Survey analysis

Standard Kano categorisation

Whichever category has the most votes is the primary category for that feature, highlighted in yellow. Multiple categories may have the same count, or a category might win only by a very small margin. We apply Fong's test ^[1] to see if the primary category is a statisically significant winner (90% confidence level), indicated with an asterisk *.

[1] Fong, D. (1996). Using the self-stated importance questionnaire to interpret Kano questionnaire results.The Center for Quality Management Journal, 5, 21 – 24.

Distribution between categories

To improve on the basic Kano categorisation, we show the spread of responses between the different categories as a radar chart, and compute a less strict statistical significance test using a Z-Score ^[2] to a level of 1.5 standard deviations. Categories meeting this threshold have a tick, otherwise they appear with a confidence % level.

[2] Z-Score on Wikipedia.

Continuous matrix analysis

This analysis plots features on a grid to show the nuance within the categorisation, a method proposed by DuMouchel ^[3]. You may get slightly different answers compared to other approaches - the others are 'lossy' (we lose information by reducing responses to a single category), whereas this is more sensitive and allows for the strength of each expression (eg. tolerate vs dislike vs expect) within a category.

[3] William DuMouchel, “Thoughts on Graphical and Continuous Analysis”, Center for Quality of Management Journal, Fall 1993

Customer satisfaction coefficients

This chart condenses all the information about each feature down to a single relative measure of how customer satisfaction could change if the feature were absent or present ^[4]. Red / negative numbers show the dissatisfaction if the feature was absent, green / positive numbers show satisfaction if it was present.

[4] Mike Timko, “An Experiment in Continuous Analysis”, Center for Quality of Management Journal, Fall 1993

Prioritisation

This sequence of features will maximise overall positive impact on customer satisfaction, based on the analysis of primary categories and satisfaction coefficients. Note: where features have multiple primary categories you may see them repeated in the list. You can choose which primary category you wish to go with.