Amazon Review Mining

How to Mine Amazon Reviews and Supercharge Your Copy

Let’s be honest.

Writing your own copy can feel like you’re getting your teeth pulled.

You spend hours staring at a blank screen trying to come up with the right words, but you can’t seem to find them.

Finally you say “screw it” and write down the first thing that comes to mind.

Big mistake.

Copy isn’t just words on a page. Copy is very strategic. It’s the difference between a sale and a lost customer. It’s the difference between a thriving email list or a list that resembles a wasteland.

It pays to spend time on your copy.

When you take the time to learn the language your customer uses, they’ll feel like you truly understand what they’re going through. They feel like you’re one of them. They’ll feel like you have what they want.

But how do you find that language?

The answer is: mining Amazon reviews.

The Secret to Writing Irresistible Copy

When most people write copy, they talk about what they know. They talk about themselves and the business side of things.

But good copy isn’t about your business.

Good copy revolves around your customer. Good copy tells your target audience why they should care.

You can’t do that if you don’t know the exact words and phrases they use.

Sometimes the language can be subtle. It could be something you wouldn’t think twice about when you read it.

Take a look at this example.

Amazon reviews comparison

They’re both good products, they’re both roughly the same price, and they’re both getting great reviews. One of them, however, has waaay more reviews than the other.

If you look at the reviews for the Jerzees one, you’ll notice people talk about ‘hoodies’ more than they talk about ‘hooded sweatshirts’

Amazon reviews language comparison

Now I’m not saying it gets less reviews just because it says ‘hooded sweatshirt’ instead of ‘hoodie’, but it certainly helps.

When you use the same language your customer uses they’ll feel like you – get – them. They’ll feel like you’re part of the same group and that you know them.

Now, you might be wondering…

Does this work for service based businesses?

Yup. People “hire” books and products to help them do something. If you’re an accountant, you can check out books on accounting to see what people are saying in the reviews.

The goal isn’t to find a product that does exactly what you do. The goal is to find a product your target audience might be reading.

How do you spot fake reviews?

After you’ve done this a couple of times, it’ll be pretty easy to spot the fake ones. Later in this guide we’ll go over how to separate reviews so you only use the good ones.

Now let’s mine some reviews, shall we?

The Step-by-step guide to Amazon review mining

Step 1: Find products that solve the same problem you’re trying to solve

The first step to mining Amazon reviews is to find a couple of products your target audience would use.

For this guide, I’ll use the example of a marketing consultant trying to work with startups. You might check out books like…

  • The Lean Startup by Eric Ries
  • Traction by Gabriel Weinberg and Justin Mares
  • The $100 Startup by Chris Guillebeau
  • Growth Hacker Marketing by Ryan Holiday

Find books or products that have a decent amount of reviews ( preferably over 150 ). The more reviews the better. You can use books that don’t have a lot of reviews, but your data won’t be as accurate as when you have a 100 + reviews

Step 2: Create a Google Sheet

Before we can start gathering our reviews, we need to prepare our spreadsheet. We’re going to create something that looks like this:

mining spreadsheet

Note: If you want a copy of this doc, you can get it by clicking here.

Notice it has a 7 different headings:

  • Link – This is where you’re going to put the link to the book. This just makes it easier for us to  where we left off.
  • Review – We’re going to paste reviews in here.
  • Character count –  This is a way of finding out how passionate people are about a subject. If they don’t care, they’re not going to write a long review. Don’t worry, Sheets can easily calculate this for you.
  • Stars –  We want to keep track of the amount of stars the review had. 2 to 4 star reviews tend to much more nuanced than 1 or 5 star reviews.
  • Multiplier – Because 2 to 4 star reviews tend to be more nuanced, we’re going to give them more weight. Every review that isn’t a 1 or a 5 star review gets a 1.5 multiplier.
  • Responsiveness score –  This is the final score we’re going to give the review. The responsiveness score = character count x multiplier.
  • Hyper – Here we’ll determine if a person is hyper responsive or not. Don’t worry if this doesn’t make sense yet, I’ll explain later on.

Create a tab for every book you use. If you do all the reviews at once you’re going to get messy data and it’s going to make your life a lot harder.

Step 3: Copy the reviews to Google Sheets

Before we can check out figure out what words to use in our copy, we first need to gather a bunch of reviews in our spreadsheet.

There are two ways of doing this.

Method 1: The manual way

The good thing about this method is that it doesn’t require any technical knowledge whatsoever.

The bad thing about this method is that it requires you to copy & paste the ever loving shit out of the reviews, which is about as exciting as watching paint dry.

If you plan on doing this, grab a drink, put on some music, and grind it out. Or pay someone to do it for you.

Method 2: The automated way

The second method is to use scraping to gather the reviews automatically. You can use a tool like Parsehub or Import.io to help you gather reviews quickly.

The downside of this method is that you need technical knowledge. If you don’t know what you’re doing you might end up getting your IP banned, which is no bueno. Site admins generally don’t like bots crawling through their website.

Since this isn’t a tutorial on how to scrape, I’m not going to go into much detail. If you want to learn more about how to scrape a website, you can read these articles:

Step 4: Decide who you’re going to listen to

Who would you rather take advice from?

This person?

Amazon bad review

Or this person?

Amazon good review

The second person, right? He spent a lot of time going in detail and talking about what was good and what was bad about the book. The second person seems much more engaged than the first person.

We call these people hyper responsives.

To figure out the language our target audience uses, we’re going to listen to the hypers in the group. The question is: how do we determine who’s a hyper and who isn’t?

To figure out who’s more engaged, we’re going to weigh the reviews. We want to give nuanced reviews ( 2,3,or 4-star reviews) more weight because they tend to be more objective.

Then we’re going to look at the responsiveness score. In this case, if the reviews responsiveness score is over 1000, I’m going to consider them a hyper.

If you find that none of the reviews have a score over 1000, you might lower it a bit. The goal is to separate the fake and the low effort reviews from the juicy ones.

Step 5: Turn the reviews into word clouds

In this step, we’ll take the reviews that we’ve gathered and turn them into word clouds. It allows us to see what words are and what words aren’t being used.

There are plenty of tools out there that can help you turn your reviews into word clouds. Here are a couple of them:

Create two word clouds for every book you review. One cloud for the hypers and one for the non-hypers. This is so we can see how both groups talk about the subject.

Below are the word clouds I got for Growth Hacker Marketing by Ryan Holiday.

Hypers:

review word cloud - hypers

Non-hypers:

review word cloud - non-hypers

Step 6: Compare word clouds

At this point, you should have a bunch of different word clouds.

Pull them up next to each other and look at what they’re saying. Look for a couple of things:

  • What words are being used a lot?
  • What words could be there but aren’t?
  • Are there any differences between hypers and non-hypers?

Do this for every book you’ve picked and really go through the word clouds to see if you can come up with some good words to use.

For the Growth Hacker Marketing book, there are a couple of differences:

  • The hypers in the group use the word ‘Viral’ more
  • The hypers talk more about Dropbox ( one of the most well known examples of growth hacking) and Instagram.
  • The hypers talk about products more than non-hypers

Taking a closer look at the reviews

Instead of just looking at what words and phrases people use, we’re going to take it a step further. We’re going to take a closer look at the reviews and find answers to three questions:

  • What are they trying to achieve?
  • What are they trying to avoid?
  • What stops them from achieving their goals?

Once you have the answers to those questions, you can really amp up your copy. Use those hopes, dreams, and fears to write copy that speaks to what your visitor wants.

So open up your spreadsheet again and create another tab. Add the three questions to the sheet.

review copy mining

Now go through the reviews and copy any memorable phrases into the new tab. Anything that can tell you what they want to achieve, avoid, or what is stopping them.

Turning Your Findings Into Copy That Connects

It doesn’t matter how many golden nuggets you find if you’re not going to use them.

The goal of mining Amazon reviews is to find the language your customer uses and echo that same language back to them. Doing so will allow you to make a deeper connection with them.

A quick word of warning though. The goal isn’t to just copy and paste the reviews on your site. The goal is to find specific words or memorable phrases you can use. Don’t use these reviews as a substitute for writing copy.

Now that we’ve got that out of the way, let’s take a look at what you should optimise first.

Optimize your home, about, and landing pages first

Before you go all crazy optimizing every little bit of your website. We’ll take a look at what’s more important to optimize.

In this case, it’s your home, about, and landing pages.

If your site is like most people’s website, these pages get 80% of the traffic. You can use the language in blog posts and drip campaigns, and it will help you. However, you’ll probably find the biggest benefit in optimizing your home, about and landing pages.

Optimize headlines first

If your headline fails to capture people’s attention, the rest of your copy doesn’t matter. You can have the best copy in the world, if nobody reads past your headline all that hard work was for nothing.

So optimize your headline first. You can use the words and phrases you found in the reviews to make your headline more interesting and attractive.

If reviews talk about ‘getting traction’ instead of ‘growing your business’, use traction. It will capture the attention of the people you are targeting.

Lead and close with the most important benefit

If you did your research, you’ll have a bunch of different reasons why people are trying to achieve their goal.

Maybe they want to grow their business so they can save for a trip. Maybe they want to grow their business because they want to help more people. Maybe they want to grow their business because it would allow them to take some time off and spend more time with their kids.

There are a million different reasons why people want to do something. It’s your job to lead with the most important benefit. Pick the most detailed, most vivid example you can find and make it even more vivid.

The goal is to make people FEEL like you have a solution. It’s important that you have a good solution, but most of the buying decision are made with feelings, not with thoughts.

The reason you want to lead with the most important benefit is simple. It’s called the serial positioning effect.

Serial Positioning graph

With the serial positioning effect, people are more likely to remember the first and the last items on a list. This is because the top item gets the most attention and the last item is the most recent.

If you add the top benefit first and the second most important benefit last, you’ll make your copy a lot better.

Counter objections

If you look through your research, especially the more mediocre reviews, you’ll find some objections can come up that you can use in your copy.

If you can prove to someone your product doesn’t have the objection, they’ll be more likely to buy your product.

For example. Let’s say you found a negative review for a book. The reason why it’s negative is because the book doesn’t give the person ‘actionable, step-by-step’ advice.

You can adjust your copy in a way that shows people that you know most guides don’t give step by step advice, but you do.

This will make you stand out from the crow and show potential buyers you know your stuff.

Make your call to actions sound beneficial

Your calls to action are the bridge between a visitor and a customer. It’s the difference between having no subscribers and having a list full of people who love to engage with you.

So make sure they count.

You can use the info you’ve found throughout this process to make your CTAs as powerful as can be.

A good call to action is two things:

  • It’s actionable
  • It’s beneficial

CTAs like “submit” are garbage because they aren’t beneficial at all. It doesn’t tell your visitor what they can expect and what’s on the other end of the bridge.

In order to improve that, look at what your headline says and provide a actionable call to action that tells them what they can expect.

Ready to Take Your Copy to The Next Level?

While mining Amazon reviews shouldn’t be the only research that you do, it can be an excellent method of initially finding out what your audience is saying and how they talk about your subject.

Once you know what they talk about, what they want to achieve, and what’s stopping them from getting what they want, you’ll have a leg up from the competition and you’ll be able to write copy that connects to your audience on a deep level.

Here’s what I want you to do:

  1. Pick 3 – 5 books that solve the same problem you’re trying to solve.
  2. Download the Google Sheet and start mining reviews.
  3. Let me know in the comments below if you’ve found anything interesting.

No comments yet


Add a Comment

Your email address will not be published. Required fields are marked *

Comment *

Name *
Email *
Website