Surveys

The why, then
how many.

Quantitative surveys with screening, conditional logic and recruitment, in many languages, next to the interviews that explain the numbers.

Interviews find the problem. Surveys tell you how big it is. Do both in one study.

An AI survey tool with online survey recruitment built in: design, field and analyse quantitative research in multiple languages, then connect every number to the quotes and clips behind it. That is mixed-methods research without two vendors.

Mixed methods

Twenty interviews.
Two thousand answers.

Interviews surface the issues in people's own words
The survey measures each issue across your market
Every chart links back to the interviews that explain it
One report, one team, one timeline

Why did you abandon checkout?n = 2,000

Unexpected fees at checkout

62%

Delivery time unclear

41%

Could not compare plans

33%

Payment method missing

18%

“I got to the end and the total was ten dollars more. I just closed the tab.”

From interview P08Linked to finding 1

Survey features

Built for real fieldwork.

Screening

Reach the exact audience by market, age, segment or behaviour.

Conditional logic

Branching so every respondent sees the right questions.

Many languages

Field one survey across markets and read the results in one language.

Recruitment included

Respondents come from the same verified sources as interview participants.

Quality checks

Attention and consistency checks, with flagged respondents reviewed.

Crosstabs and charts

Segment comparisons and stakeholder-ready charts in the report.

For AI agents

Surveys at scale,
from your agent.

The Cookiy Skill and CLI turn any question into a full research program inside Claude, Codex or Cursor: participants, interviews, surveys and the report. Open source, MIT licensed.

QualAI interviews with real people, from five to hundreds per study

QuantMulti-language surveys with screening, conditional logic and recruitment, at the scale you need

View on GitHub · ★ 1.5k Agent Skill and CLI →

claude

> /plugin marketplace add cookiy-ai/user-research-skill
> /plugin install user-research@cookiy-ai
✓ installed cookiy user-research

> Interview 20 ICU nurses about shift handover,
  then survey 2,000 nurses in the US and UK to size it.

● Discussion guide + survey drafted .. done
● 20 ICU nurses verified, interviewed  20/20
● Survey live in 2 markets ...... 1,312/2,000
● Themes, tone and crosstabs ...... running
→ One report: the quotes and the numbers

Keep exploring

Related: AI user interviews, Pricing research and Concept testing.

AI user interviewsIn-depth interviews with real people, a hundred a day if you need them.Learn more →

Pricing researchWhat people will pay, and the reasons behind it.Learn more →

Concept testingFind out which idea wins, and why, before you build it.Learn more →

About surveys.

Can Cookiy run quantitative surveys?

Yes. Design the survey with screening and conditional logic, field it to recruited respondents in many languages, and get results with crosstabs next to your interview findings.

How large can a survey be?

As large as the audience allows. Consumer surveys scale to thousands of respondents; niche B2B and expert audiences are sized with you before launch.

What is mixed-methods research?

Combining qualitative interviews (the why) with quantitative surveys (how many). A common pattern: 20 AI interviews to find the issues, then a survey to measure how widespread each one is.

Can I run surveys from my AI agent?

Yes. The Cookiy Skill and CLI let Claude, Codex or Cursor design, field and analyse surveys at scale. See Agent Skill and CLI.

Which question types are supported?

Choice, rating-scale and open-ended questions, with conditional branching. Interesting open ends can be followed up in an AI interview.

How do you keep survey data clean?

Screening before the survey, attention and consistency checks during, and review of flagged respondents after.