The Hidden Cost of Returns: How Predictive Analytics Can Save Your Bottom Line

Share:

According to Shopify, ⅓ of all products purchased online are returned. 

41% of customers place online orders with the intent to return some of the items. 

80% of shoppers are less likely to shop from the same retailer again due to a poor returns experience. 

One thing is clear. Customers have made up their minds – they now expect returns to be fast, easy, and free.

A true catch-22. 

Offer free shipping and returns, and your brand will eat the costs. Don’t offer free shipping and returns, and your brand will lose customers. 

And while we can’t change the consumer mindset, as digital retail marketers, we can strategically spend our ad dollars to minimize customer returns and maximize profits.

It’s all about attracting the right customers and keeping them for life.

So, let’s talk about how.

Customer Returns Meet Free Shipping

Amazon’s scale and prevalence created a new standard for online shoppers.

Nearly 80% of people who shop on Amazon say they do so because of fast and free shipping. 

36% because they have a Prime membership, and 23% because of the free and low friction return experience.

And with 37.6% of total eCommerce sales in its pocket, Amazon’s model has forced other retailers to follow suit, even when it is financially straining.

The True Cost of Returns.

While consumer-friendly, offering free shipping and returns comes at a steep price.

On average, processing a return can cost up to 59% of the item’s original selling price. 

When a customer returns something, there are costs for return shipping, reprocessing the item into inventory, or even discarding the item at times.

Last year, product returns cost US retailers $743 billion.

And this cost doesn’t even include the advertising dollars wasted acquiring orders with good top-line metrics but low post-return value.

So, what can we do about it?

Well, most retailers typically do one of three things.

1. Tell shoppers to keep unwanted purchases, alleviating the need to process returns and manage inventory that will likely end up discarded. 
2. Attempt to reduce the cost of processing a return, which may not be feasible for smaller retailers. 
3. Offer better product information or digital “try it first” features.

But what if there was another way?

This is the very question we asked set out to answer. 

How can we help digital retailers be proactive and not reactive? 

What if we could predict how much the customer will return at the time of order without waiting until the end of the return window?

It turns out we can do just that.

How Predictive Analytics Optimizes Advertising Campaigns

Leveraging first-party order and returns data, we developed customer purchase history profiles and over 100 unique features tailored to each order’s details, such as the number of sizes of the same product ordered or the total number of different products in an order.

We then incorporated these features into a custom predictive model and began our analysis.

This approach allowed us to identify customers with a higher likelihood of returning their orders.

So, what insights did we uncover?

Customers who have returned significant portions of their previous purchases often continue this behavior. 

Additionally, we identified other telling patterns: ordering the same item in multiple sizes or purchasing several items within the same category (e.g., dresses) often signals that a customer is planning to try on items at home and return those that don’t meet their expectations.

Why Does this Matter?

Understanding customer behavior before the return window closes allows digital retailers to take preemptive actions that enhance profitability. By accurately predicting order values from the start, businesses can leverage this knowledge to:

1. Optimize Ad Spend: Use predictive insights to guide ad platforms; tailoring campaigns toward customers with a history of higher post-return revenue ensures marketing budgets are spent effectively.
2. Build Targeted Audiences: Identify and focus on customers who are less likely to make returns. Create audience segments based on return propensity, allowing more strategic investments in customer acquisition.
3. Enhance Customer Retention: Develop customer bases that are both profitable and loyal over time, ultimately minimizing inflated revenue figures that don’t translate to long-term gains.

This predictive approach isn’t just a reactive cost-cutting measure; it’s a way to shape more profitable customer interactions from the start, setting a precedent for smarter growth strategies and customers that stick around.

Proactive Measures for Maximized Profits

So, before celebrating a successful campaign with record-breaking sales and sharing the news company-wide, consider how much of that revenue might be lost to returns.

With the help of predictive analytics, your brand can attract high-value customers while still offering free returns and shipping. 

It’s a win-win.

Do you want to go even deeper and learn exactly how we built this custom predictive analytics model? We like the way you think. Download our guide or reach out to us today.

Kelly Weeks

Kelly Weeks is Director of Analytics at Adlucent, overseeing advanced and predictive analytics for the agency. A dynamic leader with 15 combined years of experience in marketing, data analytics, business operations, and retail, she is skilled at turning raw data into insights and building systems to use that data at scale. A technology and machine learning enthusiast, she loves exploring patterns in data, finding meaning in numbers, and using AI to make everyday tasks— like writing this bio— easier.