> For the complete documentation index, see [llms.txt](https://docs.askyazi.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.askyazi.com/build-a-study/choose-a-study-type.md).

# Choose a study type

Compare surveys, AI interviews, recurring and diary studies and multi-activity studies so you can pick the right method before you build.

Every piece of research in Yazi is a study, and every study runs over WhatsApp. Before you build, decide which research method fits your question. This page compares the options and points you to the full guide for each one.

{% hint style="info" %}
**In short:** use a **Survey** when you need numbers, an **AI interview** when you need to understand why, a **Recurring** survey or diary when you need to see change over time, and a **multi-activity study** when you need several of these in sequence for the same people. Pick the method first, then decide whether it is one activity or several.
{% endhint %}

## The methods at a glance

When you click **+ New Research** and choose **Build manually**, Yazi asks for a **STUDY TYPE** (**Single campaign** or **Multi-activity**) and a **RESEARCH METHOD** (**Survey**, **Interview**, **Recurring** or **Group Discussion**). If you choose **Build with AI**, Yazi suggests the method from your brief. See [Create a study](https://docs.askyazi.com/build-a-study/create-a-study.md) for both routes.

<figure><img src="https://358667285-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FzMRQUXiGQlBwWdg6WEmG%2Fuploads%2F3SFlVtCiV00y4cC1H5ng%2Fnew-research-manual.webp?alt=media&amp;token=07372011-ef5a-4e40-a660-25e484e10123" alt="The Create Research screen with Build manually selected, showing Study type and Research method"><figcaption></figcaption></figure>

| Method                                    | Best for                                   | How questions work                                                | Typical sample            | Participant time                      | What you get                                                |
| ----------------------------------------- | ------------------------------------------ | ----------------------------------------------------------------- | ------------------------- | ------------------------------------- | ----------------------------------------------------------- |
| **Survey**                                | Measurement, tracking, benchmarking        | You write every question in advance                               | 250 or more               | 3 to 8 minutes for 20 to 28 questions | Coded, countable data, charts and cross-tabs                |
| **Interview** (AI interview)              | Motivations, attitudes, experiences        | A short survey, then an AI moderator asks follow-ups in real time | 12 to 100                 | 8 to 20 minutes                       | Structured answers plus transcripts, voice notes and themes |
| **Recurring** (recurring survey or diary) | Behaviour over time, in-the-moment logging | The same short set of questions repeats over days or weeks        | 20 to 200                 | 2 to 5 minutes per entry              | Time-stamped entries per participant                        |
| **Multi-activity study**                  | Journeys, phased research, mixed methods   | Two or more activities (any mix of the above) that open in order  | Depends on the activities | Spread over several days              | One study with results per activity                         |
| **Group Discussion**                      | Asynchronous moderated group conversation  | Participants post in a shared room that a researcher moderates    | Small groups              | Over a set period                     | A discussion transcript and media                           |

Sample sizes and times are guidance from studies that ran well, not platform limits. There is no hard cap on participants or questions. The practical limit is drop-off, which rises noticeably past about 28 questions on a survey.

{% hint style="warning" %}
**Group Discussion is early access.** Contact Yazi before planning a study around it. See [Group discussions](https://docs.askyazi.com/study-types/group-discussions.md).
{% endhint %}

## Which one should I choose?

* **"I need percentages I can report."** Survey.
* **"I need to understand why people do something."** Interview. The survey section gives you the numbers, the AI section gives you the depth.
* **"I need to see what people do day to day."** Recurring survey or a diary built as a multi-activity study.
* **"I need a baseline survey, then a week of diary entries, then an interview."** Multi-activity study.
* **"I need a web form rather than a WhatsApp chat"**, for example for a panel. Ask about Live Forms, which are early access. See [Live Forms](https://docs.askyazi.com/study-types/live-forms.md).

## One activity or several

A study with one activity is just "a study". The Create Research screen labels this option **Single campaign**, because some screens still use an older label.

A study with two or more activities is a multi-activity study. Each activity can use a different method and can open right away, on a schedule, or when you push it to a participant. Participants move through the activities at their own pace, and they pick their language once for the whole study.

|             | **Single campaign**     | **Multi-activity**                                        |
| ----------- | ----------------------- | --------------------------------------------------------- |
| Activities  | One                     | Two or more                                               |
| Methods     | One                     | Can mix survey, interview, recurring and group discussion |
| Set-up time | Faster                  | Longer, because you plan the schedule between activities  |
| Best for    | One clear research task | Journeys, diaries, phased or longitudinal work            |

See [Multi-activity studies](https://docs.askyazi.com/study-types/multi-activity-studies.md) for scheduling and progression.

## How questions reach the phone

Separately from the method, each survey or survey section is delivered in one of two ways:

* **In-chat**: one WhatsApp message per question, answered with buttons, lists, text or media.
* **WhatsApp Flow**: a form that opens inside WhatsApp and shows several questions on one screen.

You choose this in **Study configuration > Participant experience > Builder Mode**. It affects which question types you can use, how logic works and what the WhatsApp messages cost. See [In-chat vs WhatsApp Flows](https://docs.askyazi.com/build-a-study/in-chat-vs-whatsapp-flows.md).

## Changing your mind later

You can convert a study from one type to another from the type chip at the top left of the study (for example "Survey" with a dropdown), or by asking Ask Yazi. The questions and any answers stay with the study, and Yazi shows you what the change means before applying it, for example that an AI interview stops asking your written questions one by one.

You cannot convert a study that has already gone out to participants, because anyone partway through would be stranded.

## Limits and things to know

* **One kind of response per project for billing.** A survey and an AI interview count as two kinds of response, so on most plans they run as two projects. See [Plans, billing and usage](https://docs.askyazi.com/start-here/plans-billing-and-usage.md).
* **A study cannot be converted once it has gone out to participants.** Duplicate it instead and convert the copy.
* **Ranking questions are not available** in any method. See [Question types](https://docs.askyazi.com/build-a-study/question-types.md) for the workaround.
* **WhatsApp has its own limits** on options, rows and media that apply to every method. See [WhatsApp limits](https://docs.askyazi.com/build-a-study/whatsapp-limits.md).

## Related

* [Create a study](https://docs.askyazi.com/build-a-study/create-a-study.md)
* [Surveys](https://docs.askyazi.com/study-types/surveys.md)
* [AI-moderated interviews](https://docs.askyazi.com/study-types/ai-interviews.md)
* [Diary and recurring studies](https://docs.askyazi.com/study-types/diary-and-recurring-studies.md)
* [Multi-activity studies](https://docs.askyazi.com/study-types/multi-activity-studies.md)
* [In-chat vs WhatsApp Flows](https://docs.askyazi.com/build-a-study/in-chat-vs-whatsapp-flows.md)
