For 2022, collecting customer data will be the most important element of your Amazon marketing strategy. Without data, and more importantly the insights derived from it, you’re firing blind -- possibly reaching your target audience, but not optimizing the results.
As far as data is concerned, Amazon has historically been viewed as a “black box” -- but this is changing. Amazon’s increasing reliance on 3P Sellers as a revenue source has corresponded with an increasing flow of information from Amazon to Sellers. 1P Vendors have access to the same data reporting tools as always, but the general increased volume of customer data within the Amazon ecosystem amplifies the need for everyone to apply data and stay competitive.
Here, we are going to outline a strategy that will help you get ahead of the curve and use your Amazon customer data to your advantage in 2022 and beyond. If you want the more advanced version, we have written an eBook on exactly how you can make sense of your Amazon customer data.
Step 1: Know why you are looking for data
To have success with any data-based operation, you need to understand your goals. That starts with identifying the main reasons data will help your brand. For example, you could be looking to launch new products, increase sales for slow-moving products, reduce a high ACoS, or win a greater share of a market.
Identifying your main business objectives allows you to narrow in on the data that you find important. If you take a manual approach to data analysis, this is particularly important.
The next step is to segment your customers. This is critical to applying your data in a way that makes sense. It can also help you collect data. You should be looking for data that allows you to segment personas and craft “typical” pictures for different customers of different product lines. Depending on your product portfolio, some of the segmentation you could look to build includes:
Life Stages: For example, will single adults shop differently than adults who have recently had a baby or have grown-up kids.
Behavior: Understand why and how customers act. For example, When? How Often? Who/What?
Geography: If you’re looking for customers close to a physical location.
By identifying key goals, you have a template that you can then start to fill out with information. By applying that data to a segmented framework, you can start to make sense of that data in a way that will then let you take action. But if you want to do that, you need to know how to extract data from Amazon in the first place.
Step 2: Keep up to date with the data Amazon makes available
If you want to build better customer profiles, you need to understand what data is available and how it’s presented. For better and worse, Amazon is constantly updating what customer data it makes available, and the channels through which that information is delivered.
For 3P Sellers, 2019 was a year of big change. For registered brands, Amazon rolled out a new interface, Amazon Brand Analytics (ABA), free-of-charge. This allows eligible retailers to analyze product performance in terms of item comparisons, demographics, search terms and basket analysis.
Amazon Marketplace Web Services (Amazon MWS) remains the only source of transactional data. As a Seller, you may already have a system set up to use Amazon MWS. Conversely, it’s possible that you are fully unaware of its existence. As an integrated API, Amazon MWS requires Sellers to either build or purchase an application to transform the raw-data feeds into actionable information.
The challenge of interfacing with Amazon MWS is exactly why ABA has been hailed as such a big win for Sellers on Amazon. There is great info contained within ABA. However, if you want a comprehensive view of your Amazon customer data, you will need to supplement ABA information with data pulled from Amazon MWS, along with PPC campaign reports. It’s also worth noting that ABA data is currently only available for the last year -- meaning you will need to export that data to craft long-term trends. The most basic solution here is to export that data as .CSV files, but this can become rather unwieldy.
Third-party analytics tools remain the most powerful way to collate all of the information available on Amazon -- and the increased access to data is improving the functionality of these tools as well. But the basics remain the same. Once you understand what you want from your data, you need to understand what data Amazon has made available. Then you can start to think about how to put that data to use.
Step 3: Analyze reviews and reviewers to add to demographics and customer profiles
As we all know, reviews and approval ratings are key to success on Amazon. A higher rating and more positive reviews both drive higher conversions and higher organic rankings. Tools can help you keep track of this, and it's also a pretty simple thing to monitor manually -- depending on the size of your product portfolio.
While you are checking your reviews and ratings, start building up profiles of who is reviewing your products, what else do they review, how do they rate competitors and how often.
When adding this information to your customer profiles, also look at returns and fulfillment trends. Not only with this will give an indication of customer concerns, it can flag possible misunderstandings generated from your product descriptions. You can also see whether they are specific issues that only affect a particular segment or are more widespread.
Create a benchmark for like-for-like products to identify products that maybe are not well presented in their product listing. You can use this information to change the branding and messaging of your listings and see how that affects returns. You can also use this information to align your products more closely with the demographics and personas making the purchases. With enough diligence, you can use reviewer data to get very knowledgeable about product-specific buyers, and tailor listings and ad campaigns accordingly.
Step 4: Integrate buying trends into your customer profiles
An important outcome that can be achieved using customer data is the discovery of specific buyer trends -- or ‘buying trajectories’. These are likely purchase patterns that different customer profiles tend to undertake. This can be calculated in abstract, and overlaid onto the demographic profiles you develop. However, it’s often most practical to craft these trends in regards to the product purchased.
For example, looking at buying trends might tell you that an individual who buys your scented candle pack has a 50% chance of buying a candle holder within the next week. You also might discover that the same person is likely to buy scented bath soap within the next month. Understanding trends like this can help you prioritize PPC bids. It also gets you a long way towards calculating customer lifetime value (CLV), a critical metric for understanding the real ROI of customer acquisition.
The transactional data required to calculate buying trajectories is hidden within Amazon MWS. The complexity of transforming that data into accurate trends requires sophisticated programming. The best third-party tools for crafting these trajectories deploy AI and machine learning to analyze not only Amazon MWS data, but pull reports from across the Amazon ecosystem.
By combining persona data, alternate purchase behavior, basket analysis, long-term trends and purchase outcomes, you build up a unique picture of how your customers compare products and what is important to them when buying.
Step 5: Use your insights to re-evaluate your goals
As you uncover customer data insights, you will shed light on trends and opportunities that you hadn’t previously considered. Although it’s important to start with goals in mind, you should never pass up new opportunities.
Data analysis isn’t something you do once and then forget about, it’s a process that you have to maintain. That means using the conclusions you reach through data analysis to shape how you move forward. Your first review of customer data might be targeted to support a specific product, or simply be a general audit. But that doesn’t mean you need to maintain that specific focus forever.
Regardless of your original intentions, you should take the conclusions you reach and use them to better define the outcomes you are looking to achieve. Over time, you will develop an ever-improving process and unlock more secrets hidden in your Amazon customer data. Think iteratively -- not singularly.
Data is the key to targeted efficiency, but data can also overwhelm
A clear picture of your customer data is the key to effective decision-making and taking targeted actions on Amazon. It can shape your PPC bid decisions, it can impact how you prioritize the listing of products, and it can even help shape the future products you develop. Data lets you understand your customers. Even before the era of ‘big data’, understanding your customers was key to business success.
The more you look to enrich your customer data, the more complex the process becomes. What’s more, analysis is never a one-off activity -- it’s an ongoing process. You need a structured approach to succeed. This is also why software and automation are key to success. The more manual tasks you can strip out of your data analysis, the more resources you will have to allocate to taking action on those insights.
Analysis software will also help implement advanced techniques such as search term isolation to better match your bids and search terms to what you now know your customers are searching for. It will also identify new cross-selling and bundling opportunities for particular customer personas and segments. Not only will this create value, results and insights, but will move your business into a new data-driven culture.
No matter how it’s achieved, what is important is having a robust process for capturing, storing, processing and analyzing data are all key to unlocking the real game-changing sales opportunities locked within Amazon customer data. Start with goals, build customer profiles, and then use the information you have to go back and shape the goals you started with. Through iteration, you can continually hone in on the right answer and build data-based profiles that will help you succeed.