Card sorting and MaxDiff in one tool, with automatic quality checks

From a messy list to a ranked one

Most teams run these two methods in two different tools: a UX tool for the card sort and a survey or conjoint tool for the MaxDiff, with the list retyped in between. Here it is one study feeding the next.

1. Sort

An open, closed or hybrid card sort shows how customers group a long list of needs, features or benefits, and what they call each group. Readable at 60 cards and beyond, built for a phone.

2. Carry

One click turns the finished sort into a MaxDiff study. Each group becomes an item to rank, with Graham’s one-line description carried across, so nothing is retyped.

3. Rank

People pick the most and least important from a few items at a time. Every item gets a 0–100 importance score with a confidence range, and a must-have line shows where to stop.

4. Check

Every response is judged automatically: speeding, tapping the same spot, sorting without reading. Rejected panel responses are replaced at no cost.

What the MaxDiff side gives you

Best–Worst scaling, also called MaxDiff, is the method for ranking a list too long to rate. Rating scales tell you everything matters; forced choices tell you what matters most.

Feature	In SortedResearch
List size	5 to 200 items; 20 to 30 is the sweet spot. Five per screen by default, each item shown three times per person.
Scores	A 0–100 importance score per item from a choice model, anchored by a short rating question so the numbers mean something on their own, with a ± confidence range.
Must-have line	Separates the items that clear the bar from the ones that don’t.
Beyond the ranking	Polarization (items that split the room), a TURF shortlist (which few items reach the most people) and scores by segment.
Custom topics	Rank on importance, or on a question you write: most convincing message, most trustworthy name, easiest feature to understand.
Reports	Graham drafts a client-ready report, downloadable as PDF, Word or a branded PowerPoint, plus Excel and CSV data.

Quality checks on every response

The checks run on every response, whoever sent the participant: people you recruit yourself as well as people bought through our panel.

Study type	Automatic checks
Card sort	✓Nine checks, including sorting faster than the cards can be read, one giant pile, every card on its own, and groups that just mirror the shuffled order.
MaxDiff (Best–Worst)	✓Five checks, including a speeder check that knows familiar items are quicker to answer, and tapping the same position screen after screen.
Pairwise	✓Five checks, including one-sided tapping, with left and right randomised so a pattern stands out.
Findability (tree testing)	✓Eleven checks, including never opening a menu and the same answer on every task.
What a rejected response costs	✓Nothing if it came from our panel: it is rejected, refunded and replaced automatically.

Pairwise, when you also need satisfaction

MaxDiff ranks what matters. A pairwise study measures importance and how well each need is met today, and plots both on an opportunity matrix, so the unmet needs stand out. It can be built from the same card sort.

About pairwise

Tree testing, when the list becomes a menu

If the groups from your card sort become navigation, Findability tests whether people can find things in it, with first clicks, paths and a confidence range for every task.

About Findability

When a different tool is the better fit

We would rather you picked the right tool. Look elsewhere if:

You need conjoint analysis, pricing research or a market simulator. Specialist tools such as Sawtooth Software and Conjointly are built for that.
You need prototype, first-click or usability testing alongside your sorts. See our comparison of UX research tools.

If the job is turning a long list of concepts, needs or benefits into a structure and a priority, with every response checked, that is what SortedResearch is built for.

Common questions

Is there one tool that does both card sorting and MaxDiff?

Yes. SortedResearch runs card sorting, tree testing, pairwise comparison and MaxDiff (Best–Worst scaling) in one platform, with automatic quality checks on every response. A finished card sort becomes a MaxDiff study in one click.

Is Best–Worst scaling the same as MaxDiff?

Yes. Best–Worst scaling and MaxDiff are two names for the same method: people see a few items at a time and pick the most and the least important, and those choices add up to a ranking of the whole list.

How does a card sort turn into a MaxDiff study?

Choose Best–Worst, then “From a card sort”, and pick a finished sort. Its final groups become the items to rank, each with Graham’s one-line description, so the list is in your customers’ own words and nothing is retyped.

What quality checks run on the responses?

Up to eleven automatic checks depending on the study type: nine on a card sort, five on MaxDiff and pairwise, eleven on tree testing. They catch speeding, tapping the same spot, sorting without reading and careless tree tests, on every response, including participants you recruit yourself. Rejected panel responses are replaced at no cost.

How much does it cost?

Plans start at $49 a month with unlimited studies, or $99 a month for a team of up to three. Recruited participants cost from $5 per completed response with no minimum. You can run one study free with up to 10 participants, with no card required.

Sort it, then rank it

Run one study free with up to 10 participants, no card required. Every response is checked, and you only pay for the ones that pass.

Try a study first About Best–Worst