Why are some products showing up in the wrong categories?

We follow the taxonomies the retailer provides, which are sometimes inconsistent or unintuitive.

Occasionally you may notice that a product you are interested in is appearing in a category that doesn't make sense to you. In some cases, the product actually is selling in multiple categories, some of which may be unrelated to the category you're interested in. In other cases, the product is miscategorized, either due to a retailer taxonomy error, a seller error, or some other unknown error.

In the first case, we've seen products appear in unexpected categories that further research showed was due to the product expanding into a new market with a new purpose. However, most of the time when we see unexpected categorization, it is due to an error on the retailer side.

For example, here is a screen capture of the Home Improvement category on Walmart.com taken 11 October 2023:

What's going on here? We're not certain: it could be an error in the Walmart.com taxonomy, or perhaps some of these items are popular in an obscure subcategory under Home Improvement.

In any case, this makes a mess of both the top level Home Improvement department, as well as any of these products' taxonomies.

What can I do about this?

Unfortunately this sort of error happens across all retailers from time to time. Trying to tease apart which products are appearing in new-but-unexpected categories and which products are simply the victims of a retailer error is difficult to automate at scale.

Analytic Index aims to provide an accurate view of the digital shopper experience across many retailers. When we gather data from these retailers, we capture the products as they appear across all departments and categories, which is then reflected in our data.

In most cases, the retailer (or seller) eventually notices and corrects the error within a few hours or days, in which case our data will also reflect those corrections. In the situation where the retailer does not correct the error, our own data will also reflect error.