So far we’ve been looking at what our competitor wants us to think about their offer.

Marketers tell you what they think you want to hear. Can’t blame ‘em.

But customers? They tell you what actually matters. They reveal the gap between promise and reality—and that gap is where opportunities hide.

We’re switching gears today and looking not at what our competitors say about themselves. But what their customers say about them.

I'll show you how to systematically mine customer feedback to understand what people actually value, what frustrates them, and what they wish existed.

Let's get started:

The Unfiltered Truth

Marketing messages are crafted to persuade. Customer feedback is raw truth. When someone's angry enough to leave a 1-star review or thrilled enough to record a video testimonial, they're showing you what actually matters.

This feedback reveals:

• What people really struggle with (vs what marketers assume)

• What actually delivers value (vs what sounds good)

• What's missing from current solutions

• What would make people switch providers

Think of every review of your competitor as a free consulting session with your target market. And you didn’t even need to convene a focus group!

Let’s gather up some truth bombs then. Start with the obvious places:

• The competitor’s website testimonials

• Google reviews

• Facebook page reviews

• Trustpilot or industry-specific review sites

• Course platforms (if applicable)

You can use an agent like Manus to go get all this if easier.

But also check:

• Reddit discussions about the company

• Facebook group mentions

• Twitter conversations

• YouTube comments on their videos

Collect everything for now. Yes, even the glowing 5-star reviews that feel like they were written by their mom. Patterns matter more than individual opinions.

The Two-Sided Intelligence

Now it gets a bit Goldilocks…

5 star reviews don’t tell us that much because they are too glowing.

1 star reviews don’t tell us that much because they are too angry.

3 or 4 star reviews though?

Ooo I love a 3/4 star review.

Why? Because for someone to give a 3 or 4 star review actually means they’ve thought about it. “It’s good but…” or “It’s terrible but…”.

We get far more insight here than at the extremes.

Before AI I’d have said only collect the 3 and 4 star reviews and focus here. But we can get AI to filter and sort for us so collecting everything remains valid. Just be aware for yourself that the gold is in the 3 and 4 stars.

The Review Analysis Framework

Feed all reviews into your AI project with this prompt:

I've collected customer reviews for a competitor's offer. Help me extract deep insights about what customers actually experience versus what's marketed.

Discard overly negative or overly positive reviews where nothing substantial is said. Instead focus attentions on 2-4 star reviews where there are substantial points made about the company and their product and services.

Analyse all reviews to identify:

1. Satisfaction Patterns
- What consistently delights customers
- What repeatedly disappoints
- What surprises people (good and bad)
- What expectations aren't met

2. Value Perception
- What features get praised most
- What people say is "worth the price"
- What they wish was included
- What they don't use or mention

3. Hidden Pain Points
- Problems people didn't expect
- Struggles not addressed in marketing
- Support issues that arise
- Technical or delivery frustrations

4. Transformation Reality
- Actual results people achieve
- Timeframes versus promises
- What success really looks like
- Who succeeds vs who struggles

5. Language Patterns
- Exact words customers use
- How they describe their problems
- How they describe solutions
- Emotions they express

Provide everything in a report for me.

And create a "Voice of Customer" document that contrasts marketing promises with customer reality.

Video Testimonials: The Deep Dive

Video testimonials are a completely different beast. People reveal so much more when they're talking versus writing.

If testimonials are on YouTube, NotebookLM becomes your secret weapon. Upload the video URLs and it can analyse the transcripts at scale.

Use this prompt with NotebookLM:

Analyse these video testimonials for deeper insights:

1. What specific moments or features do people get emotional about?
2. What stories do they tell unprompted?
3. What results do they emphasise vs downplay?
4. What do they say they struggled with before finding this solution?
5. What words/phrases appear across multiple testimonials?

Look for what's NOT said as much as what is. What aspects of the offer do they not mention?

This is supplemental to the text reviews. Go for quality over quantity here too - look for videos where customers really go into details rather than just “I love it.” That tells us diddly squat.

What's Next?

Finally we bring it all together. We'll synthesise everything—offer structure, marketing messages, and customer reality—into your leapfrog strategy.

You'll learn how to position against established competitors without competing directly, and how to build something that makes their offer feel outdated.