Cloud Your First Project

Your First Project

A project is one study: its survey code, its responses, and its analysis, all in one place. Creating a project sets up a Git repository and a database for you automatically — you do not wire up anything yourself.

New to writing surveys as code? Skim Core Concepts first; it explains the model you will see in the project's files.

How a project can start

When you create a project you choose how it starts — three options:

• Example — the complete "Digital Life & Wellbeing 2026" study: a full questionnaire (consent, screening, quotas, every question type, skip logic, a custom thank-you page), two analysis scripts, and about 300 sample responses so the Database, Dashboard, and Analysis screens are alive immediately. Pick this for your first project — you can see the whole loop before writing anything.
• Template — a minimal survey plus two starter analysis scripts (cleaning and final tables), with no sample data. A tidy starting point for a real study.
• Empty — a one-question starter survey plus the project scaffold (scripts/, outputs/, reports/), and nothing else.

Create a project

What you'll need: the admin or owner role. On the Free plan you can have up to two projects.

Steps

1. Open Projects.
2. Click New project.
3. Enter a name, then choose how to start — Empty, Template, or Example (Example is recommended for your first project).
4. Click Create.

Result: the project opens. If you reached your plan's project limit, the button is disabled with a prompt to upgrade — see Plans & Billing.

What gets created for you

Behind that one click, the platform provisions:

• a Git repository with a starter commit (your survey and scripts),
• a database that will hold this project's responses,
• automatic validation on every commit (a status dot by the project name), and
• the default branch main.

What's in the repository

Open Repository to see the project's files:

Path	What it is
siamang.yaml	The project's configuration: the survey entry point, deployment environments, and which analysis scripts to run. See Project Config (siamang.yaml).
survey/questionnaire.py	The survey itself, written as Python.
scripts/	Analysis scripts (in the example: cleaning.py — data cleaning, and tables.py — frequencies, a crosstab with a chi-square test, and a Markdown report).
outputs/	Where analysis runs write tables and files.
reports/	Generated reports (Markdown / HTML).
README.md	The project's notes, shown on the Dashboard.

Take the example for a spin

If you started from the example study, you can run the whole pipeline right away:

1. Repository — open survey/questionnaire.py and read the survey as code.
2. Deployments — deploy the pilot environment and open the public link. See Deploying a Survey.
3. Database / Dashboard — watch responses (next to the ~300 samples).
4. Analysis — click Run all and open the report.

The Cloud Quick Start walks this same loop end to end.

See also

Repository & Editing · Project Config (siamang.yaml) · Deploying a Survey · Cloud Quick Start