MaxDiff Analysis: Find True Preferences, Not Just Ratings

WHY IT EXISTS: Standard rating scales (e.g., 1-5 stars) are ambiguous. One person's "4" is another's "3", and users often rate many things as important. MaxDiff was designed to break these ties and force choices, revealing a clear, ranked order of preference instead of a cluster of vague high scores.

THE MENTAL MODEL: Think of it as a tournament for features. Instead of asking "How much do you like each of these 15 things?", you show small groups of 4-5 items and ask "Which is best, and which is worst?". By repeating this with different combinations, the system can deduce the full ranking of all 15 items more efficiently and accurately than by asking for a full ranking directly.

HOW IT WORKS: A respondent is shown a small subset of items from a larger list. They choose the single best and single worst item from that subset. The "best" choice implies it's preferred over all others in the set; the "worst" choice implies all others are preferred over it. For example, if a user sees {A, B, C, D} and picks A as best and D as worst, the system learns five preference pairs: A>B, A>C, A>D, B>D, and C>D. This process is repeated with different subsets, and statistical analysis of all choices generates a relative preference score for every item.

WHEN TO USE IT: Use MaxDiff when you need to rank a long list of items (more than 7-10) and need to understand the relative importance or preference. It's ideal for prioritizing product features, marketing messages, or brand attributes where you need to force trade-offs among many options.

WHEN NOT TO USE IT: Don't use it for very short lists where a simple ranking or paired comparison would suffice. It's also not the right tool if you need to understand the absolute value or willingness to pay for something; it only provides relative importance, not standalone valuation.

ONE CANONICAL EXAMPLE: To prioritize 15 potential new app features, you could run a MaxDiff study. A user might see a screen with "Dark Mode", "Offline Access", "Social Sharing", and "Custom Notifications". They select "Offline Access" as best and "Social Sharing" as worst. This single screen provides several data points about their preferences. After many such questions with different combinations, you get a clear, ranked list of all 15 features based on collective user preference.