Long online questionnaires
Concept tests, diaries, satisfaction, diagnosis, maturity, B2B. Studies that need dozens of questions to answer the brief, with respondents who answer alone, on their own time.
Immersive research
The long questionnaire, turned into an immersive format designed for each study.
Survey Studio is immersive research. Every study gets its own format, designed by the Cassi.ai team around the brief: a game with missions and badges, an environment the respondent walks through, a narrative in scenes, cards, boards. The questionnaire keeps its rigor and the form disappears. The researcher follows fieldwork live and receives one results board and one report, whatever the format.
Illustrative data. The scenes play on their own.
01 The work
Halfway through the questionnaire, the respondent is still clicking and has stopped reading.
Every long online study reaches this point. The questionnaire is sound, the sample is right, and the answers of the second half are worth less than those of the first.
The same grid of radio buttons for every study
Whatever the brief, the collection tool presents the questionnaire the same way: screen after screen of the same grid, a percentage bar at the top. A packaging test looks exactly like an employee survey.
Speeding, straight-lining, drop-off halfway
The respondent rushes, marks the same column all the way down the grid, or leaves in the middle. The researcher pays in data quality, in cleaning, and in extra fieldwork days to replace the completes that were lost.
A format designed for the study, in place of the form
The Cassi.ai team designs the immersive format around the brief: missions and badges for a brand diary, a store the respondent walks through for a packaging test, a deck of cards for a diagnosis by area. Underneath, one engine keeps the rigor: five question types and conditional logic validated before field.
The questionnaire stays as rigorous as the researcher wrote it. The format the respondent lives in is designed for that study.
02 What it answers
The questions a researcher asks before and during a long fieldwork.
From the respondent's side, a format designed for the study. From the researcher's side, one engine with rules, one cockpit and one report that can be traced back to the data.
"Will they reach the last block?"
A format designed per studygamification, environment, narrative, boards
No two studies get the same experience. The team designs the format that fits the brief and the respondent: missions with levels and badges, an environment with hotspots that open questions, a narrative in scenes, a deck of cards, a board. In every format, progress is visible and celebrated at each block.
"Does my questionnaire fit?"
Five question typesplus conditional logic
Whatever the format, the engine renders five question types: single choice, multiple choice, short text, long text and rating scale. Each format dresses them its own way, a scale as a ruler, an open answer as a sticky note or a speech bubble. Conditional logic is validated before the study goes to field, so no one lands on a question they should have skipped.
"Can I field it in more than one country?"
Three native languagesPortuguese, English, Spanish
Every text of the experience is written natively in the three languages, from the mission briefing to the error message. Each respondent answers in their own language and the data lands in one structure.
"Who still has not answered?"
Fieldwork cockpitlive, per respondent
The cockpit shows who is pending, who has started and who is done, and how far each block has been completed. The follow-up goes to the right people while fieldwork is still open.
"Who can open this link?"
The link is the credentialexpires in 30 days
Each collection link carries a long random token. Only its hash, a one-way fingerprint, is stored, and the link expires after 30 days. Public routes have a rate limit, a cap on requests per address, and each organization's data lives in isolation.
"What do I take to the client?"
One results board, one reportthe same for every format
After publication, whatever format the respondent lived in: a decision matrix, a heat map by area, a radar and an action roadmap in three horizons. The reading is written by a deterministic analyst, fixed rules applied to the data, so the same data always produces the same reading. The PDF has a cover, a table of contents and eight chapters.
03 Method
Six steps from the questionnaire to the published report.
The questionnaire is a definition that the engine renders. The immersive format is designed on top of it, by the Cassi.ai team, for each project.
- 01
Bring the questionnaire
You send the questionnaire and the study design. The Cassi.ai team turns them into a definition with blocks, questions, scales and conditions.
The rigor of the questionnaire is preserved.
- 02
Design the format
The team proposes the immersive format that fits the brief and the respondent: gamified missions, an environment to walk through, a narrative in scenes, cards, a board. Each block of the questionnaire becomes a piece of that format.
One engine underneath, whatever the format.
- 03
Validate the logic
Every condition and every skip is checked against the definition before a single link is issued.
No study goes to field with a broken skip.
- 04
Issue the links
Each organization gets its own space and its own collection links, in the languages the study needs.
Long random token, stored as a hash, valid for 30 days.
- 05
The respondent enters the format
A mission map, a store aisle, a deck of cards: the respondent moves through the format and each finished block is celebrated. Anyone who leaves comes back to the same place.
Progress is celebrated. A score is never shown.
- 06
Follow and publish
The cockpit lists who is pending, who has started and who is done, with completion by block. After publication, the results board and the report come out of the same version, the same for every format.
A published version never changes. A correction becomes a new version.
04 What is behind it
Rules of the platform, stated with their scope.
Each number below describes how the product works today. None of them is a promise about your completion rate, which depends on the sample and on the questionnaire.
5question types in the engine, with conditional logic validated before field
3native languages in every text: Portuguese, English and Spanish
30days until a collection link expires, with no reuse afterwards
8chapters in the PDF report, after the cover and the table of contents
What it never does
- Never ships a standard questionnaire. Every study gets a format designed for it.
- Never shows a score to the respondent. Progress is celebrated, results are for the researcher.
- Never mixes data between organizations. Each one works in isolation.
- Never changes a published result. A correction is published as a new version.
- Never lets a model write the reading without rules behind it. The reading is rule-based and reproducible.
Who it is for
- Research firms
- Insights teams
- HR and transformation consultancies
- Brand teams
Where it fits: concept and packaging tests, brand and product diaries, customer satisfaction and experience, employee and organizational diagnosis, maturity assessments, and B2B surveys with long questionnaires. Each study gets its own format, configured per project.
05 Questions
Questions and answers
What is Survey Studio?
Survey Studio is a Cassi.ai product for immersive research. Each study receives an immersive format designed for it by the Cassi.ai team, in place of a standard questionnaire: gamified missions, an environment the respondent walks through, a narrative in scenes, cards or boards. Underneath runs one engine with five question types and three native languages. The researcher gets a live fieldwork cockpit, a results board and a report in Markdown and PDF.
How is it different from a standard survey tool?
A standard collection tool presents every questionnaire the same way: a sequence of grids with a percentage bar. Survey Studio presents each questionnaire in a format designed for that study. A brand diary can be a map of missions with badges, a packaging test can be a store aisle with hotspots, a diagnosis by area can be a deck of cards, one card per area. The rigor of the questionnaire stays the same, the experience of answering it changes with the study.
Which immersive formats are available?
The format is designed per project, so the list grows with the studies. Formats the team designs include gamification with missions, levels and badges, environment immersion with a scene and hotspots that open blocks of questions, narrative in scenes, decks of cards, and one example already in field: a cork board where each block is a pinned note and a field notebook where each question is a page. Each block of the questionnaire becomes a piece of the chosen format, and progress is always visible.
Which kinds of study does it fit?
It fits studies where the questionnaire is long and drop-off hurts: concept and packaging tests, brand and product diaries, customer satisfaction and experience, employee and organizational diagnosis, maturity assessments and B2B surveys. Two studies already run on it, an operational diagnosis by area and an AI maturity assessment. Every other category is designed per project.
Can I build the questionnaire and the format myself?
Not today. Survey Studio has no self-service builder. The questionnaire is a definition that the engine renders, and the Cassi.ai team sets it up for each project from the questionnaire and the study design you send. The immersive format is designed on top of that definition, also by the team.
Which question types, logic and languages does it support?
The engine has five question types: single choice, multiple choice, short text, long text and rating scale. Conditional logic, the rules that show or skip a question according to earlier answers, is validated against the definition before the study goes to field. Every text of the experience is written natively in Portuguese, English and Spanish, and all answers land in the same data structure.
How are the links and the data protected?
The collection link is the credential. It carries a long random token, only the hash of that token is stored, and the link expires after 30 days. Public routes have a rate limit, which caps the requests accepted from one address. Data is isolated per organization and never mixed.
Who writes the reading of the results?
A deterministic analyst: a set of fixed rules applied to the data, so the same data always produces the same reading and anyone can trace a sentence back to the numbers. The results board is the same for every format: a decision matrix, a heat map by area, a radar and a roadmap in three horizons. Published results are immutable per version.
Bring the questionnaire that loses respondents halfway.
Send the questionnaire and the study design. We propose the immersive format that fits your study and show what the cockpit and the report look like for your category.