Using Feature Architecture to Enable Solution Selling

Vasco Drecun, Research Director

Executive Summary
D.H. Brown Associates, Inc. (DHBA) was recently engaged by a leading medical devices manufacturer to review the company's practices of commercial product definition and assess the potential of using a feature architecture to enable solution selling. The established approach in the company relied on catalog identifiers for describing products commercially, as an extension to the manufacturing representation. Manufacturing effectively extended and imposed its view as a loosely implemented, quasi-standard on marketing and sales, and ultimately, the customer. The result was a manual and error prone reconciliation across both the sales and manufacturing environments.

A feature architecture provides flexibility without proliferating the number of identifiers. The feature architecture builds on a hierarchical, constraint-based structure that presents an implicit logic. That logic, which employs relatively few rules, is not only much easier to understand and remember, it is also far easier to implement.

Catalog identifiers, in contrast, supply a highly complex classification scheme that must be delineated in full detail, and cannot possibly be memorized given the accumulation of detail. To maintain the integrity of different configurations, the catalog-based paradigm requires another layer of complexity that defines an explicit set of rules that is heuristic in nature and does not constitute an obvious functional description for the product. Whereas features fulfill the same requirements with a single, logical set of rules, a second layer of rules must be created with identifiers to reasonably structure the first layer, which consists of a compilation of details. Moreover, the catalog rules need additional graphical documentation with usage guidelines and online configuration tools where rules are typically organized in a tool-proprietary format. In this manner, the rule-based configuration engines can interpret them. These tools often do not provide a reconcilable release control, and eventually when replaced, do not provide an easy extraction of the product knowledge contained in them.

By comparison, it is far more effective to jump directly to the set of rules enabled by a feature architecture that stores them and relates them to the functional product breakdown, while avoiding a superfluous level of complexity and accumulated detail with identifiers. A structured and standardized commercial product representation offers benefits. The approach is based on a feature architecture reconciled with standardized manufacturing product representations. Benefits include:

• Increased sales responsiveness and productivity. This is derived from two key elements: greater flexibility in defining product functionality at different levels of granularity when assessing competitors; and the ability to prepare quotations for complex solutions faster and more accurately with a customer-specific level of abstraction.
  
• Superior product profitability due to the enhanced flexibility and ease of proactive margin analysis. More precise and flexible modeling of value propositions that can be benchmarked throughout the product lifecycle provide a broader range of options and configurations that may be customized for premium prices.
  
• Improved responsiveness to competition due to improvements in the management of customer expectations and options early in the ordering process. The ability to model competitive products without requiring commercial identifiers means that head-to-head competitive comparisons may be based upon a functional logic already defined. Catalog numbers based upon identifiers provide no logical structure to begin an analysis, whether that effort relates to competitive analysis, product planning, or any other category.
  
• Better product planning based on the rich metrics enabled by hierarchical functional structures.
  
• Reduced errors in order fulfillment due to the synchronization of the commercial and manufacturing release processes. This is accomplished by managing all product structures and rules through a standard based upon a non-proprietary schema.
  
• A smaller workload in product release and engineering change management due to an accurate relational model between the feature architecture and the physical architecture of the product.

In general, a high-level, shared representation in the feature architecture, as supported by the Twelve-Fold WayTM described in many DHBA publications, provides a dramatic simplification of the problem. The mapping of features, functions, and attributes requires one level of logical reconciliation between the domains, replacing the multiple manual levels of reconciliation of detailed lists. With that single reconciliation, the architecture defines the rules and maintains the logical structure to automate the fulfillment of other requirements currently met manually by means of the catalog identifiers.