The Perks of Return Analytics: The Mystery of the Jilted Jewelry

First in an occasional series of stories about life in the trenches of retail product returns

It was a busy day at the jewelry retailer’s returns department. Lots of tiny boxes opened to reveal the same SKU coming back in droves. Processors noted each returned order and took a quick look at the product condition. Still very good – same as with two different styles of necklace that also had high return rates.

Duly noted and compiled onto a spreadsheet, those SKUs would have joined the thousands of products across hundreds of retailers that come back daily with nothing apparently wrong and no way to figure out why they got returned. In consumer electronics, they call this “no-fault found,” and according to research by Accenture, 68% of returned electronics products fall into this category. Retailers across categories experience this frustrating problem every day.

But this retailer uses return analytics. By applying natural language processing to the return data, the return analytics tool detected a common denominator across the seemingly unrelated SKUs with high return rates: They all featured tassels. Tassels, apparently, suddenly went from a “do” to a “don’t.”

The retailer reacted quickly, pulling several SKUs off the website and canceled open orders. By being proactive on one insight, they avoided thousands of costly returns.

Returns have a lot to teach retailers about customers and what they like and dislike about products. But many retailers maintain rudimentary returns tracking processes that:

Force consumers to choose among pre-determined reason codes that don’t always apply and leave no room for detail or explanation.
Compile reason codes into massive spreadsheets that make it extremely difficult to spot patterns.
Inadequately inspect returned goods and spend little time planning for efficient returns.

Return analytics uses data and root cause analysis to quickly identify difficult-to-detect trends such as the sudden unpopularity of jewelry with tassels so retailers can react quickly to spiking product return rates in their tracks. Rich insights gleaned from data turn returns from mysteries into clear patterns that retailers can use to fix problems and understand their customers as never before.

First in an occasional series of stories about life in the trenches of retail product returns

It was a busy day at the jewelry retailer’s returns department. Lots of tiny boxes opened to reveal the same SKU coming back in droves. Processors noted each returned order and took a quick look at the product condition. Still very good – same as with two different styles of necklace that also had high return rates.

Duly noted and compiled onto a spreadsheet, those SKUs would have joined the thousands of products across hundreds of retailers that come back daily with nothing apparently wrong and no way to figure out why they got returned. In consumer electronics, they call this “no-fault found,” and according to research by Accenture, 68% of returned electronics products fall into this category. Retailers across categories experience this frustrating problem every day.

But this retailer uses return analytics. By applying natural language processing to the return data, the return analytics tool detected a common denominator across the seemingly unrelated SKUs with high return rates: They all featured tassels. Tassels, apparently, suddenly went from a “do” to a “don’t.”

The retailer reacted quickly, pulling several SKUs off the website and canceled open orders. By being proactive on one insight, they avoided thousands of costly returns.

Returns have a lot to teach retailers about customers and what they like and dislike about products. But many retailers maintain rudimentary returns tracking processes that:

Force consumers to choose among pre-determined reason codes that don’t always apply and leave no room for detail or explanation.
Compile reason codes into massive spreadsheets that make it extremely difficult to spot patterns.
Inadequately inspect returned goods and spend little time planning for efficient returns.

Return analytics uses data and root cause analysis to quickly identify difficult-to-detect trends such as the sudden unpopularity of jewelry with tassels so retailers can react quickly to spiking product return rates in their tracks. Rich insights gleaned from data turn returns from mysteries into clear patterns that retailers can use to fix problems and understand their customers as never before.

First in an occasional series of stories about life in the trenches of retail product returns

It was a busy day at the jewelry retailer’s returns department. Lots of tiny boxes opened to reveal the same SKU coming back in droves. Processors noted each returned order and took a quick look at the product condition. Still very good – same as with two different styles of necklace that also had high return rates.

Duly noted and compiled onto a spreadsheet, those SKUs would have joined the thousands of products across hundreds of retailers that come back daily with nothing apparently wrong and no way to figure out why they got returned. In consumer electronics, they call this “no-fault found,” and according to research by Accenture, 68% of returned electronics products fall into this category. Retailers across categories experience this frustrating problem every day.

But this retailer uses return analytics. By applying natural language processing to the return data, the return analytics tool detected a common denominator across the seemingly unrelated SKUs with high return rates: They all featured tassels. Tassels, apparently, suddenly went from a “do” to a “don’t.”

The retailer reacted quickly, pulling several SKUs off the website and canceled open orders. By being proactive on one insight, they avoided thousands of costly returns.

Returns have a lot to teach retailers about customers and what they like and dislike about products. But many retailers maintain rudimentary returns tracking processes that:

Force consumers to choose among pre-determined reason codes that don’t always apply and leave no room for detail or explanation.
Compile reason codes into massive spreadsheets that make it extremely difficult to spot patterns.
Inadequately inspect returned goods and spend little time planning for efficient returns.

Return analytics uses data and root cause analysis to quickly identify difficult-to-detect trends such as the sudden unpopularity of jewelry with tassels so retailers can react quickly to spiking product return rates in their tracks. Rich insights gleaned from data turn returns from mysteries into clear patterns that retailers can use to fix problems and understand their customers as never before.

The Perks of Return Analytics: The Mystery of the Jilted Jewelry

Related articles