Feedback Collection

Feedback collection is the core purpose of CapySays. A feedback device runs the Collector — a lightweight, self-contained app that displays configured screens to customers and records their responses. This page explains how the whole system works, from the moment a customer walks up to a device to the moment their rating lands in your dashboard.

How Feedback Collection Works

The Collector is a single-file, offline-resilient app that runs directly on the feedback device (a tablet, phone, or any browser-capable touchscreen). It does not depend on the Django admin interface, Tailwind CSS, or any of the management-side tooling — it is purpose-built to be fast and reliable on low-end hardware with intermittent connectivity.

When a device is set to Active, it:

1. Loads the Collector from the device's short URL (/d/<magic-word>).
2. Fetches its configuration from the CapySays API — screens, rating dimensions, staff members, and scheduling rules.
3. Displays the appropriate screen to customers based on the current time and activity state.
4. Submits collected ratings to the API and, when offline, queues them locally for later sync.

Customers interact with the device directly. No login, no app download, and no account is required from them — they simply walk up and tap.

The Customer Experience

1. Customer approaches the device. The Collector displays the first active screen for the current time and schedule.
2. Staff selection (optional). If the screen has Show Staff enabled, the customer can tap a staff member's name or photo from a list before rating. This step is optional — the customer can skip it.
3. Rating. The customer selects their rating using the screen's configured mechanism (stars, emoji, thumbs, or NPS scale). If the screen has multiple rating dimensions (e.g., Service, Cleanliness, Value), the customer rates each one in sequence.
4. Submission. The Collector sends the response to the API (or queues it if offline) and returns to the ready state.
5. Next screen. If the device has multiple active screens, it cycles to the next one automatically, allowing the following customer to respond to a different prompt.

The entire flow is designed to take seconds. There are no account prompts, no loading spinners between steps (beyond initial config load), and no required fields beyond the rating itself.

Screen Types Explained

Screens are the building blocks of the customer-facing experience. Each device can have multiple screens, each with its own type, schedule, and configuration. Screens are displayed in priority order.

Rating Screens

Rating screens are the primary feedback mechanism. They display a question or prompt and ask the customer to respond using a rating mechanism.

Rating mechanisms:

Rating screens can also have multiple rating dimensions — for example, a single screen might ask the customer to rate Service, Cleanliness, and Value one after another. Each dimension uses the same rating mechanism and records a separate value in the submission.

Message Screens

Message screens display text (and optionally an image) to the customer but collect no input. They are useful for:

• Welcome or thank-you messages between rating screens
• Announcements (e.g., "New menu available — ask a team member!")
• Brief informational notices

The customer acknowledges the message by tapping a button and the Collector moves to the next screen.

QR Code Screens

QR code screens display a scannable QR code. The customer can point their phone camera at the code to open a URL — useful for:

• Linking to a longer external survey
• Pointing to a menu, product page, or landing page
• Redirecting to an app download or loyalty programme

No data is recorded on the CapySays side when a customer scans the code. The destination URL and any tracking is handled externally.

Screen Scheduling

Each screen has a schedule that controls when it appears on the device. Only screens whose schedule is currently active will be shown to customers.

Schedules are evaluated in real time. When a screen's schedule opens or closes, the Collector picks up the change on its next config refresh (approximately every 2 minutes) without any manual intervention.

If no screens are currently scheduled for a device, the Collector displays a neutral holding state and continues to poll for updates.

Offline Resilience

The Collector is built to keep working even when the device loses its internet connection. This is important for environments where Wi-Fi is shared, spotty, or subject to brief outages during busy periods.

When the device is offline:

• Feedback is saved locally in an in-device queue (using the browser's local storage).
• The Collector continues to function normally — customers can still approach the device and leave ratings as usual.
• There is no visible error state shown to customers. The experience is uninterrupted.

When connectivity returns:

• The Collector automatically detects the restored connection and uploads all queued feedback to the API.
• No manual action is needed from staff or administrators.

If the device is installed as a PWA (Progressive Web App):

Installing the Collector to the device's home screen enables Background Sync, which allows the device's operating system to sync queued feedback even when the Collector app is not the active foreground window. This provides an additional layer of reliability for high-traffic installations.

Monthly Response Cap

Your CapySays plan includes a monthly limit on the number of feedback responses your organization can collect.

• Each rating submission counts as one response toward the monthly cap.
• The cap resets at the beginning of each calendar month.
• When the cap is reached, feedback devices will stop accepting new submissions. The Collector displays a holding state rather than the rating flow.
• Submissions already queued on the device (from when the device was offline) are still synced and counted on reconnection — plan your capacity accordingly.
• Upgrading your plan increases the monthly response limit immediately.

You can monitor your current usage from the Workspace utilization summary on the organization overview page.

Heartbeats

Active feedback devices send a periodic heartbeat to the CapySays API to report that they are online and functioning. This happens automatically in the background while the Collector is running.

• The Last Seen timestamp on the device management page reflects the most recent heartbeat.
• If a device has not been seen recently, it may be powered off, have lost its network connection, or have experienced a hardware problem.
• Heartbeats are lightweight — they do not count toward your monthly response cap.

Use the Last Seen timestamp as a quick health check when managing a fleet of devices. If a device falls silent unexpectedly, it is worth checking the physical hardware and network connection at its location.

Feedback Data

Every submission captured by the Collector is stored and made available through the CapySays dashboard.

Each submission records:

• The device, location, and branch it was collected on
• The screen that was displayed
• The timestamp of the submission
• The staff member selected (if any)
• The activity in session at the time (if applicable)

Each rating within a submission records:

• The dimension label (e.g., "Service", "Cleanliness")
• The numeric rating value

All submission data is accessible through the Dashboard analytics views. If your plan includes data export, you can also download the full feedback history as a CSV file from the organization edit page.

Related pages

• Devices — configuring screens and devices
• Dashboard and Reports — viewing feedback analytics
• Activities and Scheduling — linking screens to activities
• Glossary