Skip to content
Download app as it will allow you to screenshot reviews and copy the text automatically and directly from the photos – will save you a ton of time.Go to your Shopify store and look at the reviews for your main productScreenshot the reviews and paste the text into the "Review mining" document. Aim to collect at least 20 of these.Try to collect as many reviews as possible - don't be biased or overthink this.Then go onto Google, search up your product, or the solution it solves, and screenshot reviews/comments people have left that are directly related to your product. Aim to collect at least 20 of these.Go onto Reddit, and search up a niche or forum directly related to your product's niche and the solution it solves. Copy and paste comments and reviews that people leave on this thread. Aim to collect at least 20 posts/comments.Optional but recommended: Do the same for Facebook – join a Facebook group that your audience would be in, and copy and paste their comments and posts. Aim to collect at least 20 posts/comments.Then look at your direct competitor – and copy and paste their customer reviews. Aim to collect at least 20 of these.Once you've filled out the entire document (everything should be text so the AI can read it – that's why I said to use Shottr to extract the text) then go to ChatGPT and paste this prompt:
How to do review mining
MASTER REVIEW MINING PROMPT(copy everything below into ChatGPT before pasting reviews)
You are a senior CRO research analyst.
I am about to paste raw customer reviews, Reddit comments, Google reviews, and competitor reviews below inside a single PDF document file.
Your job is NOT to summarize them.
Your job is to extract structured marketing intelligence from them.
You must analyze the reviews and output your findings in the exact structured format below.
Do not write fluff.Do not rewrite the reviews.Extract patterns, language, behavior, objections, and positioning insights.
––––––––––––––STEP 1: EXACT LANGUAGE EXTRACTION
Pull out the most powerful direct phrases customers use.
Output:
Top 20 exact phrases (verbatim)
Phrases that sound like headlines
Emotional phrases
Transformation phrases
Skeptical phrases
Do not paraphrase. Keep them raw.
––––––––––––––STEP 2: REPEATED PATTERNS
Identify the strongest repeated themes.
Output:
Top 10 repeated benefits
Top 10 repeated complaints
Top 10 repeated desires
Top 5 emotional drivers (fear, relief, status, insecurity, etc.)
Any surprising recurring theme
––––––––––––––STEP 3: OBJECTION ANALYSIS
Extract objections and friction points.
Output:
Top 10 objections ranked by frequency
Speed objections (delivery, results timing)
Price objections
Skepticism objections
Trust concerns
Product quality or durability concerns
“Didn’t work for me” patterns
For each objection, briefly explain the underlying fear behind it.
––––––––––––––STEP 4: RESULTS VS EXPECTATIONS
Analyze outcome alignment.
Output:
Where results exceeded expectations
Where results matched expectations
Where results failed expectations
Time-based expectation patterns (how long people expect it to take)
––––––––––––––STEP 5: USAGE & EXPERIENCE
Analyze what customers say about using the product.
Output:
Ease of use patterns
Friction in usage
Setup issues
Confusion points
Breakage or durability issues
Taste/feel/texture mentions (if relevant)
––––––––––––––STEP 6: PACKAGING & UNBOXING
Output:
Packaging praise patterns
Packaging complaints
Unboxing experience comments
Perceived quality signals
––––––––––––––STEP 7: SHIPPING & SPEED
Output:
Shipping speed praise
Shipping delays
Expectation gaps
Mentions of tracking/customer service
––––––––––––––STEP 8: BUYING BEHAVIOR
Analyze behavioral clues.
Output:
Quantity patterns (single vs bulk buying)
Repeat purchase indicators
Bundle preferences
Why customers chose that quantity
Buying urgency signals
––––––––––––––STEP 9: WHO THEY’RE BUYING IT FOR
Identify buyer identity patterns.
Output:
Bought for self
Bought for partner
Bought as gift
Bought for parent/spouse
Bought for professional use
Highlight any surprising buyer patterns.
––––––––––––––STEP 10: RECOMMENDATION & USE CASES
Output:
What customers recommend it for
Specific use cases mentioned
Situations where they say it works best
Situations where they say it doesn’t work
––––––––––––––STEP 11: HEADLINE & ANGLE EXTRACTION
Based only on the extracted patterns above:
Output:
15 headline ideas grounded in real language
10 angle themes (clear positioning directions)
5 strong positioning statements
5 “anti-competitor” positioning angles
5 objection-handling headline ideas
Do NOT invent angles that are not supported by review data.
Want to print your doc?
This is not the way.
This is not the way.
Try clicking the ··· in the right corner or using a keyboard shortcut (
CtrlP
) instead.