Get to a working dashboard in under an hour.

Quickstart

For most teams under 100 people, the path from signed-up to "showing the team in our Friday review" is four steps:

1. Create a tenant at prodview.app/signup. Free for 21 days, full feature set, no card.
2. Install the agent on your own device first (commands below).
3. Roll it out to the team — either by sending a one-line install script in Slack, or via your MDM.
4. Open Overview. Within ~5 minutes of agent install, data starts populating.

Install · macOS (12 Monterey or newer)

# Install the agent
brew install prodview/tap/agent

# Register against your tenant
prodview-agent register --tenant "acme" --token $PV_TOKEN

# Start (also runs on every boot)
prodview-agent start

# One-line install
curl -fsSL https://get.prodview.app/agent.sh | sh -s -- \
  --tenant "acme" --token $PV_TOKEN

# Notarised .pkg for MDM, manual, or restricted-network installs
curl -O https://dl.prodview.app/agent/0.2.1/prodview-agent.pkg
sudo installer -pkg prodview-agent.pkg -target /

Install · Windows (10 / 11)

# PowerShell as Administrator
iwr -useb https://get.prodview.app/agent.ps1 | iex
prodview-agent register --tenant "acme" --token $env:PV_TOKEN

Agent runs as a Windows service (prodview-agent.exe) under NT AUTHORITY\NetworkService. Auto-update is on by default and respects WSUS policy.

Install · Linux (Ubuntu 20.04+ / Debian 11+)

# Add the apt repo
echo "deb [signed-by=/usr/share/keyrings/prodview.gpg] https://apt.prodview.app stable main" \
  | sudo tee /etc/apt/sources.list.d/prodview.list

curl -fsSL https://apt.prodview.app/gpg | sudo gpg --dearmor -o /usr/share/keyrings/prodview.gpg
sudo apt update && sudo apt install prodview-agent

Deploy via MDM

Silent install scripts ready to drop into your fleet manager:

Data model

ProdView aggregates raw OS signals into a small, opinionated data model. Every metric in the dashboard rolls up from these four tables:

Raw signals (keystroke counts, screen contents, message bodies, file metadata) are never written by the agent — see the privacy model for the explicit exclude list.

App categories

Every app and domain is classified into one of five buckets:

• Productive · code editors, design tools, terminals, docs
• Communication · Slack, Zoom, Gmail
• Neutral · file browsers, music, OS utilities
• Distracting · YouTube, Reddit, social
• Idle · no input for ≥3 min

The default taxonomy covers 15,000+ apps and 4M+ domains. You can override per-team — Figma is Productive for designers and Neutral for sales, for instance.

Focus time

A focus block is a run of ≥25 contiguous minutes on a single Productive-category app with no idle, no app-switch, and no meeting overlap. We treat it as the closest proxy for actual deep work that we can compute without seeing screen contents.

Privacy model

The full data manifest is published and version-controlled at github.com/prodview/manifest. Auditors get read access to the live production schema.

API · authentication

All endpoints under https://api.prodview.app/v1 require a bearer token. Tokens are scoped (read / write / admin), can be IP-allow-listed, and rotate on a 90-day cadence by default.

# Set your token
export PV_TOKEN="pv_live_a82f1c3..."

# Sanity check
curl https://api.prodview.app/v1/me \
  -H "Authorization: Bearer $PV_TOKEN"

People endpoints

GET/v1/people — list every person in your tenant.

GET/v1/people/:id — one person, current activity, last 28 days.

POST/v1/people/:id/pause — pause data collection for a person.

# Get the current activity for one person
curl https://api.prodview.app/v1/people/p01 \
  -H "Authorization: Bearer $PV_TOKEN"

# Response (abridged)
{
  "id": "p01",
  "name": "Amelia Okafor",
  "team": "Platform",
  "activity": {
    "active_hours_today": 7.4,
    "focus_hours_today": 5.2,
    "idle_ratio": 0.12,
    "score": 88
  }
}

Metrics endpoints

GET/v1/metrics/:name — query any rolled-up metric with a flexible window + groupBy.

curl "https://api.prodview.app/v1/metrics/productive_hours?window=P28D&groupBy=team" \
  -H "Authorization: Bearer $PV_TOKEN"

Security events

GET/v1/events — list security events with filters by severity, type, person.

POST/v1/events/:id/ack — mark as triaged + add a note for the audit log.

Webhooks

Every metric and alert can fire a webhook. Payloads are signed with HMAC-SHA256 using your endpoint secret.

# Create a webhook for high-severity events
curl -X POST https://api.prodview.app/v1/webhooks \
  -H "Authorization: Bearer $PV_TOKEN" \
  -d '{
    "event": "security.event.high",
    "url": "https://events.pagerduty.com/v2/enqueue",
    "secret": "whsec_..."
  }'

Currently emitted event types: security.event.{low,medium,high}, person.overwork.detected, device.stale, policy.changed.

Policy & retention

Per-tenant policy YAML controls collection intervals, retention windows, off-hours definitions, blocked-domain lists, and screenshot policy (off by default).

SIEM export

Push every event in real time to Splunk, Datadog, or any HEC-compatible endpoint. Sub-second latency for high-severity events.

Terraform provider

terraform {
  required_providers {
    prodview = { source = "prodview/prodview", version = "~> 0.2" }
  }
}

resource "prodview_policy" "default" {
  screenshot_interval = "never"
  retention_days      = 425
  off_hours_window    = "19:00-07:00"
}

Troubleshooting

Most issues we see fall into three buckets. Run prodview-agent doctor first — it prints a coloured triage report.

Still stuck? Email support@prodview.app — we answer in <4h on Business and <1h on Enterprise.