An ecommerce platform is a software solution used to create a storefront, administer back-end functions, and integrate with other systems (such as digital analytics, BI, databases, CRM, and marketing technology). One or more ecommerce experiences may be delivered from a single ecommerce platform. For example, a mobile site and web site would be served by one ecommerce platform used to create, deploy, and maintain both of them. Ecommerce platforms provide the technical foundation for building the digital experience and managing it—or connecting and integrating with the technologies to power it. You can think of an ecommerce platform as the central nervous system and/or circulatory system for an ecommerce business. The ecommerce platform enables the totality of functionality that creates the ecommerce experience, such as the product catalog and pages, other content pages (like the home page and landing pages, the shopping cart, and promotions).
Ecommerce platforms are available for businesses of all sizes. For smaller businesses, there are a number of software-as-a-service (SaaS) products that can provide the end-to-end functionality needed for running an ecommerce site. For example, these platforms may provide a template-driven, build-it-yourself set of features. Larger businesses and retail corporations may choose to build their own end-to-end ecommerce platform using technologies such as Java, .net, PHP, Ruby, and more (Adams 2015). Still other companies choose to buy a software product that may be hosted (in the cloud) or run by the company within their data center. Whether to choose an on-demand or on-premises solution or a custom, licensed, or managed SaaS will entirely depend on the company. Other companies may choose to integrate open source ecommerce technology or an assortment of pre-integrated ecommerce products used for specific functions that work together based on shared technology protocols. Typically a company will hire a systems integrator to implement and customize a commercial ecommerce platform. Other companies choose to do their own integration.
An ecommerce platform, at the most fundamental level, contains a consumer layer (the storefront), a data integration layer, an analytics layer, a data layer, and an operations and monitoring layer:
• The consumer layer represents the channels for the user experience, which can be a web site, mobile, or other Internet-connected device (Internet of Things, Internet of Everything). The consumer layer will also contain the technology for the shopping cart, product catalog, product pages, and any other programmatic or API-based technology that must be integrated to render and fulfill the ecommerce experience to the end user.
• The integration layer is the set of technologies that bring together the services and technologies in the consumer layer so that the experience actually works and isn’t just a set of nonfunctional screens that display in a browser or smartphone. The integration layer contains the rules and instructions for integrating user data at the session level with shopping, transaction, and service capabilities such that the ecommerce experience is enabled as one shared operational environment that unifies the ecommerce capability from the start of the visit to the purchase.
• The analytics layer is a set of analytics and data science technologies and tools used to analyze and work with data: from cleansing to preparation to modeling to analysis to visualization and so on.
• The data layer is where the data is stored, processed, and queried after being sourced from wherever it comes from. The data could be generated by the operational processing and system integration of services that support your ecommerce platform. Or the data could come from third parties, such as partners, resellers, or data vendors. The data layer contains the traditional relational database management systems, data warehouses, NoSQL technologies, and newer processing frameworks like Hadoop and Spark.
• The operations and monitoring layer contains tools for managing the platform and technologies and can include tools that enable ecommerce operations, such as CRM, inventory, merchandising, finance, customer service, and so on.
From the layers, different applications can be developed that provide for ecommerce site functionality and for the features that create the ecommerce experience. Applications and features in an ecommerce platform can be categorized as follows:
• Content. The content management system, including videos, images, and other publishing capabilities such as advertising, including external sources of content such as social media, partners, and resellers.
• Search. Search-based navigation, querying, and filtering to find products to buy and also to find other content.
• Catalog. The collection of attributes about products used to create the product catalog and content pages, like the product page. The catalog includes the product specifications and details, the pricing and discounts, promotions and offers, the product images (from content), the taxonomy and categories (nodes), the product bundles, and cross-sells.
• Customers. The customer management technologies, including the customer record and profiles that contain customer data, such as gender, name, addresses, e-mail, phone number, first order date, last order date, segment, and other relevant customer data to your business.
• Shopping. Order capture functionality, such as fast-buy (one-click) experiences, long-form checkout and shopping carts, quick-order and one-page shopping cart, and other checkout and buying features and functionality.
• Ordering. Order functionality, including order capture, order management, order verification, order tracking, and so on.
• Payments. Processing, security and fraud, and financial and tax management technologies are payment-related functions.
• Inventory and merchandising. Tools for inventory management, preorders, back orders, and allocation and inventory release to customers.
• Fulfillment. Functionality for shipping and address verification, warehouse management, delivery status, and order tracking.
• Customer service. Capabilities for online self-service, call center, order returns, product exchanges, and site credits.
• Analytics and data science. Capabilities for data collection, storage, processing, preparation, modeling, analysis, reporting, and visualization that enable ecommerce analysis. This includes descriptive, predictive, and prescriptive capabilities created using business intelligence and data science, such as data mining, machine learning, and artificial intelligence.
Another way for the analyst to begin to understand how to think about analyzing the data in an ecommerce platform is to consider it as a set of components. Each component has related functionality and features for a business function that is provided by the underlying ecommerce platform and its many layers. These components are for marketing, user experience and interface, orders and transactions, customers, and data (Kumar 2015):
• Marketing components. In ecommerce the marketing team is concerned about making potential customers aware of the brand, acquiring them to come to the site, ensuring that they have a positive site experience that leads to an order. And then they are concerned with using marketing to retain that customer and make him loyal so he orders again and again. An ecommerce platform, thus, will have functionality that ensures that marketing activities can be executed and managed over time and the resulting data captured for analysis. Enterprise ecommerce platforms have specific marketing functionality already integrated, such as shopping cart abandonment identification and recapture, social media features, and functionality for personalization. Other platforms may provide integrations with ecommerce products that provide video, product photos and zoom, Facebook integration, product up-sell, product reviews, and more. It is important for the analyst to determine how an ecommerce platform supports the analytical needs of marketing to ensure appropriate measurement.
• User experience (UX) and user interface (UI) components. The UX and UI components are the look and feel of the site or mobile experience. The UX will generally be mapped out in what is called the “information architecture,” which identifies the sitemap, the different categories for the site, how many templates will be needed, and what the content, elements, and calls to action will be on each page of the experience. To do this work, a designer will often work hand in hand with a technologist who understands the ecommerce platform. The designer will be responsible for determining the visual elements of the user experience, the font and typography, and the logos and branding, and for specifying the high-level layouts of pages, like the product page and landing pages. From this design work “wireframes” will be created that bring together the design decisions and create a two-dimensional view of what the experience should look like. These wireframes act as the blueprint for where the elements of the user experience, such as the images, content, links, and interactions, will go. From the wireframes interactive mock-ups are created that are “pixel-perfect,” meaning they show the look and feel of the site as it should appear in a browser. The interactivity provided is generally based on hyperlinks that show how the experience flows when the links are clicked. The final mock-up will include each specific design element, define the actions that occur when interacted with by customers, and provide a set of design assets (jpegs and gifs and so on) that can be used by development to code the site. The final mock-up is usually a file, like a Photoshop file, that is then coded by a development team using the technology of their choice (for example, Java and CSS).
• Transactional components. The product catalog, content management system (CMS), shopping cart, payment processing, and order fulfillment systems are required for an ecommerce transaction to occur. The transactional components of an ecommerce platform include order management, payment processing, shipping and warehouse management, pricing and inventory management, and even fraud and risk management. All these transactional components can produce data for analysis. For example, the shopping cart experience can be tested for conversion optimization. The order process from warehouse to customer can be tracked and measured. You can analyze customer interactions with the product catalog. The impact of shipping and warehouse management on order fulfillment is another useful area for analysis.
• Customer components. In an ecommerce business, the customer relationship management (CRM) system often contains the master customer record—also known as the “golden” customer record—which contains personally identified information about the customer (such as the customer’s name, address, preferences, products purchased, and so on). For more information about the customer record, please see Chapter 9, “Analyzing Ecommerce Customers.” CRM systems may integrate with or stand alone as campaign management systems enabling the company to send communications, such as e-mails, that contain specific content relevant to the customer profile. CRM systems may provide an analytical layer and the capabilities to pass information to other systems for more complex integrations. For example, a CRM system may create a unique customer ID and pass it to a digital analytics system. CRM capabilities are also important for enabling customer-centric marketing automation.
• Internal and external data components. An ecommerce experience is dependent on being able to store data. The data could be about customers and their behavior, products, suppliers, transactions, and other events specific to the business. Traditionally, these data would be captured in a CRM database or back-end database with a custom data model. For this discussion, the data component is all the data and systems that house it. The data component will contain customer data, product data (SKU, description, picture, price, and so on), order data, promotional data (name, type, discount, code), inventory data, and so on. The platform can be configured to support data business logic applied to data. For example, when a product catalog for a specific brand is displayed, related products are shown. The product images, descriptions, text, and so on are stored in the database, and the logic for rendering the related products is also coded as business rules. For more information about the data model in ecommerce, see Chapter 5, “Ecommerce Analytics Data Model and Technology.”
I have simplified the components in an ecommerce platform to fall under marketing data, user experience and interface data, transactional data, customer data, and internal/external data in order to help you understand what to analyze. As an analyst working on an ecommerce site, you will be asked questions about the flow of the site and its impact on conversion. You will be asked about orders and products. You will be asked questions about customers and their data—and you will have to work with ecommerce platform data or related data, and perhaps the ecommerce platform itself, to do so.
The ecommerce platform supports the operation of business functions that an analyst will be asked to help with via research and analysis. However, the businesspeople you interact with will have varying knowledge of the ecommerce platform. Some will work directly and daily with it, whereas others will work only infrequently and indirectly with it. What’s certain, from an analyst’s perspective, is that these people will all want data to inform their business activities and will have varying abilities in their capacity to understand and analyze the data you present. It’s also unlikely that the businesspeople will have full knowledge of an ecommerce platform. They may not even understand that term. Your role is to understand how they use the platform (however abstractly), determine what data they generate and/or want to analyze, and help them answer business questions that guide them in the execution of their jobs and the fulfillment of their business goals.
It is common for an ecommerce platform to have tools that businesspeople can use to participate in business operations and/or the deployment of the site. The tools will vary by platform, but the roles working directly and indirectly with the platform are fairly consistent from ecommerce company to ecommerce company. The business tools supported by the ecommerce platform can include the following:
• Product management tools. Crossing both the business and the IT sides of the business, product managers define the look and feel, features and flows, and overall consumer experience for ecommerce. Depending on the size of your business, the product manager may own the entire web site, or a number of product managers at various levels of seniority may be responsible for various functions or applications within it. For example, a product manager might own the entire shopping cart experience and another may own the product catalog and product detail pages. Yet another product manager may own search.
The ecommerce platform will be well understood by most, if not all, product managers because they defined the functionality. Product managers will likely have some knowledge of how to directly use it to create new pages or modify an experience. However, the product manager will expect IT to build and create the user experience to the specifications and product requirements they create. Thus, although product managers won’t directly operate on the platform to build functionality, they will likely use any prebuilt reporting that comes from the platform. That usage may be less than ideal when the data coming out of the platform differs, in some way, from the curated data coming from analytics. Make sure you address this potential point of confusion by effective data governance and meet the data and analytical requirements from product managers—or else they will get it themselves and go around the analytics team.
• Merchandising planning tools. The “merchandising” team is responsible for promoting products so they get sold. The brands, products, promotions, and other aspects of the ecommerce experience that impact products are in the domain of merchandising. This team may use the ecommerce platform to enter orders and check the status of inventory on-order, in-site/store, in-flight, and so on. They will want data about inventory turnover, sell-through, inventory aging, stock outs, and more. The merchandising team will expect you as an analyst to help them find, source, analyze, report, and visualize data that supports their deep understanding of the products sold, including buying behaviors, current trends, seasonality, and promotional impact.
• Inventory and warehouse management tools. This role may not exist at your company because the work may have been contracted out to a third party, and that third party provides data and reporting related to supply and inventory levels back to your company. For those companies that operate or manage their own inventory and warehouse operations, the analytics teams may assess the demand side of the business using data mining and predictive analytics to provide detailed insights that can help frame and forecast demand for the supply side. For example, the inventory and warehouse managers will want to understand peak demand times, the biggest supply risks, and the duration from order receipt to warehouse fulfillment. Other managers may want to understand the impact of seasonality and variability in demand on their resourcing and space allocation.
• Promotional tools. These are most frequently used by merchandisers and marketers to create, deploy, and change the promotional codes, coupons, and the discounts and offers available. This functionality is typically baked into almost all ecommerce tools. People developing and deploying promotions will expect the analytics team to provide self-service tools and reporting about the usage, sales, profitability, and margins resulting from orders that use a promotional code. In addition, the marketing team will want to dive deeply into the data about customers who take advantage of promotions to understand that customer segment. Marketing and merchandising will want data for determining the best products to discount, for adjusting the frequency and content of offers, and for identifying the right customer segments to target.
• Content management tools. The content management system is important to an ecommerce platform. Although a content management system may be a core part of the platform or integrated with it, the analytics team will want to have a strategy that enables tracking of the clickstream, events, and behavior of people within specific content. Data collection should be automatically tagged in your content management system by native features or via tag management.
• Marketing tools. Capabilities for marketing are integrated to work with an ecommerce platform. Marketing tools run the gamut from e-mail service providers to marketing automation vendors to digital analytics to campaign management to attribution. Marketing tools in an ecommerce platform can enable testing and optimization of key ecommerce flows and functionality. Marketers will want the analytics teams to help them find, transform, and use the data marketing tools integrated with the platform.
• Search tools. Tools that allow for the natural language or operator-based searching and querying of the ecommerce platform can be enabled and tuned within the ecommerce platform. The queries entered into the search experience may be tracked and stored by the ecommerce platform, and the rules that map certain pages, brands, or categories to specific keywords may be established within the ecommerce platform. Naturally, business stakeholders will seek to use analytics to help them understand how customers are expressing themselves about the product mix and what queries are most effective in helping people find and buy specific products.
• Security and fraud awareness tools. Although it is common for a large ecommerce sites to have dedicated resources for site security and fraud detection, the ecommerce platform may provide functionality that enables security (like SSL) and have basic business rules baked into it that automatically detect potential fraud and suspicious orders and provide alerts about it. For example, a large amount of orders from specific IP addresses or the creation of new accounts that subsequently have a high rate of credit card declines are types of suspicious behaviors and transactions that could throw a flag or generate an alert to be investigated. The frequency of these alerts and other data and details about the security of the site or fraud on it may be requested by stakeholders from the analytics team.
• Mobile tools. These are tools for creating and rendering mobile ecommerce experiences on mobile devices, smartphones, tablets, and even apps. Mobile browsing, as opposed to shopping and buying, can be a primary use-case for an ecommerce mobile site or app. As such, people such as product managers and merchandisers will seek to understand data from the analytics team about the types of devices, what functionality is used on mobile, and the difference in the behavior of mobile browsers versus mobile shoppers. Mobile can also be a way to drive online users to physical store locations or enable collaboration with customer services (via mobile chat).
• Social tools. Tools for embedding social media functions into ecommerce experiences are provided to business users in ecommerce platforms. Simple integrations via a code snippet may be supported. Other ecommerce platforms provide deeper integration with common social networks, such as Pinterest, Twitter, Facebook, Instagram, and Snapchat. Although inbound traffic, behavior, and conversion resulting from tagged social media campaigns will be primarily tracked in your digital analytics tool, the ecommerce platform may also track some of this data and potentially merge it with known customer data from CRM.
Given the different levels of knowledge that businesspeople will have about the ecommerce platform—from mostly none to a deep level of understanding—it is necessary for an analyst to have an approach for analyzing the usage of ecommerce platform, the core functionality embedded within it, and other integrated technologies and tools. Presented next is an approach the analyst can take for figuring out what to do next. You will note that this process is aligned with the processes described in Chapter 2, “The Ecommerce Analytics Value Chain”:
1. Start with the business questions.
2. Identify the type of analytical output required from the platform and the timelines for it.
3. Determine the user experience and functionality created by the platform that needs to be tracked.
4. Ensure that the data model exists or the tools deployed are sufficient to support the analytical need.
5. Instrument the platform and its functions, as necessary, to collect the data.
6. Create a data model, implement it, store the data, and govern it to ensure accuracy.
7. Apply analytical and data science methods to analyze the data, and then socialize the analysis.
The ecommerce platform is central to the functionality of an ecommerce site. It contains data for analysis and will integrate with other systems that contain data. While it is an abstract technology that will “look” very different at different companies, it is important for ecommerce analysts to understand and work with it.