Your analytics command center for QoE and network quality data.
A web-based platform that brings together all your measurement data — video streaming quality, web performance, speed tests, CDN delivery, and network diagnostics — in interactive dashboards. Detect anomalies automatically, drill into single sessions, and manage your entire probe fleet from one place.
Product Overview
The Surfmeter Dashboard is where measurement data becomes actionable. Browse aggregated KPIs, drill into individual sessions, and filter by client, study, geography, or network. Anomaly detection watches every probe and service in the background and groups related alerts into incidents, so you look at problems instead of noise. On the admin side, manage clients, registration keys, capabilities, and user access from a single interface.
Key Features
Browse video, web, network, speed, and conferencing measurements in one place. Filter by client, group, tag, study, URL, domain, ISP, ASN, country, or city, pick the KPIs to plot, and set the aggregation interval and time range — with Lucene queries for anything more specific.
Every probe is compared against its own baseline and against the median of the whole fleet. When a KPI drifts far enough for long enough, the dashboard marks the episode and shows you exactly what changed and when.
Related anomalies are correlated into a single incident — service-wide when several probes see the same problem, probe-level when one location misbehaves across services. Severity, affected probes, and KPI shifts are on the overview.
See response times broken down by content delivery network across the page and video requests of your measurements — Akamai, Cloudflare, Fastly, Google, Amazon, Microsoft, and more — and find out which delivery path is holding quality back.
Ask questions about your data in natural language. The LLM-powered assistant queries your measurements, explains individual results, and surfaces trends — all context-aware with your current filters and time range.
Manage clients, client groups, tags, and studies. Add users and roles, define measurement capabilities, issue registration and API keys, and track license usage — all from a single admin interface.
Each KPI is tracked against the probe's own baseline and against the median of all other probes measuring the same service. When the rolling median stays outside the expected range through the detection window, the dashboard marks the episode on the chart — so you can tell a genuine local degradation apart from a problem the whole fleet is seeing.
Anomalies rarely happen alone. Surfmeter groups them by what they have in common: a service-wide incident when several probes degrade on the same service, a probe-level incident when one location goes bad across services. Each entry shows severity, how long it has been running, the affected probes, and how far the KPI moved.
Every key metric is graded with traffic-light indicators — green, yellow, or red — based on scientifically defined thresholds. P.1203 MOS scores, Core Web Vitals, Lighthouse scores, and network KPIs are graded automatically, and each metric explains what it measures and where its thresholds come from, with references to the underlying ITU-T standards and research.
Every video session becomes an interactive timeline — initial loading, quality switches between SD and HD, and stalling events mapped across the whole playback. You can see at a glance how long a stream took to start, which resolution it settled on, and where the experience broke down.
Surfmeter identifies the CDN behind every request your measurements make and aggregates response times per provider. Compare Akamai, Cloudflare, Fastly, Google, Amazon, Microsoft, Meta, and others over time to see whether a quality drop comes from your own service or from the network delivering it.
A normalized score across all measured types — video, web, network, speedtest, and conferencing — ranks your clients from best to worst. It is the fastest way to see which locations need attention, and which measurement types a given probe is actually covering.
The built-in assistant turns a question in plain language into queries against your own data and answers with numbers you can check — "How was Netflix quality last week?", "Which clients had the most stalling?". It respects your current filters and time range, and one click on any measurement gives you a plain-language explanation of what its KPIs mean.
Get a personalized demo and see how Surfmeter fits your use case.
How It Works
The Dashboard works with data from all Surfmeter products — Automator probes, Mobile SDK devices, and Player SDK integrations.
Measurement clients send results to the Surfmeter platform in real time. Data is enriched with geolocation, ISP, CDN, and device context on ingestion.
Use the measurement explorer to browse by type (video, web, speed, network, conferencing), filter by client, study, time range, and any combination of dimensions.
Anomaly detection runs continuously against per-probe baselines and the fleet median, and groups what it finds into service-wide or probe-level incidents on the reports page.
Open any measurement to see the full session — color-graded KPIs, the playback timeline, and CDN breakdown — or ask the AI assistant a question about your data in plain language.
Data & KPIs
Use Cases
Proven in Practice
“We're using Surfmeter to gain unprecedented visibility into our network performance through automated, active measurements that simulate actual streaming video quality, exactly as end-users experience it”
Gcore, operating a global CDN with 210 points of presence across six continents, needed deeper insights into the interactions between streaming applications and their edge network. Traditional server-side monitoring couldn’t capture what end-users actually experience. With Surfmeter, Gcore gained real-world QoE visibility by measuring actual video streams from the same platforms and players that end-users rely on.
Results
Get a walkthrough of the Surfmeter Dashboard and see how it turns raw measurements into actionable insights.
Thank you! We'll get back to you shortly.
Something went wrong. Please try again.
