Configure research measures and read their results

Start with the right questions

The Kano pair and price-sensitivity question types are each a whole method in one question: add one, and its analysis appears in that question's results with nothing to map. The mappings below are for surveys that ask the same things as separate questions, such as the Kano and Van Westendorp templates.

Kano needs paired questions about a feature being present and absent. Van Westendorp needs four numeric price thresholds in the same currency and billing period. Add those questions before configuring their mappings under Settings → Fielding.

Use the standard batteries from the question library for SUS, UMUX-Lite, TIPI, SWLS, WHO-5 and CES. Changing a validated battery’s wording, order or scale can change what its score means. Unmodified library batteries receive their standard scores.

Map a Kano feature

Map each feature’s present and absent questions to their matching five-answer scales. Keep the response labels consistent: Like, Expect, Neutral, Tolerate and Dislike. An ordinary satisfaction question is not a substitute for the pair.

Analysis classifies complete pairs as attractive, one-dimensional, must-be, indifferent, reverse or questionable. Satisfaction is (A + O) / (A + O + M + I); dissatisfaction is −(M + O) / (A + O + M + I). Reverse and questionable pairs are shown but excluded from those coefficients. Read the distribution as well as the largest category, especially when counts are close.

Map the four pricing questions

Map too cheap, bargain, expensive and too expensive in that order. Plumeform uses complete, nonnegative rows with strictly increasing thresholds; equal, reversed or incomplete thresholds are excluded. The result reports valid and excluded counts.

The acceptable range uses crossings of the too-cheap and inverse-bargain curves, and the too-expensive and inverse-expensive curves. Indifference and equal price resistance are different crossings. Their names describe this method; they do not establish a revenue-maximizing price. If a crossing is missing, collect or inspect the data instead of substituting a price.

Verify the analysis population

These measures use the current published version, active filters and unweighted answers. The complete count can be smaller than the dashboard response count because a method needs all its mapped items. Review exclusions before quoting a score.

Pilot the exact published questionnaire and inspect a few known answers. For SUS, ten neutral ratings of 3 score 50; alternating 5 and 1 in standard item order score 100. SUS scores run from 0 to 100 but are neither percentages nor percentiles. Use the local calculators to check exported numeric rows without uploading them.