Study Settings

Opening the study settings

Open the study in the Study Builder (Edit in the studies list) and select the Settings tab. It has six sub-tabs: General, Schedule, Messaging, AI, Security and Danger zone. The Settings tab always opens on General.

Most settings on these sub-tabs are saved by their own buttons or take effect immediately. They do not depend on the Save button in the Study Builder header. Options that require a paid plan are checked against the plan of the study's author. Account-level settings, such as the password and the encryption vault, are described on the Account, security and billing page.

General: study status and templates

Study Status

The Study Status section shows one card per status: Draft, In Review, Active, Completed, Paused and Archived. Clicking a card changes the status immediately, except Active: a study becomes Active only through Validate & Activate in the Study Builder header, which runs the activation checks and schedules the enrolled participants' notifications, so the Active card opens it. Moving an Active study to another status cancels its unsent notifications. The meaning of each status is described under Study statuses.

Create Template

A template stores the study design for reuse. It copies the last saved version of the study's description, surveys (including their notification schedules), sensor settings and participant management settings. Informed consent, prescreening, engagement settings, branding and encryption settings are not part of a template.

1. Save the study, so that the template contains the current design.
2. In Settings → General, enter a Template Title. Titles do not need to be unique.
3. Click Create Template. The message "Template created successfully!" confirms the result.

The template appears under My Templates at /dashboard/templates, where it can be deleted with the trash icon. Creating a study from a template requires the Basic plan or higher (see Starting from a template).

Schedule: automatic closing and follow-up

Scheduled Close

A study can close at a set time. Close at sets the date and time when the status changes to Completed; its unsent notifications are then canceled. The time is entered in the browser's local time zone, and the saved value is shown below the field with the time zone. A background job checks the schedule every five minutes, so the change can take effect up to about five minutes after the set time. Click Save schedule to store the time or Clear schedule to remove it. This option requires the Basic plan or higher. There is no scheduled opening: data collection starts when the study is activated with Validate & Activate.

Scheduled Follow-up

A follow-up wave links this study to another study owned by the same account. When the toggle is set to Enabled, participants who complete this study are invited to the selected Follow-up study after the number of days in Delay (days after completion), from 1 to 365. A daily job creates the invitations. Click Save follow-up config to store the setting or Clear config to remove it. If the account has no other study, the page asks you to create a second study first. This option requires the Pro plan or higher.

Messaging: email and SMS branding

Emails and SMS messages sent by the study can carry the research team's identity. The branding is applied around the content that each notification schedule defines. Empty fields fall back to the SMAAT defaults.

• Sender name: the "From" name of emails.
• Reply-to email: the address that participant replies go to.
• Logo URL: an image shown at the top of emails (at most 48 px high).
• Accent color: the colour of email headings and buttons.
• Email footer / signature: text added above the unsubscribe line.
• SMS signature: text appended to each SMS. It adds to the message length and therefore to the number of SMS credits used.

Click Save branding to store the fields. Branding requires the Pro plan or higher. The delivery channels themselves are configured per schedule (see Notification Channels).

AI: provider key, limits and privacy

Studies with AI questions or AI-tailored surveys need an AI provider. When an AI question runs, the app sends each message to the SMAAT backend, which calls the configured provider with the study's key. Participants never see the key. The AI configuration panel has these settings:

• API key: Set, Replace or Clear the provider key. The key is stored encrypted, and only its last four characters are displayed.
• Model: a preset OpenAI model, or Custom (other provider)… with the exact model id of another provider.
• Daily token cap (per study): when the day's tokens exceed this number, further AI calls are refused until midnight UTC. A value of 0 disables the cap.
• Provider base URL: the HTTPS address of an OpenAI-compatible provider, for example a gateway or an EU-hosted endpoint. Examples are listed under AI providers.
• Participant AI disclaimer: text shown to participants before their first AI question. A platform default is used when it is empty.
• Identifier redaction (privacy): replaces email addresses and phone numbers with placeholders before messages are sent to the provider. It is pattern-based and does not detect names, addresses or other details in free text. Stored responses keep the original text.
• Content moderation (safety): screens participant messages with the provider's safety classifier and records the result in the AI usage log. Flagged messages are not blocked; the AI usage panel lists those flagged in the last seven days, with the date, the participant code and the question.
• Crisis resources (optional): text, such as a helpline number, shown with AI items in sensitive studies.
• Extraction confidence threshold: for structured extraction, the fraction of goals that must be filled before the app stops flagging the result to the participant (0 disables the check).
• AI-tailored surveys: drafts survey content for each participant from their earlier responses. Every draft must be approved in the study's AI Review tab before it is sent. Enabling it requires a provider key, server-side encryption (the drafting engine has to read responses) and a confirmation checkbox.

The AI usage panel below shows the tokens used today, the tokens and estimated cost of the last seven days, a daily chart and, with content moderation on, the flagged messages.

Security: data encryption

The Data encryption card in Settings → Security offers three mutually exclusive options. Changing the option affects newly collected data only. Data already collected keeps the encryption it was saved with.

How encryption works, as a video

An animated explainer (3:54): who holds the key under each option, what end-to-end encryption covers and what stays readable, where the data is decrypted, your encryption vault and recovery code, sharing with colleagues, and what changes when you switch options. The narration is a synthetic voice; turn on captions in the player to read along.

Recorded in October 2026. Screens in newer versions of SMAAT may differ slightly. Chapters, transcript and all video tutorials →

• End-to-end encryption (default): data is encrypted on the participant's device with a key that belongs to the study and is unlocked through your account. Data is decrypted in the browser while you are signed in, and no key files are involved. This option requires your encryption vault (account Settings → Security; see Encryption vault). Without a vault, selecting it shows "Set up your encryption vault first (Settings → Security)."
• Server-side encryption: SMAAT manages the keys. Data is encrypted in transit and at rest and protected by access controls, but it is technically readable by SMAAT as your data processor. AI-tailored surveys and the gamification trigger Specific Answer require this option, because they read the answers on the server, and the Data Explorer shows answers only in studies that use it.
• Self-managed study key (advanced): a key pair is generated for the study in the browser, and the private key is downloaded as a file. The private key is never sent to SMAAT. The file is uploaded again to view the data. Access cannot be recovered if the file is lost.

New studies use end-to-end encryption when the author has a confirmed encryption vault, and server-side encryption otherwise.

Who can decrypt this data

With end-to-end encryption, the card lists the project collaborators under Who can decrypt this data. Collaborators who have set up their own encryption vault are granted access automatically. Access is granted by wrapping the study key to the collaborator's account, so no key files are exchanged. Each entry shows Has access with a Revoke button, Share access, or "Hasn't set up encryption". Sharing and revoking require your vault to be unlocked in the current session (Unlock vault).

Managing a self-managed study key

When Self-managed study key (advanced) is selected, the Encryption Settings card appears below. Until a key pair exists, it shows the warning "No keys generated yet — participant data cannot be sent", and the activation dialog repeats it. Until then, the SMAAT app keeps participants' data on their phones and sends it, encrypted, once the key exists; phones with an app version older than 1.4.7 send it without encryption.

1. Click Generate Encryption Keys. The browser generates the key pair. The public key is stored with the study ("Public Key (Stored with Study)") and the private key is shown under "Private Key (For Researcher)". The private key stays in the browser and is not sent to SMAAT.
2. Click Download Private Key to save the key as a .pem file. The private key is displayed only once and is not shown again after leaving the page.
3. Store the file securely, for example in a password manager. It is needed to decrypt the study's data on the Data page.

Regenerate Keys replaces the key pair after a confirmation ("Warning: Regenerating Encryption Keys", Yes, Regenerate Keys). The old private key still decrypts data collected before the change, and the new private key decrypts only data collected after it, so both files have to be kept.

Delete Public Key removes the public key after a confirmation ("Warning: Deleting Public Key", Yes, Delete Public Key). Data collected afterwards stays on the participants' phones until a new key pair is generated (app versions older than 1.4.7 send it without encryption). Data encrypted earlier can still be decrypted with the private key.

Danger zone: deleting a study

Deleting a study is permanent. It removes the study design and everything collected in it: participants and their enrolments, survey responses, sensor data and uploaded files, consent records, invitations, the waitlist, notifications and messages. Payment records are kept for accounting without the study link. Download any data you need before you delete the study. The deletion takes three steps in Settings → Danger zone → Delete Study:

1. Click Confirm Deletion.
2. Type the exact study name in the field "Please type the study name to confirm deletion:" and click Next. If the name does not match, the message "Study name does not match." appears.
3. Tick "I agree to delete the study with all its participants' data." and click Delete Study.

After deletion, the studies list opens. If the study is only to be closed, set its status to Completed or Archived instead.

Next Steps