Hi there, @Tankster. Welcome to the Community! 

Perhaps you could export the desired email replies from Microsoft Outlook directly as a CSV, then import them into Zapier Tables, or another database type of app? 

For that I’m thinking you’d likely want to set up a Zap that triggers when a New Record (Zapier Tables) is added to the table, and use a Classify Text (ChatGPT) action to identify what would be the most relevant category to put it in to make it easier to search for later. Then you’d have an Update Record (Zapier Tables) action that sets the category based on the output from the ChatGPT action.

That way when you import the old emails into the table the Zap will trigger and categorise them for you. To have new email replies added to the database going forward I’d recommend using the New Message in Folder trigger for Microsoft Outlook followed by a Create Record (Zapier Tables) action. That would allow you to move any replies into a specific folder in Outlook and have the Zap automatically add them to the table.

What are you looking to do with the information from that point? Just use it internally to search for similar cases/questions? Or would you be wanting to make this information available publicly for folks to get an answer, like via a chatbot for example?

With the aggregating of the email addresses I’d have thought that the approach would be to export the current list of contact email addresses from each of the various sources into the a single table that you’d manage them in. And set up Zaps to trigger when a new contact is added to any of those existing sources and then add their details to the table as a new record. How does that approach sound? 

Looking forward to hearing from you on this!