Thanks @leancto! I have experimented with exactly that use case of tagging recurring needs. But I took it out before I released the product because I wanted to find a way to make sure people are confident in sending away their data to the open ai api. But I have an approach now that I am happy with where you simple select the transcripts you want to analyse, and the app clearly outlines what will happen with the data (like with how it currently works with the conversation tips), so I do have that planned. In terms of sharing amongst team members. That can be done in a few different ways. The way hunch works is that you select a folder on your device, and hunch will just create and edit markdown files. So if you choose a folder from a share drive over google (or however your team is setup) the markdown files will be accessible by anyone on your team that has access to that folder. Keen to get any thoughts on this, whether it sounds in the ballpark of the type of thing that would work.

@leancto I've just submitted updates to the Apple app store for both the iPhone & Mac versions of Hunch. There will be a new feature where you can use the open ai api key to select a set of your customer transcripts and analyse them for common needs. I'll let you know when the updates get through the review process and go live on the app store. Thanks again for the feedback 🙏

@leancto both iPhone & Mac apps through the App Store review process now and live in the App Store. Hope the analysis of needs works well for you.

Thanks @marion_fowler, really appreciate that. Yea I'm really hoping to help people learn the customer conversation skills. I think with the guide in there, people can make some big gains. Let me know if you have any ideas to make it better.

@ben_harper6 Thanks 🙏

Hey @matteo_rider, to use the ai function, you create your own open ai api key, and just paste it into the input field where it asks. This way people can just use it as much as they want, with the prices on api calls pretty decent. Also, it means I can make hunch free without any subscription. It is current using o1-mini from OpenAI for the model. I wanted it to be reasonably quick, but also clever in how it formulates the tips with the Hunch guide. Keen you get your thoughts on whether you think this setup works or, or have questions/thoughts. Thanks for the question.

thanks @timscullin! You've been an inspiration 🤙

@bradely_charlesbois yes totally, they are all actually just on your device saved as markdown files in the folder that you select

@ethan_lennon I think there is value for larger companies. I've been working in a mix of both startups (2 days / week) and corporates (3 days / week) for the last while, and in the larger organisations there often aren't the tools for people to get practices at customer conversation, or sit with transcripts of conversations to form their own thoughts. Add hashtags/hunches to notes, and find insights. In terms of data security, which is often more considered in larger organisations, Hunch runs everything on device. And nothing will leave the device unless you want to use the ai tips. I try to put a warning in that screen before you send data off about what is involved with sending data to the open ai api. But without using that, there is still the guide there that I have put all the key things I think are important for good customer conversations. Keen to get your thoughts on that?

@kiwi_designer thanks crystal!

Hey @juan_smith1, good question. These tips are currently general guidelines. I've work for the last 2o years doing customer conversations over a range of different industries, and org types (ie. startups, corporates, government departments), and the general guidelines will be a good start. But there are some interesting things I could look at around specific industries, and I'm keen to look into that. There could be a way for you to enter a few contextual factors about your setup, like industry, and place within the org, and then I could mix that into the open ai api call?