[davb]

I work at a company doing retail return analytics (pretty niche) and we've found that looking at returning customer segments gives a wider picture. In many cases, high returning customers can be your most profitable in the long term even after considering cost of returns. Particular segments keep more than any other (from a financial perspective) despite being returning a disporportionate amount of the things they buy. They're the most loyal, most tolerant and often the highest value customers.

[losteverything]

What are your "return segments" or whatever you call them?

I work in a store that has, on average, 200-250 returns a day (1/10 sales/traffic). It's a very popular store.

I would segment this way (without any data, just envelope) in no particular order:

Buy A Lot & Return Some or All

Occasional Honest Jane

Mad As Hell With Wild Expectations

Scammers (part 1); Return counter is a bank

Scammers (part 2): Free lease of equipment

Scammers (part 3): Online Returns

Scammers (part 4): Garage Sale Pickups

Scammers (part 5): Anything Electronic or Gift Card-esque

Scammers (part 6): Stolen Goods

Scammers (part 7): "Last 15 minute Jack"

Shoplifter

I clearly see that we affect customer behavior if we turn the screws on our policy.

[bitL]

Clothing retail is super low margin unless you sell high-end luxury goods. This approach eats basically all the profit and it's not worth for us to have an average gain of $0.01 on such a customer; we go gladly without them. It's also like some photographers buy an outfit for a photoshoot and return it back right after they are done. We can then only sell it off for scraps on eBay etc. giving us both headache and unnecessary work. If Amazon wants those customers and burn money on them, let them.

[davb]

We work primarily with fashion retailers. Our customers have fairly high margins even though their clothes range from low end to high end.

Wear and return is certainly prevalent though our customers see it more often in occasion-wear (for example bridesmaid dresses). Their average customer values (even their highest returning customers) far exceed $0.01. I'd be really interested in learning why things are so different at your organisation (not that I doubt you but it would make for a really interesting dataset!)

We generally try to help clients integrate returns prediction into their business rules. For example if you've got scarcity of stock, perhaps prioritise customers who are most likely to keep items. Or if you're handling customer calls at the contact centre, prioritise customers who are returns sensitive (those who probably won't shop with you ever again if they have a poor returns experience).