Skip to content

Latest US poll: 63% say gas prices are affecting their household budget a great deal or a moderate amount

See the poll

Idea screening

Find the few ideas worth a full test

When you have more early ideas than budget, a screen shows which ones buyers care about. Each person reacts to part of the list, and you get a ranking and a shortlist to take into a full concept test.

Talk to salesWatch a test run
5audience groups compared side by side, at most
309ready-made question templates to start from
20question types, each with its own chart

Why screen first

Too many ideas, one budget

A workshop can produce twenty ideas in an afternoon: flavors, features, product lines, offers. Testing each one in full costs too much, and choosing in a meeting tends to favor the ideas the team likes best. A screen is a quick first cut with real buyers. It does not pick the final winner. It drops the ideas that are unlikely to win, so the full test is spent on the right few.

What can go on the bench

A short nameOne or two plain sentencesA picture, if every idea has oneFlavorsFeaturesProduct linesOffersServices

The screen, start to finish

Watch a long list become a shortlist

  1. 01Every idea written the same way
  2. 02People choose, not only rate
  3. 03The ranking builds up
  4. 04The cut

Step 1 / 4

Every idea written the same way

A short name and one or two plain sentences, so the wording and the polish do not pick the winner.

Study set-up

Idea screening

  • StimulusEighteen treat ideas, each a name and two sentences
  • WhoUS dog owners who buy treats
  • DesignEach person sees six ideas in a shuffled order
  • TaskPick the most and the least appealing from small sets
  • ThenShort open answers on the leading ideas

Set up with our team, or yourself in CatalyseAI

Step 2 / 4

People choose, not only rate

From small sets, people pick the most and the least appealing idea. That forces a trade-off and spreads the ideas apart.

Set 2 of 4

Which of these is the most appealing? And the least?

MostLeast

  • Freeze-dried salmon bites
  • Treat advent calendar
  • Bone broth ice pops
  • Calming chew for vet visits
Next

"My dog goes mad for fish, so the salmon bites, easily."

"Ice pops? Not sure what that even is."

"Vet visits are a nightmare. I would buy that today."

Step 3 / 4

The ranking builds up

Ideas are ranked by preference share. You see which ones lead clearly, which are tied and where the shortlist ends.

Preference share, so far

  1. 1Freeze-dried salmon bites11.8%
  2. 2Dental chew with mint10.6%
  3. 3Calming chew for vet visits9.9%
  4. 4Single-protein training treats9.1%
  5. The cut
  6. 5Pumpkin and oat biscuits6.4%
  7. 6Bone broth ice pops6.2%
  8. 7Senior soft bites6.0%
  9. 8Seaweed skin and coat chews5.8%

Step 4 / 4

The cut

The ideas above the cut go into a full concept test. Unclear ideas are flagged for a rewrite, not dropped.

Verdict

Four ideas go forward

Two more get a rewrite before they are judged.

  • Four ideasShortlist
  • Two ideasRewrite
  • Twelve ideasPark

The ranking

Every idea, ranked by preference share

Each row is one idea, in order of preference share. The columns are the measures the report names.

Base: Dog owners who buy treats; each saw six of the eighteen ideas.

Preference share adds up to 100% across the list. Appeal, newness and clarity are the share who chose the top two boxes.

Idea screening · scorecardBrand A · idea screen report
#Idea
Preference share
Appeal
Newness
Clarity
  1. 1Freeze-dried salmon bitesShortlist11.8%71%52%84%
  2. 2Dental chew with mintShortlist10.6%68%41%88%
  3. 3Calming chew for vet visitsShortlist9.9%66%63%79%
  4. 4Single-protein training treatsShortlist9.1%64%48%81%
  5. 5Pumpkin and oat biscuitsTied6.4%55%37%86%
  6. 6Bone broth ice popsUnclear: rewrite6.2%52%66%38%
  7. 7Senior soft bitesTied6.0%54%40%83%
  8. 8Seaweed skin and coat chewsUnclear: rewrite5.8%50%69%41%
  9. 9Low-calorie diet bitesTied5.5%51%35%85%
  10. 10Travel treat pouch4.6%46%44%80%
  11. 11Puppy teething rings4.3%44%39%82%
  12. 12Sweet potato jerky4.1%45%33%87%
  13. 13Breakfast bars for dogs3.4%39%58%64%
  14. 14Treat advent calendar3.0%37%55%77%
  15. 15Insect-protein crunchies2.6%30%74%59%
  16. 16Birthday cake mix for dogs2.4%33%61%72%
  17. 17Cheese-flavored puffs2.3%31%46%75%
  18. 18Treat-filled squeaky toy2.0%29%42%70%
  • Preference share How often an idea is picked as the most appealing, against how often it is picked as the least, across the small sets.
  • Appeal How much people like each idea they see.
  • Newness Whether the idea feels different from what is on offer today.
  • Clarity Whether people understand the idea from a short description. An unclear idea can lose for the wrong reason, so it is flagged rather than dropped.

The shortlist

Example: eighteen ideas for a pet-treat range

A US pet-food company leaves a workshop with eighteen treat ideas, each written as a name and two sentences. Each dog owner in the study sees six of them in a shuffled order and picks the most and least appealing from small sets. The ranking shows four clear leaders, a cluster of ties in the middle and a tail of ideas few people want. Two of the middle ideas score low on clarity, so the team rewrites them instead of dropping them. The four leaders go into a concept test.

  • Four ideas

    Shortlist

    Salmon bites, the mint dental chew, the calming chew and the training treats lead clearly.

    They go into a full concept test.

  • Two ideas

    Rewrite

    Bone broth ice pops and the seaweed chews score low on clarity, not on appeal.

    Reword them and screen them again.

  • Twelve ideas

    Park

    A cluster of ties in the middle and a tail of ideas few people want.

    Keep them on file for later.

Ask the studyAnswers from your own data

Which ideas are tied in the middle of the list?

Ideas five to nine sit within a point of each other. Two of them, the bone broth ice pops and the seaweed chews, score low on clarity, so rewrite them before you drop them.

It checks its numbers against your data and draws the chart.

Step by step

How a screen works

From a long list to a shortlist, step by step.

  1. 01

    Write every idea the same way

    Give each idea a short name and one or two plain sentences. If some ideas have pictures, give them all one. When every idea looks alike, the wording and the polish do not pick the winner.

  2. 02

    Split the list across people

    Nobody reads twenty ideas carefully. Each person sees part of the list in a shuffled order, and every idea is seen by a similar number of people.

  3. 03

    Ask people to choose, not only rate

    Rating scales tend to call everything good. So people also pick the most and the least appealing idea from small sets, which forces a trade-off and spreads the ideas apart.

  4. 04

    Rank the list and draw the cut

    Ideas are ranked by score. You see which ones lead clearly, which are tied, and where the shortlist ends. Short open answers on the leading ideas say why they appeal.

  5. 05

    Test the shortlist in full

    The ideas above the cut go into a concept test, where each one is developed further, shown on its own and judged on a full set of questions.

  6. Then: decide

    Two more get a rewrite before they are judged.

    Plan your study

What you take away

What you get

  • Output 01

    A ranked list

    Every idea with its score, from first to last.

  • Output 02

    A shortlist with a clear cut

    The ideas worth a full test, and the ones to drop or park for later.

  • Output 03

    Gaps and ties

    How far apart the ideas are, so two ideas a point apart are not treated as a win and a loss.

  • Output 04

    Reasons in their own words

    Open answers on the leading ideas, grouped into themes.

  • Output 05

    Results by group

    Compare up to five audience groups, such as heavy and light buyers, with markers that show which differences are real.

  • Output 06

    Downloads

    Charts as images or slides, and the data as CSV or an Excel workbook.

FAQ

Questions, answered

How is a screen different from a concept test?

A screen looks briefly at many rough ideas and ranks them. A concept test looks at one to five developed ideas in depth, with each person seeing one idea and answering a full set of questions. Most teams screen first, then test the shortlist.

How many ideas can go into one screen?

Tell us how many you have. Lists of about ten to thirty suit best-and-worst choices. A longer list can be split across more people.

Do the ideas need to be finished?

No. A name and a short description is enough. Keep every idea at the same level of finish, so the polish does not decide the ranking.

Can I run a screen myself in CatalyseAI?

Yes. Idea screening is a ready-made study type in CatalyseAI, and our team can set it up with you. Send us your list to start.

Related

Goes well with

SolutionConcept testingFind the winning idea before you build itSolutionMaxDiffRank a long list by what matters mostSolutionOpportunity assessmentFind the market gaps worth investing inOverviewAll solutionsEvery study we run, by the decision it supports.

Got a long list of ideas?

Send us your ideas as they are. We will suggest how to word them, how to split them across people and how many people you need.

Talk to sales

Nearest office: Orlando, FL (corporate office)