Rank a short list and score items on a slider in your interviews

Some research questions ask for an order, not a score: which of five benefits matters most, or which step of a checkout causes the most friction. Others need a finer measure than a 5-point scale, such as a 0 to 100 rating or a price on a range. Both are now built into the AI-moderated interview, so each answer comes with the participant explaining it.

How ranking works

Ranking is its own question type. Add 2 to 8 text items, and add labels for the top and bottom of the list if you want them, such as "Most important" and "Least important". Each participant sees the items in a shuffled order and taps or drags them into a ranking of the full list. Then the AI research assistant asks why they put their top item first and their bottom item last.

How the slider works

The slider is a new scale in Carousel questions, alongside the point scales. You set any minimum and maximum, with up to 4 decimal places, and a step that divides the range evenly. Each end gets a label. A new slider starts at 0 to 10 in steps of 1. As with any Carousel question, there's no limit on the number of items.

Read the results with their caveats

Ranking results show where each item was ranked, its mean rank with a 95% interval, and how often it was ranked first. A warning appears when fewer than 30 people completed the ranking, or when participants seem to have ranked items in the order they appeared on screen. Slider results show the average for each item with a margin of error. Both question types feed into Question coding, reports, exports and Ask Conveo.

Research teams asked for it in four situations

These come from conversations with teams in consumer goods, food, personal and home care, insurance and media, and with research agencies. We kept them anonymous.

• Rank concepts once people have seen them. Show the concept boards, ask for an order, then hear why the winner came first.

• Find the benefits and claims that matter. Learn which product benefits or claims weigh most and least in a real purchase decision.

• Add the reasons to a feature study. Rank a shortlist of features next to a quantitative study and hear what the order means to people.

• Score on a finer scale or a price range. Rate claims from 0 to 10, or ask for a price along a set range so answers stay realistic.

What this means for your research

Both are available to every organization on every plan. Ranking is in beta: it's free during the beta and will get its own pricing later, with notice. Some limits:

• Ranking takes up to 8 items and always ranks the full list. For longer lists, use MaxDiff.

• The slider has no NPS or top-2-box. Use a point scale if you need those.

• The slider isn't available in screener questions or for questions under a random image group.

• Neither ranking nor slider answers can be piped into later questions. Point-scale answers can.

Set them up with the ranking questions guide and the slider scale guide.

Built by Irene Gonzalez, with Florian Hendrickx on ranking.