Studio Quick Start

Quick Start

This page takes you from "no account" to a published survey, your own test response in the database, a first table, and the code that reproduces it. Allow about twenty minutes. Every step links to the page that covers it in depth.

 1 Sign up ─▶ 2 New project ─▶ 3 Add a question ─▶ 4 Preview ─▶ 5 Save
      ─▶ 6 Publish to pilot ─▶ 7 Answer it ─▶ 8 Look at Data ─▶ 9 First flow ─▶ 10 Take the code

---

1. Create your account

Go to studio.siamang.org. The Sign in card offers:

• Continue with Google — one click; the account is created on first sign-in.
• Continue with Email → — type your address and press Continue. If the address is new, Studio shows Create your account: your name and a password (at least 8 characters with a lower-case and an upper-case letter, a number and a symbol — the checklist under the field turns green as you go). Press Create account, then click the link in the confirmation email.

On sign-up you get your own organization, named after you, with you as its owner, on a 30-day Pro trial. The topbar shows Pro trial · 30d.

→ Sign Up and Sign In

2. Create a project

You land on the organization's Projects tab. Press New project (or, on an empty list, Start from the example study to explore a finished study first: a questionnaire, 729 sample responses and six analysis flows with their reports — see The example study).

1. Name — e.g. Customer Pulse 2026. The dim line under it, "Its address will be /customer-pulse-2026", shows the project's address; it cannot be changed later. A name in another script is spelled in Latin letters (Опрос удовлетворённости → /opros-udovletvorennosti), and an address the organization already has gets -2, -3, … added.
2. Start from — Blank project ("Start with one placeholder question") or Template. For this tour choose Blank project.
3. Press Create →.

The project opens in the Builder. Its first Save, #1, already exists and is valid.

→ Projects

3. Add a question

In the Builder's Structure tab, the blank project's first page, page1, already holds one placeholder question: a Single choice reading "New question", with the choices Option 1 and Option 2, stored in the variable q1. Make it your first question:

1. Click the question card. The Inspector on the right shows it. In Question, replace "New question" with your wording, e.g. How did you hear about us?
2. In Options, edit the Choices — each has a Code (what is stored) and a Label (what is shown). + Option adds one; pressing Enter in a label adds the next row.
3. Open Variable: the Variable label (as in SPSS) already reads your question text, and follows it while you edit. Give it a shorter label of your own if you like — e.g. Source of awareness — and that one stays. This is your codebook entry.
4. Add a second question: at the bottom of the pages rail on the left, press + Question. The menu has two columns — Types and Presets. Choose Open text and write Anything else you would like to tell us?

Note. A question has two names, shown as id → variable in its card and Inspector header. The answer is stored under the variable name: it is the column in Data and in exports, and the name conditions, piping and quotas read. The Id (Inspector → Advanced) is the question's handle in the Builder — scripts target it, and the Logic map and validation messages show it. Plain types start with the two equal (q1 / q1); presets such as NPS (0–10) do not (q3 / nps_3), and that is fine. The main rule: an Id must not be another question's variable name, or the Save is marked errors. Studio never gives a new question such an Id (nor one of the few other names an Id may not take), and if you type one, the Id field says so at once ("This is the variable … stores its answer under — the engine refuses an id that is another question’s variable."). See Question Id and variable name.

→ Question Types · Codebook and Variables

4. See it as a respondent will

Above the canvas, switch Structure | Preview. The preview is the real survey runtime rendering your current (unsaved) document; switch between Desktop and Mobile, and click a question to jump back to it. Answers given here are not stored.

5. Save

Press Save changes (or Ctrl/Cmd + S). The Save dialog takes an optional message — First two questions — then Save.

A Save is a numbered version of the whole project. The engine validates it and generates questionnaire.py, stored with that version. The topbar badge now reads ● valid #2 (or ● warnings #2 — click it to see why).

→ History and Versions

6. Publish to the pilot environment

Open the Distribute tab. In the Publish panel on the right:

1. Current shows #2 ● valid; Draft shows no unsaved edits.
2. Set Target to pilot.
3. Press Publish to pilot.

A card for pilot appears. It shows ◌ Building with a progress stepper and a build log, then ● Live with the survey link:

https://study.siamang.org/<survey-id>/

Press Copy. That link stays the same every time you republish pilot.

New projects come with two environments: pilot (capped at 50 completed interviews) and main (capped at 1,200). Use pilot for testing, main for fieldwork. Only completed interviews count toward a cap — not screen-outs, not unfinished interviews. On the Free plan the project as a whole also stops at 1,000 completed interviews, pilot and main together. See Response caps.

→ Publishing and Environments

7. Answer your own survey

Open the link in a new tab, answer, and submit. The thank-you screen shows a Response ID — the row number of your answer in the database.

→ What Respondents See

8. Look at the data

Open the Data tab and pick the responses table in the rail. Your row is at the top — the grid shows the newest rows first: one column per variable (named after the variable, not the question's Id), plus survey_id, respondent_id, partial, timestamps and a meta column with fieldwork details.

• Insights — instant frequencies and crosstabs, no setup.
• Export ▾ — CSV, Excel, SPSS, Stata, Parquet or SQLite of the whole table. In an export the meta column becomes ordinary columns (duration_s, started_at, one url_<name> per link parameter, …).

→ Responses and the Data Tab · Data Exports

9. Your first analysis flow

1. Open Flows → New flow. Give it a title (Awareness) and press Open canvas.
2. The canvas starts with a Responses source. Select it and, in the inspector, set Environment to pilot (its default is main).
3. From the palette on the left, drag Frequencies (under Analyze) onto the canvas. Drag from the Responses node's data output to the Frequencies node's data input.
4. Select Frequencies and pick your question's Variable.
5. Press Ctrl/Cmd + Enter (Run to here). The inspector shows the table.
6. Save, then Run. The run appears in Run history with its log.

To turn it into a document, add a Report section and a Save report node — see Reports.

→ Analysis Flows · Node Reference

10. Take the code with you

• Builder → More ▾ → Export Python downloads questionnaire.py for the current Save.
• History → open a Save → More ▾ → Download a bundle → With the responses so far downloads the whole study — generated code, documents, codebook, a Methods draft, a citation file, provenance and your data — as a zip you can run outside Studio.

→ Reproducibility: Code, Bundles and Citation

---

Where to go next

• The whole journey on a realistic study: Tutorial: A Study from Start to Finish
• The vocabulary in one place: Key Concepts
• Routing and screen-outs: Logic and Branching
• Before real fieldwork: Testing Your Survey
• Working with colleagues: Organizations and Team and Working Together

See also

• Siamang Studio — Overview
• Key Concepts
• Recipes
• FAQ and Troubleshooting

---

← Siamang Studio — Overview · Studio contents · Key Concepts →