With millions of results obtained with Speedtest® each day, Ookla® has the most accurate and comprehensive view of worldwide network performance, quality and accessibility. Speedtest Awards acknowledge operators that provide the Fastest Mobile Network, Fastest Fixed Network and Best Mobile Coverage in their respective markets. Ookla's methodology for determining Speedtest Awards winners is based on a robust set of criteria that is detailed on this page.
Speed Awards MethodologyCoverage Awards Methodology
For a given location, Speedtest Awards consider only operators that provide service for a significant number of customers in that geographic area. If an operator is fastest in a small city or region, they may not be eligible to receive a national level award because they only serve a small portion of that country. Additionally, an operator must meet a minimum threshold of the number of unique devices testing each day over a six month period to be considered for a geographic area. By instituting these parameters, Ookla ensures that recognition is given to operators whose services are available to and used by a significant portion of the population in a location.
To analyze the data behind Speedtest Awards, Ookla aggregates consumer-initiated tests taken with Speedtest after they have gone through a ‘sample construction’ process that creates standardized data points that can be used for advanced statistical analysis. These standardized data points are what Ookla is referencing when the term samples is used.
Samples are constructed to control for variables and ensure that each unique user receives fair representation in the data. Each sample represents the cumulative tests results for each unique device/user per location, per calendar day. This approach improves data accuracy and reduces bias from factors like repeated testing from the same device during a short time period. It is also one of the many tactics Ookla takes to prevent any external skewing of Speedtest data that may aim to produce fraudulent or inaccurate results.
When analyzing fastest operators, Ookla solely considers top carriers (all carriers with 3% or more of total test samples in the market for the entire award period). We then determine the fastest carrier using Speed Score which incorporates a measure of each provider’s download and upload speed to rank network speed performance (80% of the final Speed Score is attributed to download speed and the remaining 20% to upload speed). The Speed Score uses a modified trimean to demonstrate the download and upload speeds that are available across a provider’s network. We take speeds from the 10th percentile, 50th percentile (also known as the median), and 90th percentile, and combine them in a weighted average using a 1:8:1 ratio, respectively. We place the most emphasis on the download speeds and median speeds as those represent what most network providers’ customers will experience on a day-to-day basis.
To determine Speedtest Awards winners for fixed networks, Ookla assesses the fastest speeds attained across a given network, regardless of the performance tier for which an individual is paying. Since access to ISP service happens over both wired and Wi-Fi connections, Ookla takes into account tests that are taken on various Speedtest applications that connect to a fixed network and analyzes them based on Speed Score. This includes tests taken on mobile phones through a Wi-Fi connection. Each top carrier accounts for at least 3% of the sample size in the geographic area.
Browse Fastest Fixed AwardsTo determine Speedtest Awards winners for mobile networks, Ookla assesses the average of test results taken using Modern Chipsets. Modern Chipsets includes all mobile tests, regardless of connection technology used, as long as they are taken on devices that are identified as being capable of achieving the fastest speeds available in a market. After this filtering has taken place, Speed Score is applied as outlined above to determine the best speeds available in a country. Each top carrier accounts for at least 3% of the sample size in the geographic area. The Fastest 5G Mobile Network Award evaluates the speeds on 5G to determine the highest Speed Score on the technology.
Browse Fastest Mobile AwardsUsing billions of measurements from hundreds of millions of mobile coverage scans collected by Speedtest® and Consumer QoETM users each day, Ookla® has developed a robust and comprehensive view of worldwide mobile coverage. We have established an accurate and consumer-relevant methodology using extensive data analytics to determine the best mobile coverage on networks around the world. Our unbiased, data-driven approach controls for extraneous variables and ensures an equitable accounting for all coverage scans.
Ookla’s data analysis process for mobile coverage involves four primary steps: collection, filtering, spatial aggregation, and summarization. The results of that process are then leveraged to calculate and determine the Mobile Coverage Award recipients within a country.
Hundreds of millions of coverage scans are received daily from Android Speedtest and Consumer QoE users from around the world.
Several data quality filters are applied to ensure the data represents true network conditions. Coverage scans are excluded from devices that are misconfigured or inactive. Because coverage is a spatially-focused metric, only scans with precise and legitimate location information are included.
Filtered data are grouped into spatial bins using the h3 hexagonal hierarchical geospatial indexing system with resolution 7. This produces hexagons with an average area of approximately 5000 m2, which is roughly the area served by a typical cell sector.
We classify a scan as “in coverage” if the device is successfully registered on a cell network when the scan is collected and “not in coverage” if the device is out of service or has only emergency service when the scan is collected.
We then assess a provider’s coverage within each hexagon. A hexagon is eligible for use in coverage calculation if it has at least 5 scans and at least 2 users. We classify each eligible hexagon as “in coverage” if at least 50% of the scans within are classified as “in coverage.” Finally, we compute the total number of hexagons classified as “in coverage” divided by the total number of hexagons in the country (multiplied by 100) to arrive at the Coverage Score™ for a particular provider.
All Speedtest Awards are based on six months of historical data within the same calendar year, called the award period. By using a six-month time period, coverage can be examined at a fine spatial scale, and any temporary changes in service can be resolved. This also reduces the likelihood that short-term fluctuations in service will significantly influence the analysis.
The intention of the Speedtest Mobile Coverage Award is to identify the mobile networks with the best coverage that are generally available to consumers in a specific geographic area. To this end, for each provider, we compute the tile share as the total number of eligible tiles (see Coverage Summary) for that provider divided by the total number of eligible tiles (across all providers). In order to be eligible for the award, a provider’s tile share must be at least 50%. If a provider’s tile share is less than 50%, it is an indication that they are not widely available in the country.
We evaluate every mobile network operator to confirm they are actively providing consumer service in the specified region. All providers must remain an individual entity during the entire award period. Mergers and acquisitions will be taken into consideration when determining provider award eligibility. If a provider no longer has a named presence in the market as a result of a merger or acquisition, they will be excluded from Speedtest Awards consideration. Additionally, test results for a merged or acquired provider will be included in the results of their new parent company on the final merger date, potentially impacting the overall Coverage Score of the parent company.
A provider is declared a winner of Best Mobile Coverage only if it consistently surpasses every other provider in the country. To determine this, and correct for any underlying biases in scan counts, we employ Monte Carlo bootstrapping methods to calculate coverage and to evaluate the statistical uncertainty in the data. A one-sided Mann-Whitney U test in both directions is executed for each pair of providers. We use an alpha value of 0.05 to evaluate the significance of the observed differences. We also require an effect size, defined as the percentage difference in coverage ratios between providers, to be greater than 5%.
Methodology updated July 2024. For previous methodology, see 2023 Mobile Coverage Methodology
Top-Rated Awards are designed to evaluate, for a specific location and time period, whether an operator has the highest star rating. We use our sampled data set of Speedtest 5 Star Ratings to extrapolate and make inferences about consumer sentiment amongst the broader population of consumers. Statistical hypothesis testing provides a formal framework to rigorously and consistently decide whether the highest rated operator can be identified from our sampled data.
To evaluate Top-Rated Awards we compare mean ratings using a modified Hsu’s Multiple Comparisons with Best (MCB) procedure. This evaluates whether a “best” group can be identified from many groups, based upon sampled data sets. In the case of Top-Rated Awards, this implies that we are able to identify the Top Provider with the highest rating. The MCB procedure is similar to performing multiple t-tests comparing each Top Provider to the best of the rest, and then combining these to make a single statistical decision.
Ookla’s Video Experience Awards for fixed and mobile networks are based on Ookla’s Video Score, which combines five measures of the quality of video experience:
When analyzing the Best Video Experience offered by providers, Ookla solely considers providers with 3% or more of total video test samples in the market for the entire award period. To ensure that the data is accurate and reflective of actual consumer experience, each video test result is subject to Ookla’s proprietary data quality filters. To ensure that each user has an equal voice, we apply a user aggregation methodology to video test results. This means that a user can submit any number of video test results during any arbitrary time period, but they will only contribute one “vote” that represents their typical video experience during a given period of observation.
Gaming Experience Awards for mobile and ISP networks are based on Ookla’s Game Score, which combines four measures of the quality of gaming experience:
When a mobile provider wins both the Fastest Mobile Network and Coverage Awards in their market, they may claim to be the “Best Mobile Network” in the country. Best Mobile Network may also be claimed when a mobile provider wins the Fastest Award and no provider is determined to have better coverage based on statistical tests.
Ookla licenses data access to help companies both understand and demonstrate the benefits of their networks. With claims backed by Ookla, you can differentiate yourself in the market with meaningful, verifiable marketing and brand campaigns that potential and existing customers trust.
Licensees can make Noncompetitive Marketing Claims about their network alone or Competitive Marketing Claims which can be made about a network in relation to others in the market; this could be made by a provider or another company using the data to make a comparison.