Summarize your data & extract insights with Airtable AI

How to summarize comments in Airtable

Summarizing comment threads is a helpful way to stay up to date on important conversations in your organization. 

This feature allows anyone with comment access or on an interface to generate a summary of the comment thread. That summary is then available to all users on that interface to view until new comments are added, when the summary can again be generated.

Summarize data with an Airtable template

Outside of in-thread comment summaries, the easiest way to get started with summarization using the AI field is to use one of our pre-built templates. There are two key templates we recommend for common use cases:

• General > Meeting summary

  • Requires as input a field with meeting notes or a call transcript

• UX Research > Feedback summary

  • Requires as input a field with meeting notes or a call transcript, and specific insights to extract

We recommend sticking with the “enable the best models” setting in the field config when doing summarization because these models are better at following directions and writing clearly and concisely.

If you want output in a different format or have more specific instructions, it’s best to use a custom summarization prompt.

Advanced AI-powered summaries 

Custom prompts

If you’ve started with a template but want to build a more flexible AI query, you can convert your templated query to a custom one by clicking the “change to a text box” button.

Now that you have a custom prompt, you can edit any part of the prompt or delete it all and start from scratch. You may want to try:

• Changing the AI persona to be specific to your use case

• Changing the instructions of how you want to summarize or extract information. Some possibilities:

  • Extract all mentions of topic X or competitor Y

  • List all the customer pain points with the product and why they’re challenging

  • Summarize areas of disagreement and what the competing viewpoints are on these areas

• Adding more examples to clarify how you want the outputs to look

• Adding in additional fields to provide context

Here’s a custom prompt written to extract very specific information from a call transcript:

Incorporating AI summaries into Automations

In Automations, you can create an action which makes a query to an LLM and gets back a response. You can then use that response in any of the ways you use Automations today - write an email, create a doc, post to Slack, write to a field in a base. In addition, you can input that response back to an LLM, creating the ability to chain LLM calls together. 

Here are two examples of how you might incorporate AI into your Automations logic:

Example #1: Send an email summary of meeting notes from the previous week

Example #2: Send an email summary of new product features that launched in the last week, or that are launching in the next week

1. Trigger: Every week on Monday morning

2. Find records from projects where project status = Shipped or where GA milestone is in the next 2 weeks

3. Send data from the project description as well as launch status to the AI to create the summary content.

4. Create an email action that sends the launch summary (along with other news and updates) out to the ProdDev organization.

Try it now

With this feature at your fingertips, it’s easier than ever to distill vast amounts of information from comments, user feedback, and meetings into concise and actionable insights. You can get started using one of our pre-built templates for simple use cases, or customize an AI-powered workflow through prompts and Automations. 

Get started using Airtable AI today! As always, we welcome your feedback and suggestions as we continue developing new AI-driven features for Airtable.