If you haven’t already, sign up for The AI Exchange Newsletter where we’ll continue posting helpful resources like this!
What if your sales call follow up emails drafted themselves?
In the below tutorial, we present a step by step how-to guide on using automation software, like Make, to begin integrating GPT into your repeatable, daily tasks.
All you need to get started is a Slack account, Open AI account, and free Make subscription*!
We understand that everyone takes notes differently and each team has different sales strategies. And while this guide is a great starting off point for automating workflows, most real life implementations will require prompt customization. That being said, be on the lookout for future resources in which we will provide guidance on prompt engineering and customization for your specific use cases.
<aside> 💡 *Please note that this is an affiliate link through Make, however, we would recommend this tool regardless because they have the best free tier + a really easy to use UX!
</aside>
create a new Slack channel for yourself
sign up for a Make account
create a new Scenario
add Slack as your first step
in the actions pane, scroll to the bottom and find “New Event”
add a new Webhook and fill in the corresponding fields
This will be necessary to ensure that your scenario will execute when a new message is sent to your Slack channel
Note: if you do not have one already, you will have to establish a connection to your Slack account
add OpenAI GPT-3 as your second step
Note: you will need to create a connection to your Open AI account by adding your API key
To use OpenAI through API, you must create a free account and generate keys.
fill in the corresponding fields for the Open AI Completion call
We have included a suggested prompt, but feel free to customize it to your specific email style and sales call strategy! Within your prompt, don’t forget to link to the received Slack message text from the first scenario step.
Suggested prompt: Using my sales call notes, draft a follow up email that schedules a follow up meeting and delivers on any action items promised during the call to the potential customer in a friendly and helpful tone. Here are the sales call notes: <input call notes>
before adding the next step, click the wrench between the first Slack step and OpenAI step to create a filter to remove any GPT generated messages
This will make more sense in step 11, but in order to ensure we don’t have a loop of sending and receiving messages, we want to only continue if the message text does not contain (case insensitive) “GPT response:”, shown below.
add another Slack module as your final step after Open AI to send back your result
fill in the corresponding fields to send a message in Slack
In the message text field, choose the GPT-3 completion by selecting “Choices”, then “text”
To ensure that we can filter out the responses generated through this Scenario, we’re going to add “GPT response:” to the front of the resulting message, as shown below.
test your integration
Test your integration by sending a test message (i.e. one of your sales call transcripts) into the slack channel and clicking “Run once” in the Make scenario builder view