use the email match approach

  • Avatar of Andreas W.
    Andreas W.
    Β·
    Β·

    Arturo O. we’d have to run it side by side I guess! True no verification built in here (did it manually), something I could add in the future.

  • Avatar of Arturo O.
    Arturo O.
    Β·
    Β·

    Yup, makes sense! Hamid, if you prompt the AI to confirm if they work at the company and to otherwise return "Not found" if it can't do this, the false positives decreases quite a bit or even gets nullified. Of course, testing it out is definitely the best way to confirm this. Still super efficient and cost-effective in any case, because you can scale this up for thousands of leads opposed to hours on end of manual work :)

  • Avatar of Channeled
    Channeled
    APP
    Β·
    Β·

    Hi Andreas W.! This thread was recently closed by our Support team. If you have a moment, please share your feedback: