There are many ways to prioritize work in a backlog. Kano offers a way to do it empirically for a customer-satisfaction driven approach.
The Kano Model is a framework developed by Professor Noriaki Kano for prioritizing work to drive quality. Though the Kano Model prescribes surveys and statistical analysis, its theories, given sufficient knowledge of a product and its users, can be readily applied.
Especially in the absence of empirical data, if the theories of the Kano Model are not understood, or the practitioner lacks product knowledge or trust with the Product Owner, these must be improved first.
Kano designed the model for use during discovery. In agile methodologies, where elicitation is ongoing, the framework is used continuously.
Attributes, Satisfaction, and Dissatisfaction
Attribute The term “attribute” is used here for its unbiased meaning. Broadly, an attribute is a backlog item or feature, a measurable value that can be delivered.
Satisfaction and Dissatisfaction Kano analyzes attributes by satisfaction and dissatisfaction: the satisfaction a user receives from having a function is compared to the user’s dissatisfaction with its dysfunction, not having the function.
Though much has been written about the range of possibilities and cross-analysis, Professor Kano offers a simplified model with three possible answers to the two considerations, function and dysfunction:
For example, bicycles are expected to have brakes. Having brakes does not make a bike more satisfying than its competitors because bikers expect brakes.
However, not having brakes, or having dysfunctional brakes, causes very high dissatisfaction. In the Kano Model, brakes are:
The goal is to identify and deliver attributes that avoid dissatisfaction and avoid spending resources on those that are exciting but offer only nominally increases to satisfaction.
Minimum – Priority 1
Minimum attributes must be delivered for a product to be acceptable. In the Kano model:
Functional: Neutral (expected)
Satisfaction is not increased. Dissatisfaction is avoided. Sometimes called “must be,” “natural,” or “expected,” Kano warns because of their perceived obviousness minimum attributes are often missed during discovery.
For a bicycle, brakes are a minimum attribute.
Performance – Priority 2
An increase in a performance attribute directly relates to the rise in satisfaction. Conversely, a decrease in performance attributes increases dissatisfaction.
An increase in the number of miles before servicing a bicycle leads to more satisfaction. Maintenance may be expected, but below a performance threshold, dissatisfaction will increase.
Attraction – Priority 3
Because they are exciting, attractive attributes are tempting to give high priority. Though not expected, these attributes can set a product apart. However, they should not be prioritized above minimum and performance attributes.
Dysfunctional: Neutral (not expected)
Kano identifies a further risk: over time, attractive attributes become expected. Expectation causes attraction to fall to performance and, eventually, minimum qualities that cause dissatisfaction if not continuously maintained.
A bicycle that includes high-performance gears in its base-model is attractive if customarily reserved for more expensive bikes. If the next release excludes them, buyers are likely to be disappointed.
Neutral - Priority 4
These attributes neither satisfy nor dissatisfy. These can be difficult to identify without user feedback. A sign is that their value is difficult to measure. Kano suggests that such attributes are likely candidates for not developing.
A product team developing a new bicycle may consider offering it in many colors. Although exciting, if buyers are indifferent, then spending resources on this is wasteful.
Often ignored, reversal attributes are those for which more is dissatisfying. For bicycles, weight is a reversal attribute; lighter is preferred. One way to address reversal attributes is to consider what would happen if they were not delivered and create their solution as a performance attribute.
An informed Product Owner should be able to answer Kano’s two foundational questions: how satisfied will users be if this is delivered, and how dissatisfied will they be if it is not? Having answered these questions, it is a matter of assessing the attribute’s type and prioritizing accordingly.