How CMO is Redefining Product Search with Programmatic Display

programatic display

        Category Merchandising Optimization (CMO) is by no means a new breakthrough. It’s been used by brick-and-mortar based retail outlets for years. CMO has been implemented by grocery stores when they selectively place high-demand items at eye level on shelves, and in end-cap displays in relevant isles. CMO is the reason that staple items such as bread, milk, and eggs are all the way in the back of the grocery store, forcing consumers to walk past all the other products that they may be tempted or reminded to buy. It has been implemented in nearly all retail stores where premium products are displayed at the front of the store, whereas clearance items are more commonly found in the back of the store. An engaged retail merchandiser manually studies the shopping patterns of their customers to optimize the store’s layout and item placement strategy. This has been considered a requirement of in-store strategy for many years, and recently the proper technology has been developed to implement a similar prioritization process in eCommerce through programmatic display.

        Without CMO, online shopping can be a lot like a clearance clothing bin, with items haphazardly mixed together. The eCommerce equivalent of this is scrolling through multiple pages of products in a vain attempt to find a specific item. This is typically where the application of product categories comes into play. eCommerce sites often have categories, subcategories, and even further subdivisions of products to organize the available items that your customer is in search of. However, often times a product applies to multiple categories, which leads to an unrealistic understanding of the products demand. For example, when searching via the “most popular” sorting method for an item such as coloring books, as explained by Linda Bustos, colored pencils may appear before the coloring book itself due to other relevant categories the pencils may be included in, such as back-to-school, art supplies, children’s, etc. Bustos goes on to explain that items may show up more often due to a self-selection bias. Items that are vaguely categorized appear more often, leading to more clicks, which in turn continues to result in priority ranking within “most popular” despite potentially being irrelevant to begin with. This can be explained via the clearance bin example mentioned earlier. Items at the top of the bin may be tried on many times (equivalent to many clicks) resulting in them being very “popular”, however, they are returned to the top of the bin (never added to cart) every time, so they are, in fact, not selling well at all.

        Category Merchandising Optimization (CMO) is the strategic, category-by-category approach to organizing merchandising product list pages. This allows for product search results to be sorted under a similar algorithm to that of Google’s search engine result ordering process. Now, an intuitive prioritization algorithm can be applied to product result pages at the level of an individual electronic storefront. This allows your products to be programmatically sorted in relation to purchasing context by category, boost and bury factors, and personalization factors, such as categories and brands previously browsed. CMO combines conversion rate optimization, the efficient comparison of A/B testing, with the prioritization algorithms of SEO to ensure your consumers are being shown the products most relevant to their inquiry.

Buying Context by Category

        Evergreen products, when compared to seasonal products, are exceptionally sensitive to the category under which they belong. Being able to automatically prioritize products which see a spike during specific seasons will likely drive conversions in real-time based on the number of inquiries the algorithm detects.
        The degree of consideration a consumer allocates when purchasing a product relies heavily on what category the product exists within. For example when purchasing single use, consumable goods the extent to which a consumer considers the product’s strengths and weaknesses is lesser than that of a product with continuous application, which requires a great deal of consideration.
        Certain categories should prioritize specific ranking factors over others. This is particularly true when considering products where the impact of social proof (top-rated, best-selling) outweighs specific company-defined factors such as price or the unique selling point of a product’s promoted features. For example, when considering a purchase such as a new pair of shoes, where the comfortability of the shoe can only be honestly expressed by consumers who have previously purchased and worn the shoe to test it. This significantly overpowers promoted features, for no company will intentionally inform you of blisters or a lack of arch support resulting from the product they are attempting to sell.

Boost and Bury Factors

        Boost and Bury rules consist of factors a CMO algorithm can analyze to increase or decrease the prevalence of individual products within your website’s search results. An algorithm that can recognize a successfully selling product will boost this product’s result listing status to the front-page results. For example, if your product receives a shout out from a celebrity you may notice traffic rapidly spiking. A boost rule within the algorithm recognizes this and promotes the product independently in real time. Contrary to this, if you are particularly close to being sold out of a product, the algorithm’s bury rule will reduce the visibility of this item so not to disappoint your customers as they search only to find out the item is unavailable or can only be purchased in an unpopular size or color.
        Boost and bury rules can be affected by factors such as the revenue per visitor and sell-through rate. Both of which would attract a higher priority and need for greater strategy to effectively optimize. If you want to boost the display rate of your house brand over the remainder of your inventory, a boost/bury rule may be applied to enforce the dominance of your own product over others.These changes can theoretically be done manually, however, it is exceptionally demanding work which often times may be too little too late, particularly in reference to pop-culture trends you may not follow. An applied algorithm manages your items prioritization without the need for you to actively adjust your search settings.  

Personalization Factors

        Personalizing your customer’s shopping experience relies heavily on traceable cookies that allow a CMO algorithm to build metadata profiles around each customer’s unique IP address. This profile is built through categories and brands that have been previously browsed. Past purchases and products that have been previously added to, or items currently in, that customer’s cart also contribute to a customer profile. Personalization factors can adjust your site’s content based on how the visitor arrived at your website, whether that means arriving via a search engine, referring site, or a specific advertisement or campaign. This is particularly useful in conjunction with an advertisement. The item displayed in the promotional material may then be instantaneously prominent upon arrival at your website.

        By studying and actively adapting to real-time trends and product availability factors you can optimize your product conversion rates, driving sales to otherwise unprecedented levels. In addition to driving conversion rates CMO will create a more intuitive and satisfying shopping experience for your customers. Manual category merchandising can be performed without the automated algorithms as displayed by Jirafe (recently acquired by SAP Hybris) which displayed a 40% lift in conversions as a result of serious category merchandising considerations.  However as this automated eCommerce trend has just recently emerged, there are a very limited number of companies offering tools for programmatic category and merchandising optimization. If you’re interested in learning more check out SAP Hybris‘ Merchandising tool, Nextopia, Bloomreach, and continue to check in with the 1Digital blog where we’ll be reviewing CMO case studies as they’re released to keep you up to date with the latest news in this exciting new field.


Sourced from:
Category and Merchandising Optimization [webinar]. Dir. Linda Bustos. Perf. Linda Bustos. The Future of Customer Engagement and Commerce. SAP Hybris, 30 May 2016. Web. 2 June 2016. <>.