- Help your customers and users buy your product or app with copy that tells them what they need to know
- For service providers, SaaS companies and ecommerce store owners
- Once you’ve got all your data in order, you can start rewriting your copy with AI + a super-targeted list of its recommendations
Use AI to analyze customer reviews, user feedback and service tickets – in minutes – so you can create offers that’ll actually sell.
You’ve poured hundreds of hours and blood and sweat and tears into creating and shipping your solution or product, but how do you know if it’s what your customers and users actually want? And if your solution is what they want – is your copy actually selling it?
Here’s what to do: Prompt AI to organize and parse all your customer reviews / feedback in a document you can use to write all the assets you need to sell it. This is what ChatGPT 4, aka GPT-4, created for me when I used the AI copywriting prompts I’m about to teach you:
Here’s how to use GPT-4 to get similar results for your business or client.
Step 1: Gather all the copy you’re currently using to sell your solution, plus all the current customer reviews / feedback you’ve got, and have AI organize it all for you
The goal of Step 1 is to gather all your data on your product or solution.
If you’re selling a service or productized service, collect all the copy have on your service offering:
- Description of your offer
- A breakdown of deliverables
- Cost of your core offer
- Add-ons and their prices
- Pricing tiers and/or packages
- Discounts and their terms
- Any other copy you’re using to sell your service or productized service.
Collect all your customer feedback data:
- Reviews
- Interviews
- Customer service tickets
- Any communication by email or in-app chat that include comments and feature or service requests
Note that at the time of writing, ChatGPT has limits to how much data it can process, especially during peak hours. Make this process more efficient by pasting only relevant data into the prompt. For example, I wouldn’t paste in a whole interview transcript; I would only paste in the good stuff like sticky copy from VOC.
Pro tip: Use the tool ChatGPT Splitter to make this step more manageable and to avoid GPT rejecting your prompt if the data you paste is too much for GPT process at a time.
If you don’t want to use ChatGPT Splitter, you can use a spreadsheet to keep track of the data you’re cutting and pasting into GPT-4. I used a Google Sheet to quickly organize the data and then just copy-pasted the data into the chat.
To demonstrate this prompt, I used the copy that AppSumo created for the sales page to sell the app EmailDelivery.com and the reviews left by AppSumo purchasers.
Step 2: Enter your data with these 2 prompts
Next, we’re going to enter all the data you organized in Step 1, so AI has all the data it needs to follow the prompt in Step 3.
Step 2 has two prompts:
The first prompt tells GPT-4 that you’re selling a product or service, and that you’re providing it with copy of the features and pricing information for it, and a direction to wait for further instructions
Copy paste this prompt:
I am selling a [SaaS app or other digital or physical product]. Below is copy on the features and descriptions of those features and how they function, pricing tiers or packages and applicable discounts and any other copy used to sell the product. Please wait for further instructions; do not do anything with this data yet.[ Your Data ] The second prompt tells GPT-4 that you’re you’re providing it with customer feedback data, and the different types you may be pasting into the chat. There’s also direction to AI to wait for further instructions.
Copy paste this prompt:
Below is customer feedback data on the [SaaS app or other digital or physical product] I am selling. Data here may include reviews and interview and survey responses, customer service tickets or communication by email that include comments and feature or service requests. Please wait for further instructions; do not do anything with this data yet.[ Your Data ] Step 3: Prompt AI to organize the customer feedback
In GPT-4, ask it to organize the feedback data. I asked GPT-4 to organize the reviews in a table by mood, and highlight each user’s criticisms, recommendations or feature requests
Copy paste this prompt:
Please organize the reviews in a table by mood, and highlight each user's criticisms, recommendations or feature requestsStep 4: Ask GPT-4 to use your customers’ feedback to improve your existing copy
Prompt GPT-4 to tell you how the copy selling your product can be improved, using your customers’ or users’ feedback.
In this prompt, I asked AI to use the positive reviews to find opportunities to highlight or sell harder on a feature in the copy. I also asked GPT-4 to use all the reviews, but with more reliance on the negative reviews, to find places in the copy that can be improved on for clarity.
Copy paste this prompt:
Please tell me how the copy selling the app can be improved, using the reviewers' feedbackGive me recommendations on what should be highlighted in the copy, especially from the positive reviews. For example, if a particular feature has positive reviews but the feature isn't highlighted or further explained in the copy, please point that outGive me recommendations what should be better clarified, using all the reviews, especially from the negative reviewsWithin minutes, AI had generated the following for me:
Read the full chat here.
Please remember that ChatGPT doesn’t know your product or service or your market as well as you do. Always take what GPT-4 generates only as way to get passed the blank page. It’s up to you as a pro conversion copywriter to determine what finally ends up on the sales page.
I tried this AI prompt on a set of silicone spatulas. Here’s what happened.
I used these prompts to improve the copy on the Amazon product page selling a set of silicone spatulas by ALLWIN-HOUSEWARE. GPT-4 gave me this list of ways to improve the copy:
Read the full chat here.
Use these prompts to quickly identify parts of your copy that you can improve so that you can create offers that your customers or users want.
Conversation