For the complete documentation index, see llms.txt. This page is also available as Markdown.

Quickstart: your first study

Build a three-question survey, try it on your own phone in WhatsApp, share it and see the results, in about 15 minutes.

This page takes you from a new account to a working study on your own phone. It uses a short survey because that is the quickest thing to build. Everything here also applies to AI interviews and diaries.

In short: sign up at app.askyazi.com, click + New Research, add three questions, set the study to Testing and open the test link on your phone. Your answers show up under Responses within seconds. You do not need your own WhatsApp number to do any of this.

Before you start

  • A computer with a modern browser (Chrome, Edge, Safari or Firefox).

  • A phone with WhatsApp installed, so you can answer as a participant.

  • A work email address. If your colleagues already use Yazi, ask them for your organisation's access code first, so you join their organisation instead of creating a new one. See Accounts and sign-in.

Build and test your first study

1

Sign up

Go to app.askyazi.com and choose Sign up. Enter your work email and a password, then type the verification code we email you. Create your organisation (usually your company name). Your account starts on Yazi's shared WhatsApp number, so you can test straight away.

2

Start a new study

On the Dashboard, click + New Research. The Create Research screen asks you to "Tell us about your research" and offers two ways to begin:

  • Build with AI (recommended): upload a brief or an existing questionnaire (PDF, DOC, DOCX, XLSX or TXT, up to 10 MB) or paste the text, then click Continue to analysis. Yazi drafts the questions for you. Always proofread what it builds. Create a study

  • Build manually: for this quickstart, pick this. Under Study type choose Single campaign (one activity), under Research method choose Survey, then click Continue. Nothing is generated for you in manual mode.

Create Research with Build manually selected, showing the study type and research method choices
3

Add three questions

In the Build tab, use the + button to add questions. Try one of each:

  1. A Single Select question, for example "How often do you buy coffee?" with three short options.

  2. A Text question, for example "What do you like most about your favourite coffee shop?"

  3. A Voice Note question, for example "Send us a short voice note describing your ideal coffee."

Click the eye icon on any question to preview exactly what a participant will see in WhatsApp. Keep answer options short: WhatsApp cuts long button labels. See WhatsApp limits.

The Build tab of a survey: questions on the left, Study configuration on the right, and the Build, Share, Responses and Reporting tabs at the top
4

Try it in the Test Simulator

Click Test at the top right of the Build tab. The Test Simulator panel shows the study as a WhatsApp chat on a phone preview, so you can check wording and order without picking up your phone. Type your answers into the preview; Reset starts again. It does not handle voice notes reliably, so test those on your phone next. Test your study

The Test Simulator beside the builder: a phone preview of the study on WhatsApp, with Run Simulation and Reset
5

Set the study to Testing

A new study starts as Draft, and nobody can take a Draft study, including you. Click Move to testing at the top right of the study. Moving to Testing publishes your questions, so the test link works straight away. Publish and go live

A draft study: the status shows Draft and the Move to testing button sits at the top right
6

Take it on your phone

Go to the Share tab and click Share link. Open the Test mode link on your phone (send it to yourself, or use the WhatsApp button beside it). WhatsApp opens with a short message already typed, something like "Send to start #4F9A1". Send it without editing it, and the study begins.

The Share study link window: the starter phrase, the live link and the test link, each with Copy, WhatsApp and open buttons

Answer all three questions, including the voice note. Test answers are tagged as test data, so they never mix with real results.

What a participant sees: a single select question opens as a WhatsApp list
7

See your answers

Back in Yazi, open the Responses tab. Switch on the test-data chip to show test responses. You will see your answers in Table Data, your conversation in Transcripts, and your voice note (with its transcription) in Media Library. Charts are under Reporting > Graph Data.

8

Go live and share

When you are happy, click Move to live at the top right and confirm. Yazi shows an estimate of what Meta will charge for the WhatsApp messages. Then copy the live link from Share > Share link and send it by message or email, or put it on a poster or QR code of your own.

The Go live? confirmation before a study starts taking real responses

If your organisation is still a sandbox (trial) organisation, some of these actions may be locked. Locked controls show a lock icon, and clicking one shows you how to upgrade. See Accounts and sign-in.

What to try next

  • Add logic so people skip questions that do not apply to them: Logic and routing.

  • Add an AI interview after your survey questions: AI interviews.

  • Send the study to your own contact list as a WhatsApp broadcast: Send a broadcast.

  • Ask Ask Yazi (the tab on the right edge of the screen) to add questions or translate the study for you: Ask Yazi.

Limits and things to know

  • Draft studies do not run. If the link does nothing, check the study is in Testing or Live.

  • Use the test link for testing. The test link can be used again and again and keeps your answers out of the real results. The live link is for participants.

  • Send the starter message as it is. If you edit or delete the pre-typed message, WhatsApp cannot tell which study you want.

  • Your account starts on Yazi's shared number. Participants see Yazi's WhatsApp name and picture, not yours. To use your own name and branding you need your own number: WhatsApp number options.

  • Setting up a study is free. Charges apply when a study is fielded with real participants. See Plans, billing and usage.