Free conjoint analysis software for the data you already have, or the study you haven't run yet. Build balanced choice tasks here, field them in Qualtrics, SurveyMonkey or any survey tool, then upload the answers and get part-worths, importances, willingness to pay, a market simulator and 95% ranges on every number. Choice-based conjoint, MaxDiff, menu-based choice and TURF. Free, with no time limit.
Every module, the simulator, the optimizer, the tutorials and the course. The Free plan covers class projects, pilots and small studies; it doesn't expire.
| study type | free limits | you get |
|---|---|---|
| Choice-based conjoint (CBC) | 300 respondents, 12 tasks each, 5 products per task, 25 estimated values | design generator, part-worths, importances, willingness to pay, taste variation, how people switch, segments, simulator, optimizer, holdout check |
| Menu-based choice (MBC) | 300 respondents, 12 menus, 12 items | what people buy together, price effects across items, bundles, reach, simulator |
| MaxDiff (best-worst scaling) | 300 respondents, 20 sets of up to 6, 30 items | preference scores, top-3 shares, shortlist simulator, TURF, design generator |
| TURF | from a menu or MaxDiff study | the shortlist that reaches the most people, marked proven best where it is |
Bigger study? The free trial, academic (.edu) and beta tester routes run larger studies free too. See the four ways to start free.
Any tool that records which option each respondent chose. Chosŏn can generate a balanced design for you, for CBC and for MaxDiff, to load into Qualtrics, SurveyMonkey or any survey tool.
Designed in Chosŏn? Upload your survey tool's export as it is (one row per respondent and a column per task, the way Qualtrics and SurveyMonkey export it) with the design file; if the columns aren't named the usual way, you pick them. Otherwise one row per option shown. CBC: resp, task, alt, <attributes>, chosen, or a Sawtooth .CHO export. MaxDiff: resp, set, item, best, worst. Menus: resp, menu, item, price, bought. The upload check explains anything that won't fit, in plain language.
A small study fits in a couple of minutes. Every number comes with a 95% range, and the model is checked on each respondent's last task, which it didn't see.
Change a product, a price or the whole line-up and see shares move. Drop a product and see where its buyers go.
Most conjoint software assumes a product's buyers spread across the shelf in proportion to share when it changes. Chosŏn learns from the choices which products people treat as close substitutes, so line extensions, flavors and pack sizes split share the way shoppers do. It's on by default, free.
Where a new variant's volume really comes from. Read more →
Where a dropped product's buyers go. Read more →
A ranging test where the right answer is known. Read it →
Yes. Studies within the Free limits are free with no time limit and no card. Larger studies run on the free trial, academic and beta tester routes; paid plans are for people who run large studies often.
No. Run it in any survey tool, then upload the responses. Chosŏn doesn't host surveys; it analyzes them.
Yes. Enter your attributes and levels and Chosŏn builds balanced choice tasks, one version per respondent, with prohibited combinations if you need them and a report on the design's balance and efficiency. MaxDiff designs too.
Start with the free course and the tutorial projects. Each one walks through a full study on example data where the right answer is known, so you can see what the model recovered.
Your projects are visible only to your account. To have a project and its data deleted, email info@quantecarlo.com.