Each Pulse type asks the panel a different kind of question, so Panel emphasizes a different result view. Use this guide to choose the mode that matches your decision and know what evidence will come back.
Choose a study type
The mode changes both what you enter in Pulse and the headline analysis in Panel. Audience, context, assets, evidence, segments, demographics, and persona interviews remain available across study types.
Choose the mode before entering the study content. The selected mode controls the input fields and the primary result treatment.
A/B compares exactly two variants.
Focus Group collects reactions to one or two prompts.
Survey / Statement tests one statement or situation.
Ranking orders two or more options.
Coin Allocation forces budget tradeoffs.
Freeform Focus Group asks an ordered questionnaire.
Start with the decision you need to make: pick a winner, explore reactions, score one statement, order several options, force tradeoffs, or investigate several open-ended questions.
A/B
Use A/B when exactly two variants compete for the same job: two headlines, offers, concepts, or creative directions. Each respondent evaluates both options and records a preference.
Enter: Option A and Option B.
Primary result: probability for each option, a winner only when the configured margin is decisive, and a favor count including neutral responses.
Then explore: score distributions, preference-split demographics, segments, evidence, and individual reactions.
Winner, modeled probabilities, and favor count all point in the same direction in this example; the shared evidence and audience analysis remain available below.
Winner reports a decision only when the probability gap clears the threshold.
A probability / B probability show the modeled chance each option wins.
Favor count shows parsed preferences, including respondents who chose neither.
Focus Group
Use Focus Group for qualitative reactions to one prompt or a side-by-side pair of prompts. It is designed for exploration, not a forced A/B winner.
Enter: one prompt for a focused discussion, or two prompts to compare reaction patterns.
Primary result: 0–10 reaction distributions across the prompt set, with low, median, high, and mean summaries.
Then explore: evidence, respondent segments, option spotlight, and verbatim persona interviews to understand why reactions differ.
Focus Group results compare the shape and strength of reactions without converting them into a probability winner.
Prompt legend keeps each qualitative stimulus distinct.
Reaction distribution shows where each prompt's scores concentrate.
Summary statistics make strength and spread comparable at a glance.
Survey / Statement
Use Survey / Statement when the panel should react to one claim, proposition, situation, or policy. It is the cleanest single-subject benchmark.
Enter: one statement or situation.
Primary result: one uncluttered score distribution with low, median, high, and mean.
Then explore: evidence rows, segments, demographics, and interviews behind the aggregate score.
A single-subject study removes comparison noise so you can benchmark one idea on its own.
Single distribution shows all respondent scores for the statement.
Histogram, box, and dots switch the statistical view without changing the data.
Low / median / high / mean summarize the full response range.
Ranking
Use Ranking when several options need a clear order rather than one-versus-one probabilities. Respondents score the full set and identify their first choice.
Enter: two or more options.
Primary result: ordered option scores plus the count and weighted share of respondents who ranked each option first.
Then explore: the multi-option distribution, evidence, segments, demographics, and respondent-level rankings.
Ranking distinguishes overall score from first-choice behavior, which can reveal a broad compromise versus a passionate favorite.
Rank identifies the multi-option result treatment.
Ranked #1 counts expose the panel's first-choice concentration.
Option score keeps the full 0–10 evaluation alongside first-choice share.
Coin Allocation
Use Coin Allocation when preference should have a cost. Each respondent commits a fixed dollar budget across options, revealing how much they would spend rather than only naming a favorite.
Enter: two or more options, the price of each option, a spend-dollar budget, and whether the full budget must be spent.
Primary result: dollars spent on each option, total dollars allocated, and average dollar spend per respondent.
Then explore: evidence, segments, demographics, and individual allocation decisions.
Coin Allocation makes tradeoffs tangible by showing the dollars committed to every option; spend share remains supporting context.
Dollar budget confirms spend per respondent and whether full spend was required.
Spend dollars show the amount committed to each option.
Average dollars / units separate per-person spend from purchase volume.
Freeform Focus Group
Use Freeform Focus Group when one study needs several open-ended questions. Questions remain ordered and are automatically split into context-safe batches. There is no fixed question-count field limit, but the live preflight must fit the model context, durable shared-input persistence budget, and a maximum of 32 batches across eight worker waves.
Enter: an ordered questionnaire. Add, remove, or reorder questions, then review the live execution preflight before launch.
Primary result: response coverage, sentiment, strength, persona-weighted themes, and exact supporting quotes for every question.
Study rollup: a grounded synthesis connects themes across the full questionnaire without flattening the per-question evidence.
Freeform results preserve each question's evidence while adding a study-level synthesis across the questionnaire.
Question count confirms the ordered questionnaire returned in this study.
Panel themes across questions previews the cross-question synthesis.
Coverage and ready status show whether the full panel answered each question.
Weighted themes and quotes keep the rollup grounded in respondent evidence.
Need a simple rule? Use A/B for a winner, Focus Group for comparative reactions, Survey / Statement for one benchmark, Ranking for order, Coin Allocation for constrained tradeoffs, and Freeform Focus Group for a multi-question qualitative investigation.