Manish ChowdharyFounder & CEO pioneering AI-driven ecommerce logistics

Your Ecommerce Return Rate Is Hiding the Real Problem

16 Sept 2026 · 2 min

Watch on YouTube

1:41

In this panel discussion, the speakers emphasize that ecommerce return rates cannot be interpreted as a single aggregate percentage. Instead, they must be broken down through cohort analysis that separates customers and product types. They explain that return behavior varies dramatically across consumer packaged goods (CPG) verticals—apparel typically sees high returns due to fit and style trials, while cosmetics, food, and beverage have low returnability because opened products cannot be resold. By examining cohort metrics, such as the percentage of returns per SKU, the panel shows that a product returning 85% of the time signals a poor‑selling item unlikely to become a hero SKU, suggesting the need to discontinue or redesign that product. The overall message is that a superficial return rate hides deeper operational and product‑selection problems that can be uncovered through detailed cohort analysis.

Chapters

But a first-time shopper coming in, so from your presentation, Jay, you have to distinguish, and this is what they call the cohort analysis. Jay or Joe, if you want to talk about this, that you cannot look at return as a return percentage and divide all your returns because it's a lot of different cohort analysis. And then things begin to make sense if you pay attention to that.
Yeah, no, without a doubt. And I think that everyone's making really good points. I think that within the different verticals of whatever type of CPG brand that you're going to be, your return rates are going to differ. So obviously in apparel, there's going to be a much higher return rate because you want to try a different color, a different fit, a different size. And there's kind of this knowledge within the apparel field where you can try something on and return it, and that can just be refolded, retagged, rebagged, put back in, and resold. Whereas from cosmetics, food and beverage, if you open up a package or if you use a cosmetic, that good's really not going to be able to be used again unless you know it's really lightly or a perfume or some— so you have to really look at what that good is. Now from a metric standpoint in the cohort, knowing what your return rates and what units in the metrics look like, as Gabrielle said, if a SKU is being returned 85% of the time, you can probably kind of insinuate that that's not going to be a high-selling SKU. It's not going to be one of your hero SKUs. If somebody is continuously buying that and it's being returned 85% of the time, Perhaps we don't manufacture that excuse.