How to Make Amazon’s Customers Your Customers Using Facebook Ads

Anthony Lee explains step-by-step how to use Amazon buyer data to create targeted Facebook ad campaigns, and go direct to customers

This post is by Anthony Lee, COO of SixLeaf (formerly ZonBlast), the first and largest product launch and ranking service for Amazon sellers.

When sellers start offering their own private label products on Amazon, their goal is usually to build an independent brand. They aim to use Amazon as a springboard and, in the future, make most of their sales through their own website.

The problem is that a lot of the training programs and advice available to online sellers doesn’t explain HOW to grow your brand beyond Amazon. There is just a common notion that once your brand becomes “big enough” it will naturally happen. It doesn’t work that way.

In this post, I’m going to talk you through some practical steps which really work to build your brand. You’ll find out how to leverage Amazon buyer data to find your customers on Facebook, and target them with Facebook advertising campaigns.

By doing this, you can direct existing customers, and other buyers just like them, to products on your own webstore, and build a really robust, independent brand.

Amazon Logo
The best tools for your Amazon business
Automate pricing, request feedback and reviews, improve your marketing & SEO and a lot more. Amazon consultants help with difficult issues like suspension, or help grow your business. All in the Web Retailer directory.
View Amazon Selling Tools Now

Your most valuable asset: Amazon buyer data

As an Amazon seller you have access to the most valuable information your business will ever be exposed to, and that is buyer data. The people who have purchased from you in the past did so because something about your brand or product appealed to them so much that they were willing to part with their hard-earned money.

It stands to reason then that these buyers, taken as a whole, represent your ideal customer. No matter what your market research says about your target demographic, the truth is always told when people vote with their wallets.

If you have a list of past customers, they represent the type of people you need to get your product in front of. You can use that to increase your visibility. There is no better list of people to help further your marketing efforts than a list of existing customers.

Your Amazon buyer data is exactly that. It gives you a starting point to get your offerings in front of more willing and eager buyers.

But Amazon doesn’t allow marketing to buyers!

That’s absolutely right. You don’t get buyer email addresses, so you can’t send emails to them directly. You can’t sell to them through Amazon’s own messaging system, because it’s against policy. You can’t use your listings to direct buyers to your own website, and you can’t include a flyer with orders offering a discount code to buy direct.

The magic that makes this possible is Facebook ads.

Sounds like a showstopper, right? It’s not. Technology has come a long way, and you can now target advertising to past customers, and people with similar attributes and interests, even without email addresses. The buyer data Amazon does provide is enough, and you won’t break any rules by using it because you won’t be marketing directly to them.

The magic that makes this possible is Facebook ads, and specifically features called Custom Audiences and Lookalike Audiences. Custom audiences allow advertisers to use filters and criteria to create groups of people (“audiences”) to advertise to. That in itself is pretty powerful, but the tool is so much greater than just being able to target a specific demographic by manually choosing the attributes that you think are most relevant.

One of the ways that Facebook allows you to create a Custom Audience is from a “customer list”. Essentially, Facebook is giving online sellers the ability to upload their customer data so they can be reached with Facebook ads. When your customer list is uploaded, Facebook uses that information to find their user profiles on their platform. They don’t tell you who the matching users are (that would be a big violation of privacy) but they do give you the ability to target them with ads.

It’s a great opportunity to get even more sales from people that already know you. But wait. It gets better.

Let’s say you’ve been selling on Amazon for a few months and you’ve had a couple of thousand sales. In the world of paid advertising, that isn’t a very large pool to advertise to, even though it’s very relevant to your unique products. Have no fear. Facebook has an answer to this too, and it’s called Lookalike Audiences.

Starting from a Custom Audience, built by uploading a customer list, Facebook can find other people who are similar to your customers. Basically, Facebook takes the interests and demographics of your Custom Audience and looks for people who closely match them, using the data it has accumulated on every user of the site. This comes from information Facebook users add to their profile, such as their gender and location, as well as their use of the site such as posts they have liked and groups they have joined.

Many of us have seen uncannily accurate Facebook ads before, that can even seem to predict if users are pregnant, and it’s the same highly sophisticated technology at work here. As long as you have a Custom Audience of at least one thousand people, you can create a Lookalike Audience of 2.1 million in just a few clicks!

Those 2.1 million people are just like your customers. They have many of the same buying habits, they like the same brands, follow the same pages etc. These are targeted individuals who are likely to want what you have.

Imagine it this way. Let’s say you put an ad for your kitchen product in USA Today. A huge number of people would see that. Some will want your product, but many won’t have any interest at all. If you are effectively paying per person who sees the ad, you may run out of money before enough of the people who would actually buy your product get to see it.

Contrast that with putting your ad in a kitchen and cooking magazine. Your audience would be much more receptive there. This is essentially what Facebook does with their Lookalike Audiences. They find relevant people, using your Custom Audience sample, so your advertising will be more effective.

Effective ads are good for Facebook, good for you, and good for Facebook users. If your ads work, and people respond well to them, you’ll keep running them. It’s a win-win.

Step by step: build a Custom Audience from Amazon buyer data

Building a Custom Audience out of your Amazon buyer data is actually pretty easy, just a bit labor intensive. All of the information you need can be found inside your reports in Seller Central. The only problem is, it cannot be uploaded to Facebook unedited. There are some manual tweaks you’ll need to perform.

Step 1: Run the report

  1. Go to Reports in Seller Central
  2. Click on Fulfillment
  3. Go to the Sales tab in Fulfillment Reports.
  4. Click on Amazon Fulfilled Shipments
  5. Input the date range and request download

The only problem with this is it doesn’t pull more than 30 days of data at a time. If you have a longer order history, this is where it can get labor intensive. You’ll need to combine multiple reports.

Once you have the report you need, click Download. Then open the file in Excel. It downloads as a .txt file, so if you have a PC and open it with the default program, it will likely open in Notepad. This will make the contents of the file appear jumbled and unorganized. Opening it in Excel will divide the data into cells.

Step 2. Fix up the columns

There will be a lot of columns you don’t need. Facebook won’t use them or know what to do with them. Delete all of the columns except for the five you will be using to get the most accurate Custom Audience matches within Facebook. Those are Name, City, State, Zip Code.

Now here’s where it gets a bit tricky. See, you only need the name, city, state and zip code. So why include the recipient name? Because Facebook needs the first and last name in separate fields. However, Amazon’s reports give the first and last name in the same cell. So they have to be manually separated.

You can manually separate the name into first and last, craft an Excel formula to do it for you, or use a virtual assistant to help you with this.

Now that you have a customer file with the proper fields filled out, head over to Facebook’s audience manager.

Step 3. Create a Custom Audience

  1. Click Create Audience
  2. Select Custom Audience
  3. Choose the Customer File option
  4. Choose “Add customers from your own file or copy and paste data”
  5. Upload your prepared customer file

Once you’ve taken these steps you will have a shiny, new Custom Audience. Now that you are familiar with the process, hopefully you can see the potential.

For example, imagine segmenting the list. Instead of uploading ALL of your customers, you could only upload those who have purchased your products at a significant discount. That would isolate only the promo buyers, allowing you to effectively target your own list of promo buyers in the future.

Or you could segment the buyers of a specific product, or who have spent a specific amount of money with your brand. Imagine being able to isolate a list of only people who have spent over $100 with your brand.

Now, this segmentation takes more time and effort, because before you delete all the unnecessary columns you’d need to filter through the relevant product or price information. My company has developed a tool called Bridge by SixLeaf which automates the process of segmenting buyers and creating Custom Audiences from them.

Lookalike Audiences

One thing you might notice is that your audience is a little small. Even more so if you segment it. With only a few thousand people, you don’t have a lot of reach with your ads. Sometimes this is OK. If you have a really compelling offer, you can work wonders with even a small audience like this.

In other instances, however, you’ll need a broader reach. That’s when you create a Lookalike Audience from these lists, and Facebook gets really powerful.

For example, take the idea of a list of buyers who all spent over $100 with your brand. Those people have a very specific set of interests, buying habits and income level. Facebook can go find all of the users on its platform with a similar set of interests, buying habits and income level, giving you a highly valuable, laser targeted group to run ads to.

It might sound a little like magic, but Facebook can really do that. Many people pour their lives into Facebook, so there’s many things that they know directly, and what they don’t know they can make amazingly accurate assumptions about using their algorithms.

This is what makes your Facebook ads really work. The increased relevancy will lower click costs as well as lead to more sales and engagement (depending on your goals).

Here’s a real world example. This first campaign used a Custom Audience manually created using the results of intensive research into potential buyers.

Clicks were expensive and the campaign didn’t run for long due to it not being affordable.

This next campaign used a Lookalike Audience based on a segment of full price buyers (i.e. who didn’t use a promo code).

There’s a 10x difference in click costs. Traffic increased dramatically, as did sales and profit.

So now what?

Creating a targeted audience is a key part of Facebook advertising, but only one part. How to create and optimize ads to build your brand is a topic for another day, but here are some ideas to help get your thought process going:

  • Promote products or special deals you have on your own website
  • Advertise your products on Amazon, either direct to the listing, or via a landing page where you collect emails and give a discount code.
  • Ask people to sign up for your mailing list, with exclusive weekly discounts and offers.
  • Get them to join a Facebook group related to the products you sell. Use the group to build a community and occasionally put out special product promotions.

Outgrowing Amazon

Despite all the limitations, you CAN use Amazon buyer data to create targeted Facebook ads.

Increased relevance and clicks, with decreased costs, is a recipe for growth. These audiences can be used to drive traffic to your Amazon listings and grow sales on the marketplace, which in turn helps conversions and ranking, further allowing you to grow on the site.

However, they can also be used to drive sales on other platforms that you may have more control over. And that is how Amazon’s customers become your customers.

This post was by Anthony Lee. When his nose isn’t buried deep in testing Amazon’s search algorithm, Anthony writes books about selling on Amazon. You can connect with him directly through Facebook Messenger here.

Amazon Logo
The best tools for your Amazon business
Automate pricing, request feedback and reviews, improve your marketing & SEO and a lot more. Amazon consultants help with difficult issues like suspension, or help grow your business. All in the Web Retailer directory.
View Amazon Selling Tools Now

13 comments on “How to Make Amazon’s Customers Your Customers Using Facebook Ads

  1. Hi Lee, thanks for the great article!
    Just to clarify…Amazon strips the phone number and email address of the customer from the orders data. From my testing (in the UK) it seems that facebook needs at least one of these fields to be able to match the customer with a facebook account. Am I missing something?

    1. Actually, Facebook does NOT need an email or phone number. The customer file custom audience option allows you to create an audience with just first and last name, city, state and zip code.

        1. Are you sure you are getting 0%? I ask because Facebook recently took away the audience size estimation feature from custom audiences created without email addresses. The audience is still there, still with plenty of people in it, but Facebook just doesn’t report how many. So it will show as <20 or whatever their default is.
          A great way to test this is, if the data you input is greater than 1000, input it and create the audience, then try to create a lookalike. If it lets you, that means you had over 1000 matches.

  2. The downloads we get from amazon do not include the name fields, we do not use amazon to fullfill our orders – our reports only show buyer address details along with product details.

  3. Unfortunately for this to work you have to have the name. If you don’t use Amazon to fulfill and don’t get the name in your reports, you’ll need to extract those names from elsewhere (I assume if you self fulfill you get the name from somewhere). Then input the names separately. You can also attempt to create the audience without the name. Facebook may be able to match with just address, but I’ve never tried.

  4. Hey Anthony, I love your article! I have been looking for ages to find a way to do this.

    However, I’m having some problems. About 2 days ago I uploaded a file containing around 900 customers, the audience says it’s ready but it shows and audience size of -1 one (2 days later).

    Moreover, fb won’t let me create a LLA as it claims the the audience is too small. I’ve confirmed that for greating a LLA the msource file should contain a minimum of 100 people. Does this mean that out of 900 people facebook was able to match less than 100 people.

    Any tips on how to make this work? It can have an insanely big impact on my ecommerce business as I’m transitioning from amazon to shopify!

    Once again, thanks for your great article!

    At the same time, the FB does not create the LLA audience! Let

  5. Hey there.
    Ok, first off, the reason Facebook is reporting an audience size of -1 is because right now FB has turned off that feature temporarily. They made a public post about it but basically their devs are working on fixing an exploit they found. The estimated audience size will return as a feature one day, but we don’t know when. However THE AUDIENCE IS STILL THERE.

    Now, to address your looks like issue, you actually need at least 1000 people to create a LA audience. So it appears you are just a little short.

    I hope this info helps.

    1. Hello Anthony,

      Thank you for your fast response! I have created LLA audiences in the past with less than 1000 people. Also in the link below FB mentions the source file should contain a minimum of 100 people.

      https://www.facebook.com/business/a/custom-to-lookalike-audiences

      Evenmore so, while creating the LLA fb gives me an error that the source file should atleast have 100 people.

      This would mean that out of 900 people FB was not able to match a 100. Do you have any tips on how to fix this problem?

      1. I don’t recall where I read it but I read recently the audience needs to be 1000.
        It’s possible you are reading an old bit of policy.
        Further I can confirm that I have an audience with a bit over 200 people I used to create a custom audience and it was “too small” to create a lookalike. Yet my other audience of 1800 people created a LA just fine.
        Regarding the statement that FB says your source file has less than 100 I wouldn’t trust it’s accuracy since FB disabled the audience count feature.
        It is POSSIBLE those 900 only yielded less than 100 matches, though with address information properly input I’ve never seen a match rate so low.
        My advice is to try and get your source file to over 1000 people and try again.

Leave a Reply

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