M-Lab Knowledge Base M-Lab data is a powerful resource for internet advocacy — it’s open, independently verifiable, and produced at a scale that individual communities or organizations cannot replicate on their own. This guide is for advocates, organizers, civil society researchers, and journalists who want to use M-Lab data to document ISP behavior, support communities, and build evidence-based arguments.
What Makes M-Lab Useful for Advocacy
- Independent — M-Lab is not run by any ISP, and data is not controlled by commercial interests
- Open and verifiable — anyone can check the methodology and reproduce the results
- Long record — continuous measurements since 2009 enable before/after comparisons and trend analysis
- Covers throttling — the WeHe test specifically detects whether ISPs treat app traffic differently
- Freely accessible — no subscription required to access the data
Start Here
- Welcome to M-Lab: Open Internet Measurement — what M-Lab is and why open measurement matters
- Beyond Speed: Understanding Internet Quality Metrics — why download speed headlines often don’t capture what users actually experience; latency, packet loss, and quality under load
- How M-Lab Measures Internet Speed: NDT7 and MSAK — what NDT actually measures and what it doesn’t
Detecting Throttling and Traffic Discrimination
One of M-Lab’s most directly advocacy-relevant tools is WeHe, which tests whether your ISP treats traffic from specific apps (YouTube, Spotify, Amazon Video, etc.) differently from other traffic.
To run a WeHe test now: WeHe is available as a free mobile app at wehe.meddle.mobi. Each test you run contributes to the public dataset.
WeHe works by sending real app traffic and comparing it to randomized (bit-inverted) traffic. A statistically significant difference means the ISP is applying app-specific treatment — throttling, deprioritization, or blocking — not just experiencing general congestion.
- WeHe — Traffic Differentiation Detection — how WeHe works, how to interpret results, and how to query the historical dataset in BigQuery
Broadband Speed and Coverage Evidence
For characterizing ISP performance in a community, region, or across ISPs:
- NDT (Network Diagnostic Tool) — M-Lab’s flagship speed test, with 15+ years of data. Can be filtered by geography (country, region, city) or by ISP (ASN).
- Analyzing M-Lab Data: A Researcher’s Guide — patterns for ISP comparison and geographic analysis
Accessing the Data (Free)
- Setting Up Free BigQuery Access — register for free BigQuery access with your Google account
- Getting Started with M-Lab Data in BigQuery — run your first query in under 10 minutes
Visualization (No Coding Required)
- M-Lab Observatory — interactive dashboards comparing ISP performance by geography, no SQL needed
Understanding What the Data Shows (and Doesn’t)
Before making claims based on M-Lab data, it’s important to understand the limitations:
- M-Lab Network Annotations: Geolocation, ASNs, and What They Mean — how M-Lab ties measurements to locations and ISPs, and where this can be imprecise
- FAQ: Geolocation Limitations in M-Lab Data — city-level location data is often imprecise; ISP (ASN) level analysis is more reliable
- FAQ: Test Rate Limits — M-Lab limits tests to 40/day per IP to protect data integrity; good to know if you’re organizing community measurement campaigns
Important caveat: M-Lab data comes from users who choose to run a test. This means it’s not a random sample of all internet users — people who run speed tests may be more likely to be experiencing problems. This selection effect should be acknowledged in any public-facing analysis.
The Internet Quality Barometer
The Internet Quality Barometer (IQB) is M-Lab’s framework for moving beyond speed to evaluate overall internet quality. It scores connections across real use cases — web browsing, gaming, video calls, streaming — and is designed to make internet quality more legible to policymakers and the public.
- Beyond Speed: Understanding Internet Quality Metrics — introduction to the IQB and why quality matters beyond Mbps
- IQB Framework — full framework documentation
- IQB Executive Summary — concise, non-technical summary
How Advocates Have Used M-Lab Data
- Net neutrality proceedings — WeHe and NDT data have been submitted as evidence in regulatory proceedings documenting ISP throttling practices
- Broadband mapping challenges — comparing M-Lab speed data to FCC maps to document discrepancies between reported and experienced coverage
- Digital equity campaigns — showing speed and quality gaps between neighborhoods, demographics, or rural vs. urban areas
- ISP accountability reporting — journalists and watchdog groups have used NDT data to compare ISP performance claims against actual user experience
See M-Lab Publications — Government/Regulatory Filings for examples.
Community Measurement
If you want to organize your community to contribute measurements:
- Point people to speed.measurementlab.net — each test contributes anonymously to the public dataset
- BYOS Program — organizations can host their own M-Lab node to add local measurement capacity and generate data closer to the community
Contact
For questions about using M-Lab data in advocacy campaigns, reports, or proceedings, contact support@measurementlab.net. M-Lab staff have supported civil society organizations with data interpretation and expert testimony.