Proxy testing site Proxyway has conducted the first-ever evaluation of proxies for app development & localisation.
Based on the feedback of 30 application development and testing companies, the deep-dive shows how well an often-ignored part of app development technology stack actually performs for the industry’s needs. A total of 9 industry leaders were tested and evaluated in the report.
Read it below to see how the researchers categorised and tested the proxy’s for apps and localisation purposes and discover the findings that their tests revealed.
Every new or localized app needs to go through testing. Whether it’s serving localized content or balancing massive loads, proxy networks often work as testing tools for developers. The networks that Proxyway tested are undeniable market leaders with over 100,000 nodes, going all the way up to 30-40 million monthly servers. Developers use these networks every day to perform load testing, simulate various navigation scenarios, and ensure that the software works well for users in all locales.
Chris Becker, a co-founder of Proxyway, noted that the public image of proxies too often got the short end of the stick, which is why most of the experts surveyed for the report wished to remain anonymous:
‘Whenever you think about proxies or changing one’s IP address, you think about hackers and shady people, not hard-working people from SEO, market research, or app development industries. We hope this will change with time.’
Expert-based testing criteria
Determining how good any given proxy is for app testing has never been done before. Proxyway consulted with 30 experts about what aspects of proxies they find most valuable. The experts distinguished that they look at location coverage first; they also noted price, proxy success rate, and anonymity as vital to app development & localization proxies. Based on the anonymous questionnaires, Proxyway determined weighted scores for each of these criteria:
Co-founder of Proxyway Adam Dubois commented that the most difficult part was getting the answers from top companies in the field:
‘Running the tests is the easy part. We know how to push these servers to the limit, how to measure technical stuff like speed, response codes or proxy success rate. But such technical data alone really lacks context without the know-how of people working in that particular industry. We’re very grateful to the experts for giving feedback,’ he said.
Proxyway’s team of volunteers aggregated the weighted scores into a single number, the Proxyway Standard Proxy Score (or ).
By using it, researchers could objectively compare proxy providers.
Proxyway ran dozens of automated and manual tests of the networks for over a month. They sent hundreds of connection requests per second to measure the location coverage and reliability of the networks and used four online tools to evaluate their anonymity. This gave Proxyway precise results of how the leading proxy provider networks work for app development and localization.
Here is how the aggregated scores stacked up:
Impact of the study
So what does this analysis mean for the software development industry?
First, there finally is a real-world basis for evaluating proxy networks, and the results are impressive: some providers are twice as good for app testing tasks than others. Even among the top five performing networks the usefulness score might deviate by close to 30%. Choosing the right provider can have a big impact on the expenses and success of a company’s development efforts. And, instead of testing a dozen providers, localization testing professionals can have a simple way to weed out the worst proxy networks.
The research project
The research project ran questionnaires and interviews of industry experts from Dec 14, 2019 to Jan 31, 2020. The technical proxy network tests ran throughout Jan-Feb and into March.