Ask Google, Facebook, Uber or any of the $billion startups (the so called unicorns) out there if metadata runs the world. A friend sent me a book titled, simply “Metadata.” In the jacket, the summary read “By the end of the book, readers will see metadata everywhere. Because, the author warns us, it’s metadata’s world and we are just living in it.” Now that is something I could discuss over a glass of Cabernet in wine country. The author, Jeff Pomerantz, goes on to say:
“Metadata has become a type of infrastructure, like the electrical grid, or the highway system.”
“Metadata is not just data about data – it is the means by which the complexity of an object is represented in a simpler form.”
Consider that any object can be described in dozens or hundreds of ways. The first thought was “infinite”; however, most everyday objects are commonly described in only a few ways. Depending on the level of specificity and granularity, though, the number of characteristics increases dramatically. One can describe one’s home in a variety of ways; for example, by its address, the style or color of the home, its geospatial coordinates, or the number of windows and doors. We can continue to “componentize” the structure and describe the type of siding, the manufacturer of the windows, and the roofing materials. As we go inside the home, we can describe rooms, dimensions, appliances. An appliance can be broken down into its components. Each component can have part specifications, part numbers, and materials. The objects contained in the structure can be described by a virtually limitless number of characteristics or attributes.
Each of these descriptors is a form of metadata. Other objects in our world create metadata as they carry out their functions. Our phones generate metadata as we use them – not only the calls that we make and receive but also our location, since cell phones communicate with cell towers even when we are not making calls. Every object that we interact with contains metadata – or at least had to contain metadata at some point in its existence. The only way a product can be purchased – even in the physical world – is with metadata, since price is an example of metadata. There is metadata all along the supply chain, from manufacturing and quality control to transportation and logistics tracking and inventory and stocking information at the retailer.
The online experience is essentially comprised of metadata. While shopping in a physical store provides visual cues, online shopping depends on metadata to provide those cues. Metadata is required for routine management of information but it is also the source of unique experiences.
Uber and Lyft serve the same fundamental purpose, but they provide different user experiences based on the way metadata is surfaced. The way the transportation function is carried out for each company is unique.
The same thing happens when we go to target.com or walmart.com. Each experience depends on how the metadata is structured and arranged. The differentiation of metadata provides the personality, character, tone, and unique differentiators for those providers. Metadata enables the unique experience that each brand conveys.
Moreover, without the correct framework, the experience cannot be scaled. Metadata enables the machinery of the online experience. 3M needs to sell to multiple outlets, such as Walmart, Target and Staples, but it also needs to describe its products in a standard, consistent way. This standardization allows for efficient movement of products through the supply chain and allows for the housekeeping of accounting, inventory management and transaction processing.
How then does an organization leverage metadata for uniqueness? If the nature of the product is the same, and there are limited ways to describe that product, how does a brand differentiate the message and experience for a competitive advantage? The answer lies in how the customer uses the product and how the seller can solve the customer’s problems. Many commoditized products are sold as solutions to a problem. It is less about the nature of the individual product and more about how that product is used.
Presenting those solutions in the context of the user’s goal and objective becomes the differentiator. That is accomplished by knowing more about the user and their problem than your competitor and surfacing that combination of products and knowledge in the most easily consumed manner. What other products work together? What is the advantage of one combination over another? What else can you recommend to help me with my problem?
The manner in which the products are described is less important than the relationship to other products and how they are used to achieve the user’s objective. Solutions are highly specific to the audience, industry, user, and application. Knowing the combination of details and embedding domain knowledge into the structure of product data and content architecture provides differentiation and competitive advantage.
If every company presented their products the same way in every context, every store would look the same, and every web site would organize products using the same hierarchy, descriptors, attributes and terminology. Behind the scenes the housekeeping and tracking data about the product is the same. It is on the front end of the process where metadata differentiates. That differentiation is based on the needs of the customer and the segment in which the seller competes.
WATCH: The Paradox of Product Standards & Brand Differentiation: Can You Have Both?
Managing these classes of metadata attributes is challenging. The first challenge is getting the housekeeping consistent. Services such as GS1 provide registries of product data to help remove inefficiencies and provide consistency. The next challenge is defining the ways in which products can be used to solve problems, either by themselves or in combination with other products. This step requires a deeper understanding of the users and their tasks, and knowledge of the best ways to solve the problem.
Effectively using subject matter expertise is one way that companies can compete with online giants such as Amazon, especially in the b2b space. MSC Industrial has more collective knowledge about metalworking processes than Amazon. That is MSC’s specialty and that is how the company built its business. That knowledge is the reason customers stay loyal. MSC’s job as an online business is to embed process and solution knowledge into its metadata structures so the knowledge can be surfaced at scale. It is one thing to be able to pick up the phone and call a representative who can walk the user through a series of problems and recommend solutions. The interaction may be successful but the process does not scale. It is another to capture and codify representative’s knowledge and embed it into the machinery of ecommerce – the metadata schemas and structures, and the content that supports solutions.
Building these structures also lays the foundation for further automation. Next generation search uses these structures along with natural language processing to provide answers to questions, rather than simply generating long lists of documents. Chatbots and virtual assistants use these structures to engage in dialog with users. Search is basically a conversation – a user enters a search term and the system responds with a result. The user then modifies the search (the question) or filters results providing more contextual clues as to their intent. This interaction can easily be extended to chatbots and virtual assistants. The underlying mechanism is the same and the competitive differentiators are the same – the chatbot that understands more about the user and their problem will provide the better experience. This “understanding” comes from the underlying product relationships embedded in data structures.
The actions that improve the user experience and compete in the marketplace are the same ones that will enable next generation user experiences through artificial intelligence and machine learning. Getting this foundation in place is required no matter what the application. Metadata is the enabler. Efficiency through standardization is the beginning of the process. Differentiating by understanding the needs of your users is the bigger story.
For a look into how we use customer data models, product data models, content models, and knowledge architecture to create a framework for unified commerce download our whitepaper: Attribute-Driven Framework for Unified Commerce