How often have you contemplated the following questions?

o What do customers value in your product or service(s)?

o Are there attributes/features you could eliminate so as to reduce costs and
   pass on savings to the consumer?

o What product features will have the biggest impact in generating satisfaction?

o What features are non-negotiable from the point of view of the customer?

o Where in the product should you focus your development resources?

o Are you missing an existing opportunity to “wow” the customer?

o How should you differentiate your product from the competition?

At Perceptive Sciences, we answer these kinds of questions for our clients using a suite of analysis tools that includes the technique of Attribute Mapping. This technique originates in the pioneering work of Japanese quality expert, Noriaki Kano and so it has historically been referred to as a Kano analysis. The heart of the technique is based on the following three insights:

1. Implicit notions about quality can be made explicit through carefully constructed surveys and response analyses.
   
2. All product attributes (features) do not have the same functional relation to customer satisfaction – attributes are better understood as falling into one of four basic classes (one-dimensional, attractive, must-be, and neutral).
   
3. The resulting attribute taxonomy can be used as a principled guide for optimizing resource allocation and development strategy.

Kano Theory

One of Kano’s principle contributions to the science of quality was his theory of differentiated types of quality. Simply put, the theory holds that attributes are not of a single kind. Rather, attributes fall into several broad classes and each of these influences customer satisfaction in a qualitatively distinct way. The traditional folklore was that that all product attributes have a simple proportional effect on quality – the more functional or realized the product’s attributes, the more satisfied the customer will be with the product. In the Kano typology, these kinds of attributes are termed one-dimensional, in recognition of the simple and symmetric relationship between degree-of-functionality and satisfaction. One-dimensional attributes are also called “satisfiers” or “differentiators” throughout the literature.

The paradigmatic example of a one-dimensional attribute taken from the automotive industry is “good gas mileage” – the better the miles per gallon, the higher the customer’s satisfaction, while a lower mileage per gallon leads to a decrease in customer satisfaction. This relationship is schematized by the solid black line in Figure 1 which plots customer satisfaction against the extent to which a one-dimensional attribute is functionally realized within the product.

Though many attributes can be fairly represented as being one-dimensional, Kano’s key insight was that there were other types of attributes that behaved in a qualitatively different manner. There are, for example, attributes where the relationship between attribute realization and satisfaction is neither symmetric nor linear. The dashed blue line in Figure 1 depicts the functional form for a class of attributes that Kano calls attractive (also known as “exciter” or “wow” attributes).

The attractive type is characterized by accelerating levels of satisfaction as the attribute becomes increasingly functional within the product. In contrast to one-dimensional attributes, attractive attributes do not lead to customer dissatisfaction if absent from the product. A good example here is the recent innovation introduced by Google Maps that allows users to re-center a map simply by dragging it with the mouse. Though this feature invariably generates customer delight when first encountered, its absence from competitor products has yet to produce dissatisfaction¹.

The converse of a nonlinear attractive attribute is what Kano has called the “must-be” attribute. The functional relation between realization and customer satisfaction for a hypothetical must-be attribute is depicted by the dashed red line in Figure 1. Note that must-be attributes do not increase satisfaction when they are present in the product, but can decrease satisfaction when they are absent. These kinds of attributes are so named because they must be in the product to avoid perceptions of inferior quality. In the literature, such attributes are also referred to as “non-negotiables” or “dissatisfiers”, because they only have the potential to dissatisfy customers.

A good example of a must-be attribute is the presence of an ATM in a bank lobby. Though having an ATM will not sway the customer’s perception positively (because customers feel all banks should provide this), the absence of an ATM will surely lead to dissatisfaction when customers decide they want or need its functionality during a busy lunch rush.

Finally, there is a class of attributes that generates flat functions in the realization/satisfaction space – the customer is neither satisfied nor dissatisfied by the functional state of these attributes. A Kano analysis categorizes these product features as neutrals. Typically, neutral attributes are targets for elimination as they do not influence perceived quality. However, one person’s neutral can turn out to be another person’s must-be. This stresses the importance of looking for reliable differentiation in the attribute map as a function of different demographics or customer segments prior to action.

The Process of Classifying Attributes

The first step in creating an attribute map for your product is to identify and list its primary set of attributes or features. These attributes can either be identified with a current product, or can be hypothetical or proposed features based on input from target customers, marketing, and/or the development team.

Once the key attributes have been identified, the next step is to create a metric for extracting the implicit relation between an attribute and its potential to effect satisfaction in the mind of the customer. This is a critical aspect of the technique and it requires the use of a carefully designed survey based on Kano question pairs. There many ways to falter here, and an improperly designed survey can easily lead to biases or un-interpretable data. At Perceptive Sciences we draw upon the extensive experience of our Cognitive Psychologists to create state-of-the-art metric tools that are unbiased, reliable, and adapted to meet the specific requirements of each project.

With an appropriately designed Kano-metric in hand, the next step is to collect data from a controlled sample of potential or current customers. The survey results from individual participants are then aggregated and analyzed to derive a classification of each product attribute. Typically the classification results are combined with additional assessments of the stated direct importance of each attribute (a separate survey is used to obtain these measures).

An Actionable Solution

The principle purpose of attribute typing is to provide guidance on how best to modify, streamline, and/or evolve a product or service. The resulting attribute map gives development and management a new and clear understanding of how product features differentially create value in the mind of the customer. Attribute mapping allows the team to more easily prioritize precious resources towards those features most likely to wow, delight, or generate positive perceptions of quality, while withholding additional outlays on features that are either neutral or less likely to promote dissatisfaction if absent.

Based on the information gained from the mapping, one common and rational strategy we recommend is to 1) maximize the functionality of all must-be customer requirements, 2) maintain pace with competitors on any key one-dimensional attributes, and 3) allocate some fraction of resources towards implementing important differentiating attractive attributes.

The mapping results can also provide guidance on how best to simplify or prune an existing product that over the years has accumulated a profusion of “one-off” or “add-on” features. When a product becomes overly bloated with new features its overall usability can be compromised (see our related paper on Feature fatigue). Attribute mapping highlights which features are extraneous from the customer’s value perspective, so that the product can be appropriately streamlined. Typically, this involves removing neutral attributes and sometimes even attractive attributes, especially when these are indicated to be of low stated importance to the customer.

Finally, with a large enough sample size, demographic information may reveal reliable distinctions in the attribute map by market segment. If this occurs, then strategy can be tuned using segment-specific maps that reflect the unique requirements of each class of customer. If a single overarching strategy is desired, we can provide a blended solution in which individual maps are weighted and combined so as to optimally take into account the relative frequency and importance of each customer segment.

Beyond the Attribute Map

In addition to the core method of attribute mapping just described, there are a number of elaborations that adapt the technique to better meet the specific needs of our clients. These include the competitive attribute landscape and the attribute lifecycle analysis.

Competitive Attribute Landscaping

This method combines the output of the initial attribute map with an additional typing of your top competitors’ offerings. The final deliverable includes a comparative analysis that highlights the relative strengths and weaknesses of all the maps, as well as key areas where the potential for differentiation is greatest. Results from a competitive landscape often identify currently untapped “wow” attributes that have the potential to generate new value and an edge in the market. Finally, an attribute landscape can be used to derive a detailed better-worse report for the full set of attributes. This report provides a quantitative assessment of how much customer satisfaction changes depending on the extent to which each of your specific product features is better or worse than that of the competition.

Attribute Lifecycle Analysis

This methodological extension begins with the realization that any attribute map is a dynamic, temporally evolving measure of your product. The attractive feature that today is wowing customers will inevitably be co-opted by the competition. As assimilation occurs, it is common for a once innovative attribute to be increasingly perceived as “standard fare”. Over time this leads to a more or less predictable sequence whereby attractive attributes become one-dimensional, then must-be, and eventually in some cases even neutral in their effects. This de-evolution is driven primarily by the twin forces of market saturation and customer habituation to novelty.

The attribute lifecycle analysis addresses where in this cycle each attribute is and whether it is on the cusp of a re-classification. Simply put, the lifecycle analysis assesses the maturity of all product attributes against the backdrop of current customer perceptions. This involves 1) mapping the current attribute classes for all major competitors, and 2) estimating the likelihood that any given attribute set will undergo a major reorganization in the near future. A meaningful lifecycle analysis is built on the truism that attribute types are first and foremost a representation of the product as it is conceptualized in the mind of the customer, not a veridical account of the state of the market. Only by understanding how your customers “understand” the market at this moment in time can one assess the stability of a current attribute map.

Perceptive Sciences Coproration is a science based market research, user interface, design, and user testing firm, employing experts in the fields of cognitive psychology, information sciences, and human factors studies. Perceptive Sciences serves best-in-class technology based companies and market leaders in a wide range of industries in the U.S. and Europe.

^{1. Attribute types do not in and of themselves have a fixed type, and as we will briefly discuss below, most attributes undergo a temporal evolution in which their functional relationships with satisfaction change depending on customer perceptions of the market.}
 

©2006-2008 Perceptive Sciences Corporation | 1-866-9-USABLE | Privacy Policy