Most people who sell a product know that keywords are essential for effective marketing. The right keywords will increase the likelihood of getting a brand’s message to the target audience and improve where a product appears in search results. For eCommerce, it’s common practice to follow the product hierarchy in determining which words to use in titles and product descriptions: category, brand, flavor, size, and key benefits or attributes of the product.
This is a carryover from how consumers are used to shopping in physical stores. Stores arrange their products by departments, sections, aisles, and categories. Within a category, several factors come into play to determine how products should be grouped to make it easier for shoppers to find and purchase products. Shoppers quickly scan the shelves to locate familiar products and pause if they are looking for less familiar items or for something new. If shoppers have dietary or lifestyle needs, like gluten-free, low-sodium, organic, or vegan, it’s up to them to ensure they buy the right product. Visual cues and the information on the package are crucial. For the most part, products are where they are supposed to be, and it’s easy to see if a product is out of stock – the place on the shelf is empty.
Now, let’s shift back to online shopping. There just aren’t those patterns and visual cues like there are in store to assist a shopper in completing the trip quickly. While all sites have navigation menus that are based on product hierarchies, these navigation trees can be cumbersome. Users need to know where a specific product or category might be buried. After they go through the effort, they still have pages and pages of products to scroll through. It’s no wonder why consumers almost universally rely on Search. However once a search is fired off you they are now at the mercy of algorithms powering the search. Often times search results seem almost endless and upon searching consumers might find they want to dial in these results further to their liking.
Enter filters. Filters are the most effective way to narrow down the results to only the products that have a specific designated criteria. Unfortunately, these filters don’t always work well because the underlying data are neither complete nor accurate. If a product isn’t tagged with an attribute in the underlying data, the system will filter it out even though it could be exactly what the shopper is looking for.
The same issue holds for the navigation trees. For example, on a large retailer site, a few cans of soup had automotive tires as the category instead of soup. Of course, this was not replicated in the physical store because even if the person stocking the shelves had access to that data field, they would not have followed it. Algorithms don’t have that type of override.
The data issues have several possible causes: the manufacturer or retailer incorrectly entered the data (wrong data or a typo), the manufacturer or retailer did not have the correct information to enter the data, insufficient data cleansing or validation procedures, or lack of ongoing data quality management processes. Correcting data quality requires a combination of technology, people, and processes.
Allium AI has the technology to plug into your platforms and processes. We’ve developed proprietary technology to complete the underlying product data of a retailer’s product catalog or a manufacturer’s portfolio of products based on all the visual and unstructured information you have. Since it’s based on the most recent version being sold, retailers and manufacturers can rest assured that the data is correct.
Allium’s system pulls out these keywords as key attributes to populate structured systems and leverage in compelling & optimized content to ensure your products are fully discoverable however the shopper navigates their buying journey.