How to Use Ecommerce Data to Understand Customer Behavior
Understanding customer behavior is one of the best ways to create an ecommerce strategy around customer satisfaction. It’s no secret that when customers are pleased, they’ll come back (and even spend 67% more over their first three years with you, according to Bain & Co).
So how can you best satisfy customers so they keep coming back to you?
Understand their behavior, and adapt your user experience around it.
The best way to know how your customers behave is through the data you likely already have. There are tons of KPIs that offer you a little window into what your customer is looking for, what they’re trying to achieve, and why they would leave without purchasing.
With so many KPIs and data points to track, it can be difficult to spot issues and changes in customer behavior over time. In this post, I’m going to cover:
- What key metrics concern customer behavior and what they mean
- How to spot important changes and interpret data
By the end of this article, you should have a clear understanding about how customer behavior is displayed in KPIs and how to optimize your online store in response to them.
How Customers Behave when Purchasing
There are several metrics to consider when analyzing your data to discern customer behavior. By themselves, they may not tell you very much. But when combined, they will paint a clear picture of what your customer is looking for, and where there might be friction in your UX.
Key Metrics that Indicate Customer Behavior
Unique visits will show you how many people come to your site from each particular channel. While visits are largely a vanity metric and don’t tell you very much on their own, when you combine them with other actionable metrics, they can help keep perspective.
Average visit length:
This metric in particular tells you how long customers spend on your site. If this average is particularly low, it might indicate UX problems or other friction on your site. Combine this with pageviews, conversion rate, and device metrics to get a clearer picture of the story.
Pageviews per visit:
Think about how many pages a customer needs to visit in order to purchase. If your pageviews per visit statistics are low, like average visit length, it could be a UX issue.
Visits before order:
While the average for our merchants is 5.5 touchpoints with a brand before purchase, your customers might need more or less. New customers might need more on average to build trust before buying. Knowing this average will help you understand more about your customer journey.
AOV is a great metric to track because it tells you exactly how much your customers are willing to spend, and if you can calculate this by channel, you’ll know which channels bring in the highest-paying customers.
Each of these metrics tells you something a bit different about how your customers behave While they don’t offer an in-depth look just yet, these metrics show you the surface of customer behavior:
- How they come to your site
- How long they stay on their site
- How many pages they look at
- How many visits they need before purchase
- How much they’re willing to spend
Most importantly, these metrics set us up to dive into a few other data points that will allow us to analyze customer behavior in a more profound way.
How Customers Behave On-Site:
Once you understand where your customers are coming from, you can look at how they behave along your sales funnel. This kind of profound analysis means thinking critically about your analytics and asking yourself what those numbers mean.
There are a few rates that will help you understand how your customers behave once they arrive on your site:
Bounce rate is the rate of visitors who come to your site and leave without navigating past the landing page. You could argue that perhaps those who bounce aren’t your customers, but it’s important to keep this rate in mind when looking at other metrics because it offers perspective.
For example, a high bounce rate could mean that you’re not targeting the right customers and they’re not interested in your products/store. However, it could also mean that you need to work on a landing page that entices your customers to go deeper into your site and start looking at your products.
Failed Discovery Rate (or Browse Abandonment Rate):
This metric shows you the amount of people who browsed through your online catalogue but ultimately left your site without adding anything to the cart.
There’s a few reasons a customer might do this: they could be window shopping, getting ideas for later, or maybe they didn’t find what they were looking for. A great way to combat a high failed discovery rate is through personalization- offering the right products at the right time to the right customers.
Checkout Abandonment Rate:
Checkout or cart abandonment rate means the customers have added products to their carts but left during the checkout process at some point. This is a highly valuable metric because it can indicate issues in your UX or other obstacles in your checkout process.
Perhaps the most important of all, your conversion rate shows you how many of your visitors make it all the way through your sales funnel and become customers.
These rates help give context to the metrics we looked at earlier because they have a direct impact on your revenue. Now that we’ve covered what each of these metrics mean, we can examine how to interpret them practically.
How to Interpret your Data and Spot Opportunities for Improvement
Sometimes, looking at a lot of data can be scary. It can be hard to wrap your head around the numbers when there are hundreds of data points staring you down.
I suggest breaking it into chunks and looking at it bit by bit. Look for numbers that are wildly different from the others and ask yourself why that could be.
For example, this is a typical Divvit customer behavior report with the metrics that we looked at earlier. It’s a great idea to separate these metrics by channel so you can see what kind of customer your channels are bringing in.
Let’s start with conversion rate: here, we can see that there’s a great conversion rate for nearly every channel. However, there are two channels that really stand out in both good and bad ways.
We can see that Google AdWords has a conversion rate of about 5.6%, which is quite high. In fact, it converts higher than organic Google search (5.5%), but organic Google doesn’t bring in nearly as many visits as AdWords does.
What can you do with this knowledge? You know that organic Google search and AdWords have nearly the same conversion rate, which are quite high compared to the rest of your channels. But the big difference here is that organic Google traffic is free.
Perhaps an SEO audit or new keyword strategy should be employed to bring in more traffic from organic Google so you won’t have to rely on AdWords. This is a measurable strategy you can put into place based on the data that you already have.
Now take a look at some of the lowest conversion rates:
Retargeting isn’t doing much for you, it’s not bringing in the visits and what it does bring in isn’t converting. Perhaps it’s time to reallocate the budget you’re spending on retargeting into other campaigns. Referral looks good in terms of conversion, but ultimately has a high bounce rate. Perhaps you should retarget some of the referral campaigns you have in place.
It’s also important to look at your data to make sure that your customers aren’t experiencing problems when they browse on mobile. Let’s take a look at another example:
For this case, mobile brings in three times as many visits as PC and tablet combined. Mobile looks great across all fronts, and even has the lowest failed discovery rate. However, the checkout abandonment rate is nearly twice as high as tablet, and is 32% higher than on PC.
This is a clear indication that your UX on mobile is great for navigation, however, something is going wrong when it comes to checkout. Though the conversion rate isn’t bad, removing the potential friction in your checkout could yield a conversion rate as high as tablet or even PC.
It could be as simple as not having a number pad for long credit card numbers or not offering a favorite payment option. Considering that this is your biggest contributor in visits, it’s definitely worth looking into.
These are just a few examples of how you can interpret your data to understand customer behavior. If you take a look at your data and think critically about what the metrics mean, you can spot the biggest issues that your customers are having.
Being able to spot changes in customer behavior can tell you a lot about where you need to improve. Your data always has the answers, you just need to look for them. By combining the right metrics, interpreting your data correctly, and creating concrete goals that you can test and track, you can use your customers’ behavior to your advantage.