Executive-ready dashboards
Monitor traffic, revenue, conversion rates, and active visitors from a single unified view. The overview dashboard aggregates pageviews, unique visitors, session duration, and bounce rate into time-series charts powered by Recharts, with date range selectors that let you compare any two periods side by side. Drill from summary KPIs into source-level detail without context switching: click a traffic metric to see the referrer, campaign, or channel that drove it. For example, a marketing lead reviewing weekly performance can identify that organic search traffic rose 18% while paid social declined, then pivot directly to the traffic sources panel to investigate which campaigns underperformed. The dashboard loads in under 400ms on a warm cache thanks to TanStack Query prefetching and server-side aggregation, so leadership reviews stay focused on decisions rather than waiting for data.
Funnel and journey diagnostics
Define multi-step funnels by selecting a sequence of page URLs or custom events, then analyze drop-off rates at each stage broken down by project and date range. The funnel engine runs async SQL aggregation queries against your event stream, returning stage-by-stage conversion percentages and the median time between steps. When growth numbers shift unexpectedly, you can overlay journey paths to see where visitors actually went after abandoning a funnel step rather than guessing. A practical example: an e-commerce team notices checkout completions dropped 12% week-over-week. The funnel diagnostic reveals that 68% of users who reached the payment page navigated back to the cart, suggesting a pricing or shipping cost friction point. Investigation cycles that previously required exporting raw data to a spreadsheet now resolve inside the tool in minutes.
Retention and cohort visibility
Track repeat visitor behavior over configurable weekly or monthly cohort windows to separate short-term campaign spikes from genuine product adoption. The retention grid displays return rates for each cohort as a color-coded matrix, making it immediately visible whether users acquired in a given week are returning at higher or lower rates than previous cohorts. Retention data is computed per-project, so multi-product teams can compare adoption curves across properties. For instance, a SaaS product team might discover that users acquired through documentation pages show 34% higher Day-14 retention than users from social ads, informing a reallocation of acquisition spend toward content marketing. Cohort calculations run server-side against indexed session data, keeping render times consistent even with millions of tracked sessions.
Session replay and heatmaps
Pair quantitative dashboard metrics with qualitative evidence by replaying real user sessions and viewing aggregated click and scroll heatmaps. In international mode, the tracking snippet records mouse movements, clicks, scroll depth, and DOM mutations (with form inputs automatically masked) as batched events sent to the replay endpoint. Support and product teams can search sessions by URL, device type, or conversion status, then watch exactly what happened during a frustrating experience. Heatmaps aggregate click coordinates across all sessions for a given page URL, revealing which elements attract attention and which are ignored. A UX team investigating a low-converting landing page might discover via heatmap that 41% of clicks hit a non-interactive hero image, prompting a redesign that adds a CTA button in that visual zone. Replay is disabled entirely in GDPR mode to respect strict privacy requirements.
Privacy and compliance control
Choose between two clearly defined collection modes at the project level. GDPR mode hashes visitor IPs using SHA-256 with a configurable salt before storage, reduces user-agent strings to browser and OS family only, disables session replay collection entirely, and restricts geographic data to country and region. International mode retains full IP addresses and user-agent strings, enables session replay with masked form inputs, and provides city-level geolocation with latitude and longitude coordinates for map visualization. The mode boundary is enforced at the API layer, not the client, so switching modes requires a project configuration change rather than a snippet update. This architecture means a European property can run in GDPR mode while an international marketing site uses full collection, all within the same account. Data processing documentation maps directly to GDPR Article 6 and ePrivacy Directive requirements.
Account-level access governance
Manage multiple projects under isolated account boundaries with JWT-authenticated dashboard endpoints and superadmin scoped contexts. Each account owns its projects independently, and dashboard queries are scoped to the active account context, preventing accidental cross-account data access. Superadmins can switch account context via the auth context endpoint to audit or support any account without sharing credentials. Project API keys (prefixed wa_) are generated per-project and used exclusively for tracking ingest, keeping dashboard auth and data collection auth on separate planes. For organizations managing analytics for multiple clients or business units, this means an agency can operate 50 client properties under one superadmin login while each client only sees their own data. Audit trails track context switches so compliance teams can verify who accessed which account and when.
Traffic source attribution
Automatically classify every incoming session into one of five attribution channels: Direct, Organic Search, Social, Newsletter, or Referral. The attribution engine inspects the document referrer, UTM parameters, and known search engine and social network domain lists to assign each visit to the correct channel without manual tagging. The traffic sources dashboard breaks down visitors, sessions, pageviews, and conversion rates by channel, so you can see at a glance which acquisition paths deliver engaged users versus drive-by traffic. For example, a content marketing team might discover that Newsletter traffic converts at 8.2% while Social traffic converts at 1.4%, justifying a heavier investment in email distribution. Attribution data feeds directly into the conversion scoring model, creating a closed loop between acquisition analysis and predictive lead quality.
Conversion likelihood scoring
Every tracked session receives a conversion likelihood score from 0 to 100, computed server-side by the scoring engine. The model weighs five signal categories: device type and screen resolution, browser and OS family, visitor country, traffic source channel, and engagement depth measured by page count, scroll percentage, and session duration. Scores update in real time as session activity accumulates, so a visitor who initially scores 35 after a single pageview might climb to 72 after viewing four pages and scrolling past 80% depth. Product teams can filter sessions by score range to focus replay reviews on high-intent visitors who did not convert, identifying the specific friction points that cost the most revenue. The scoring weights are calibrated against historical conversion data, and the 0-100 scale maps to percentile ranks within your own traffic distribution rather than an abstract benchmark.
Real-time live dashboard
Monitor active visitors as they browse your site with a Server-Sent Events (SSE) powered live stream that pushes updates to the dashboard without polling. The real-time panel displays current active visitor count, the pages they are viewing right now, their geographic distribution, and the traffic source that brought them. Updates arrive within one second of server-side event processing, giving operations teams immediate visibility during product launches, marketing campaigns, or incident response. For example, during a product launch email blast, a growth team can watch active visitors climb from baseline to peak in real time, confirming that the email infrastructure delivered on schedule. The SSE connection is lightweight, using a single persistent HTTP stream per dashboard tab rather than WebSocket overhead, and automatically reconnects on network interruptions.
Geographic visualization
Visualize visitor distribution across the world with interactive Leaflet map and globe views built into the dashboard. In international mode, sessions include city-level coordinates from MaxMind GeoIP lookups, enabling pin-level precision on the map with cluster aggregation at higher zoom levels. In GDPR mode, the map displays country and region-level aggregation without city or coordinate data, maintaining privacy compliance while still providing geographic insight. The globe view renders a rotatable 3D projection that is particularly useful for presentations and stakeholder reviews. A practical use case: an e-commerce brand expanding into Southeast Asia can overlay conversion rates on the geographic map to identify which cities show high traffic but low conversion, signaling potential localization or payment method gaps. Geographic data also feeds the conversion scoring model, as visitor country is one of the five weighted signal categories.