Adventures in Searchandising Part 1

As a natural extension of the word merchant, merchandising often refers to the methods, practices, and operations retailers conduct to promote and sustain certain categories of commercial activity. According to Wikipedia, the term, “merchant,” is understood to have different specific meanings depending on the context. (And remember, Wikipedia is always authoritatively right, at least according to Google.)

Yet when context is illusive, search engines struggle to understand the contextual meaning of a search query as associated with a phrase’s common extensions. For example, enter a search query for “merchant” and query refinement kicks in, asking if you want associated results for “merchant account,” “merchant bank” or the once popular but always contextually well-targeted “natalie merchant.”

For this discussion, let’s stick with the U.S. definition of merchandising, which is commonly used as point-of-sale retail marketing tactics. Unified merchandising, however, goes far beyond the point-of-sale, striving to coordinate production, supply, distribution, marketing, advertising, display, and more multifarious sales strategies, such as product positioning in movies and videos, to increase retail sales. Unfortunately, these contextually divergent yet fungible retail merchandising strategies collide to absolutely, positively tank an online retailer’s search engine results for all but brand-specific search queries.

Why, you ask?

Because merchandising strategies for brick-and-mortar retail stores don’t necessarily translate well for search engine referrals when incorporated into virtual, online retailer environments. Add a layer or 57 of behavior targeting and personalization tactics to online merchandising and you’ve got “searchandising,” which works exceedingly well for internal site search functionality and usability, but not so much for Google, Yahoo, and MSN Live Search.

Here’s just a couple of reasons why this is so.

When you walk through most physical retail stores, they have categorical signage in place that catches your eye and draws you to where you want to go. For an online retailer, this is the equivalent of internal site navigation. Both present topically oriented, high-level, visually prominent positioning of categorical contents in the store and on the page.

The words used to define the categories are critically important as is proximity with similar goods. After all, you wouldn’t expect to find the toothpaste next to the dog treats. Unless you were shopping for dog toothpaste, then you’d logically conclude that the dog toothpaste is proximate to the dog treats, not the toothpaste associated with personal hygiene products. Still, it’s the contextual relevance of the words on the sign or the words on the page that help you find what you seek.

Search engines strive to understand the contextual nuance of what people are actually shopping for based on their personal search history. To Yahoo, this historical collection tends to be session based. This means that mom’s search results for “bass” brings up shoes or recipes, dad’s search for “bass” returns fishing supplies and services, and Bobby’s search for “bass” returns items associated with music and guitars, even when they all share the same computer.

To Google, search history is based on the individual’s login details. So when there’s a shared computer in use, and Bobby leaves his login to Google’s Gmail live, mom might struggle to find the bass recipe that dad said she should dig to help prepare his catch of the day.

Google personalized search is currently user login dependent, unless default settings have been changed. This stinks, because it’s up to the user to make the search engine work better for the user. It’s a lot like having to ask for help to find something in a retail store. You can never find someone when you need them the most.

Recommended product functionality in an e-commerce Web site often stinks too, because the recommendations are based on what you’ve bought or searched the site for before. Really, how many copies of “The Long Tail of Search” does a person need? Buy that CD of “The Rosary with the Pope” for an ultra-religious aunt and you’ll be seeing the best of gospel music in your recommended items for the rest of your life.

Recommended products based on internal searching and purchasing activities almost always fails, especially when offering up different items to someone who’s shopping for a lot of different people. The results all tend to be equally irrelevant, forcing users to continually refine their search queries. That’s also how some of the major search engine work based on patterns garnered from similar search histories. Once again, over time, results become equally irrelevant.

Toward this end, online retailers have embraced content management systems that help refine navigation within the merchandise’s context. Search engines have expanded their collection of contextual suggestion to help readily refine search queries. All these factors combine to change the way we searchandize e-commerce Web sites. In Part 2, we’ll drill down what happens when the major search engines meet guided navigation and extreme pagination.

Related reading

Brand Top Level Domains