Awards methodology
Fastest Mobile Network
As part of our mission to improve internet connectivity for everyone, we celebrate the mobile and fixed network operators delivering the best quality of service to their customers.
This page describes the methodology behind two SpeedChecker Champion award categories:
- Fastest Mobile Network — the eligible operator with the highest average download throughput across valid measurements during the award period, subject to the eligibility and award decision criteria below.
- Fastest 5G Mobile Network — the eligible operator with the highest average download throughput across valid 5G measurements during the award period, subject to the same eligibility and award decision criteria.
Both awards are based on crowdsourced measurements collected from real end users, using the methodology set out below.
Data collection methodology
Our data is collected from end-user devices running Android and iOS. All measurements are executed against a CDN with a large geographical footprint that hosts a significant share of the content users actually access. This provides an end-to-end view of internet performance that is relevant to the experience delivered to users.
Measurements are accepted only from applications approved by SpeedChecker. Award measurements are predominantly generated through automated background testing rather than being initiated on demand by the user. Users therefore do not control when award tests are executed or how many measurements are contributed, making deliberate user-driven gaming through repeated testing inapplicable to this methodology. Every submitted measurement is checked against expected value ranges, and additional security controls are in place to detect and reject manipulated measurement data.
Each measurement is attributed to an operator and award market using the network information available for the measurement together with validated network-mapping methods, including IP-based mapping where applicable. Measurements that cannot be reliably attributed to an eligible mobile network operator are excluded from the award dataset. Where a device provides location information, it is used to support geographic analysis and quality control, but location is not a requirement for a measurement to be included in an award dataset.
Every award certificate states the number of unique devices that contributed to the dataset alongside the total sample count, so the balance between panel breadth and sample volume is visible for each award.
Measurement methodology
Our test methodology is based on ITU-T Q.3960 (2016), "Framework of Internet related performance measurements", and Supplement 71 to ITU-T Q-series Recommendations. SpeedChecker is an ITU associate member and has contributed to the development of the ITU crowdsourcing recommendation, ITU-T E.812.
The methodology measures the maximum bandwidth available over a given internet connection. It does this by transferring multiple parallel data streams over separate TCP connections within a fixed time window. The transferred payload is randomly generated high-entropy data, which prevents compression from distorting the result.
| # | Parameter | Unit | Range | Current setting |
|---|---|---|---|---|
| n | Number of parallel threads | # | 1 ≤ n ≤ 10 | 3 – 10 |
| Tp | Duration of pre-load | s | 0 ≤ Tp ≤ 5 | 2 |
| Td | Duration of the downlink subtest | s | 5 ≤ Td ≤ 15 | 10 |
| Tu | Duration of the uplink subtest | s | 5 ≤ Tu ≤ 15 | 10 |
| p | Number of pings during the delay subtest | # | 5 ≤ p ≤ 20 | 10 |
Award calculation — Fastest Mobile Network
The award is calculated per country and covers all mobile network operators with sufficient measurement volume in that country. The data collection period varies by award and is stated on every certificate.
1. Eligibility. An operator enters the ranking only if it has at least the minimum number of valid measurements in the period. The minimum is the greater of 1,000 measurements or 25 measurements per 100,000 population, capped at 10,000. The country dataset as a whole must contain at least 2,000 unique devices, and at least two operators must meet the applicable eligibility thresholds. Operators below the thresholds are not ranked and cannot receive the award.
2. Metric. For each eligible operator we calculate the average download throughput across all its valid measurements in the period. Download throughput is the sole ranking metric. Upload throughput and latency are measured and reported but do not affect this award.
3. Confidence interval. For each operator we calculate the 95% confidence interval around its average download throughput. The interval width reflects both the number of measurements and the variability between them, so operators with fewer or more scattered measurements carry wider intervals. Each operator's sample count and confidence interval are printed on the award certificate.
4. Result. The award is issued to the operator with the highest average download throughput, provided its 95% confidence interval does not overlap the confidence interval of any other eligible operator. This non-overlap requirement is used as a conservative award decision rule. Where the leading operator's interval overlaps that of any eligible competitor, no Fastest Mobile Network award is issued for that country and period.
Award calculation — Fastest 5G Mobile Network
The Fastest 5G Mobile Network award applies the identical calculation to a filtered subset of the same data.
From the measurements collected during the award period, we select only measurements where radio/network signalling information reported by the device identifies a 5G connection at the time of measurement. Both 5G Non-Standalone (NSA) and 5G Standalone (SA) measurements are included. Measurements identified as 4G, 3G or 2G are excluded.
The same eligibility thresholds, the same average download throughput metric, the same 95% confidence interval calculation and the same non-overlap requirement are then applied to this 5G-only dataset. An operator must meet both the minimum measurement count and the minimum unique-device requirement on 5G measurements alone, which means the award is only issued in markets where 5G is commercially deployed and carries enough real user traffic to measure reliably.
Scope of award claims
Every SpeedChecker Champion award is specific to a single country, a single named data collection period and the methodology described on this page. Awards are not transferable across countries, periods or network technologies. Any public use of an award must credit SpeedChecker and state the country and the data collection period the award relates to.