How to Automatically Export and Analyze Conversations from Custom GPTs
Are you looking to track and analyze interactions with your users across channels like web chat widgets, WhatsApp, Instagram, and more? If so, you're in luck. In this article, we'll show you how to automate the exporting of transcripts from your custom GPT deployments and provide you with a complete automation template for analyzing and extracting valuable data from them with the help of AI. By the end of this article, you'll not only know how to automate the exporting of transcripts from your custom GPT deployments, but you'll also get access to our complete automation template for analyzing and extracting valuable data from them with the help of AI.
Table of Contents
1. Introduction
2. How the System Works
3. Setting Up the Air Table
4. Setting Up the Make Automation
5. Processing the Data
6. Overview Tab
7. Conclusion
Introduction
Many businesses are deploying custom GPTs to conversational channels like web chat widgets, WhatsApp, and Instagram. However, without properly storing and analyzing this data, you're losing half of the value of your GPTs. This data can contain valuable information for use in future decision-making. In this article, we'll show you how to automate the exporting of transcripts from your custom GPT deployments and provide you with a complete automation template for analyzing and extracting valuable data from them with the help of AI.
How the System Works
We'll be using the Assistant API to request and pull all of the transcripts and messages from the interactions that our users have had with the API. OpenAI allows us to get that information at any time we want, but we need to know how to request that information from the API. We'll be building on top of all of the work we've done in our prior videos, such as our WhatsApp and Instagram deployments. We'll be using that same system, adding a little bit in that allows us to export the transcripts and then showing you the Air Table setup that we're using and also the Make automations.
Every time a conversation is started on our website, WhatsApp, or Instagram, we'll be saving what's called a thread ID. In our prior builds, we hadn't been doing anything with that thread ID apart from using it in the conversation. In this case, we'll be saving that thread ID to an Air Table and then looking up all of the messages within that thread ID, pulling them all out, packaging them up, analyzing them, and then putting them all into a nice database and spreadsheet for us to look at and analyze later.
Setting Up the Air Table
To get started, you need to clone our Ripple, which is available on our resource hub. Once it's all booted up, you need to watch our Instagram deployment video to fully understand how it's all working. This particular builder is just a basic custom knowledge chatbot. We have removed all of the tooling from this, so all it is using is a knowledge-based document with information about our AAA accelerator program. We've given it information on that, and the point of this chatbot is going to be able to ask questions about the pricing and the features.
Setting Up the Make Automation
We've already created an entire template to make this as easy as possible for you. You'll need to add a new connection to search records node and set up the air table nodes. We'll be pulling the thread ID and doing these next steps with the thread ID that we've pulled in from that specific row. We then call the OpenAI API that allows us to get the information from these threads. We then smush those labels together into an array, and then we're able to send it off to Air Table and update the record that we've been working on.
Processing the Data
We're searching for records in Air Table that are ready to be processed. We then aggregate them all into an array, loop through all of the different records, and pull the thread ID. We then call the OpenAI API that allows us to get the information from these threads. We then smush those labels together into an array, and then we're able to send it off to Air Table and update the record that we've been working on.
Overview Tab
We've also thrown in an overview tab to show you where you can go with this information once you've been able to process it. This allows you to show how you can label all of these to a certain date and then start to see some overview data of how many messages are we getting, is it trending up, is it trending down, and then you can go into Air Table interfaces, which allows you to make a dashboard out of this.
Conclusion
In conclusion, we've shown you how to automate the exporting of transcripts from your custom GPT deployments and provided you with a complete automation template for analyzing and extracting valuable data from them with the help of AI. By following the steps outlined in this article, you'll be able to track and analyze interactions with your users across channels like web chat widgets, WhatsApp, Instagram, and more.
