Transform product feedback into roadmap features

Step 1: Create a feedback form

To get started, you’ll need to have a user product feedback form. We’re going to use Google Forms. To follow along, go to Google Forms and click the “Template gallery” option in the top right corner of the screen.

Then, scroll down and select the “Customer Feedback” form template.

Customize the form, then you’ll need to submit a fake form response to create test data. To do this, click the “Preview” icon in the top right corner.

Then, complete the form with some test feedback.

Step 2: Set up the form submission trigger

Now that we have our test data, we can head over to Zapier to create the Zap automation. To do this, click the “Zaps” tab in the left-side navigation.

Then, click the “Create” button in the top right corner to create a new Zap.

For the Trigger, search and select Google Forms (or, the form builder you’re using), and set the Trigger Event to “New Form Response”.

In the Account tab, make sure to authenticate your account. Once done, search and select your customer feedback form as the Form to monitor.

Finally, in the Test tab, run a test and you should see the form submission you created from Step 1 of this tutorial appear.

Step 3: Generate an OpenAI API key

Next, we’ll need to generate an OpenAI key. This will be used in the following step within our Zap to transform the user feedback into a roadmap feature. To do this, log in to platform.openai.com, then click ‘API’ to be taken to the developer platform.

Click ‘API keys’ on the left menu, and click ‘Create new secret key’.

Give it a name and click ‘Create secret key’. Once generated, copy it to your clipboard. Also, keep it secure; do not share it publicly as it gives access to your OpenAI account capabilities.

Step 4: Add an AI step to transform feedback into a feature

Now that we have our OpenAI API key, we can add a ChatGPT step to our Zap. To do this, go back to Zapier, click on the Action step below the Google Form trigger, and search and select ChatGPT as the app. Set the “Event” to a Conversation.

In the Account tab, you’ll be prompted to add your OpenAI API key that we generated in the previous step. Add this API key to connect your account.

Then, in the Action tab, we’re going to provide a lengthy user message to instruct ChatGPT on how to transform user feedback into a product feature. Make sure to insert your own product description in the user message as well as the form feedback from the Trigger step of the Zap.

Sample user message:

You are an expert product manager tasked with transforming user feedback into a valuable product feature. Your goal is to analyze the feedback in the context of the existing product and propose a new feature that addresses the user's needs while aligning with the product's overall vision and functionality.

First, carefully read and understand the product description:

Product Description:
‍[insert your product description]

‍Now, consider the following user feedback:

User Feedback:
‍[insert form feedback from Zap trigger]

‍To transform this feedback into a product feature, follow these steps:

1. Analyze the user feedback:
a. Identify the core problem or need expressed by the user
b. Determine if this is a common issue or a unique case
c. Consider how this feedback relates to the existing product features

2. Brainstorm potential solutions:
a. Generate at least 3 possible features that could address the user's feedback
b. Ensure each feature aligns with the product's description and overall purpose

3. Evaluate and select the best feature:
Assess each potential feature based on:
a) How well it solves the user's problem
b) Its alignment with the product's current functionality and goals
c) The potential impact on other users and the product as a whole

Choose the most promising feature that balances user needs and product vision

4. Develop the feature proposal:
a. Create a clear and concise description of the chosen feature
b. Explain how it addresses the user's feedback
c. Outline the potential benefits for users and the product

5. Consider implementation:
a. Briefly discuss any technical or design considerations for implementing this feature
b. Suggest how this feature could be prioritized in the product roadmap

Present your feature proposal in the following format:

Feature Name:
[Provide a short, descriptive name for the feature]

Feature Description:
[Write a clear, concise description of the feature (2-3 sentences)]

User Problem Addressed:
[Explain how this feature addresses the specific user feedback (1-2 sentences)]

Benefits:[List 2-3 key benefits of this feature for users and the product]

Implementation Considerations:
[Briefly discuss any important technical or design considerations (1-2 sentences)]

Priority:
[Suggest a priority level (High/Medium/Low) and briefly justify it]

Ensure that your feature proposal is specific, actionable, and directly related to both the user feedback and the product description provided.

Now that the user message is set, we can define the other parameters of the ChatGPT step. We’re going to set the Model to “gpt-4o-mini” - a powerful and cost-effective model that’s perfect for this workflow. We’ll leave the rest of the settings to their defaults.

Test the step and evaluate the output. Adjust the user message as you need to get the output you want for your workflow.

Step 5: Convert feature details into structured data

Now that we have ChatGPT generating product feature descriptions, we need to transform the block of text it outputs into structured data so we can easily add the content to our product roadmap. To do this, add another ChatGPT step and set the Event to “Extract Structured Data”.

Insert the “Reply” from the previous step as the input for the “Unstructured Text” field. We’ll use “gpt-4o-mini” as the model for this step as well.

Then, we’ll instruct ChatGPT to convert the output from the previous step into JSON in the “Description” field.

Sample description:

Convert each key-value pair to JSON.

In addition, for the “Values to Extract” setting, insert a value name corresponding to each section from the previous ChatGPT step output you want to extract. If you’re following along with our previous user message, this will include:

• Feature Name
• Feature Description
• User Problem Addressed
• Benefits
• Implementation Considerations
• Priority

We’ll leave the rest of the fields blank in the Action tab. Then, test the step. If everything worked, you should see key-value pairs with the variable name and its corresponding data. In Zapier, this looks like blue-colored variable names with text data next to them.

Step 6: Add feature data to your roadmap

Now that we have our structured feature data, we can add a final step to add the data to our roadmap. We’re going to use Notion as our product roadmap in this tutorial, but you can use any product roadmap tool that connects to Zapier (e.g. Linear, Airtable, etc.).

If you’re using Notion, set the Event to “Create Database Item”.

Authenticate your Notion account, and make sure to provide Zapier with access to the Notion page with your product roadmap. Then, in the Action step, select your Notion database and map the output from the previous step to the proper roadmap column headers.

Then, test the step, and head over to Notion to see the product feature added to your roadmap.

Make sure to publish your Zap and share your feedback form with your users. And that’s it! Now, you have an automated system that can ingest user feedback and convert it into descriptive product features on your roadmap.

‍

‍
‍

This tutorial was created by Garrett.