SOAX review for multi-account ops in 2026

SOAX launched around 2019 and has been quietly building one of the larger commercial proxy networks available to independent operators. they are not the cheapest option on the market and they are not trying to be. the positioning is squarely at agencies, data teams, and operators who need granular control over which IP they connect from, not just a rotating pool of random addresses. the headline number they advertise is over 191 million residential and mobile IPs across 195+ countries, with filtering down to city and ISP level.

for multi-account operators, that level of targeting matters more than raw pool size. if you are running accounts on a platform that geofilters users, or if you need to match the ISP a residential user in a specific city would actually be on, SOAX gives you the knobs to do that. what i want to answer in this review is whether those knobs work in practice, whether the pricing is sustainable for real ops, and where the service falls short compared to alternatives.

the headline verdict: SOAX is a legitimate tier-2 provider. the infrastructure is real, the mobile proxy offering is one of the better ones i have tested, and the dashboard is functional. the pain points are pricing at scale and support response times that do not match the premium positioning. if you are running low-to-medium volume targeted ops, it is worth testing. if you are running millions of requests per day, you will find the per-GB cost structure painful.

what SOAX actually does

SOAX offers four proxy types from a single dashboard: residential, mobile, datacenter, and ISP (static residential). the residential and mobile pools are the flagship products. residential IPs come from real end-user devices that have opted into the SOAX network, which is the standard model across the industry. mobile proxies use real carrier IPs on 3G, 4G, and 5G networks, which matters for platforms that weight mobile traffic differently or flag datacenter ASNs heavily.

the targeting system is the differentiator. rather than picking a country and getting a random IP, you can filter by country, state, city, ISP name, and on mobile proxies, by OS type (Android or iOS). for each session you can choose between rotating on every request or holding a sticky session for a configured duration. SOCKS5 and HTTP/HTTPS are both supported, which covers the connection requirements for most browser automation and scraping setups.

connections go through endpoint URLs with authentication parameters embedded. the format is a standard proxy host with port, username, and password, plus optional targeting parameters appended to the username string. most tools that accept proxy lists, including Multilogin, AdsPower, and Playwright-based scrapers, can use this format without modification. there is no proprietary client required.

the dashboard shows real-time bandwidth consumption and session stats. you can generate proxy lists in bulk from the UI, which is useful if you are provisioning accounts across an antidetect browser setup. if you are looking at how this fits into a broader fingerprinting workflow, the antidetectreview.org blog has detailed writeups on pairing proxy providers with browser profiles.

pricing

SOAX uses a bandwidth-based pricing model. as of May 2026, their residential proxy plans start at approximately $99/month for 15 GB, which works out to around $6.60/GB. higher-tier plans reduce the per-GB cost: the 100 GB plan comes in around $5.50/GB, and enterprise pricing below $4/GB is available on request. mobile proxies carry a premium, typically running 20-30% higher per GB than residential on equivalent volume tiers.

datacenter proxies are priced differently, starting at lower per-GB rates and available in both shared and dedicated configurations. ISP proxies (static residential) are sold as monthly port rentals rather than per GB, with pricing that varies by country and availability.

there is a free trial available but it is limited. you will not get a meaningful read on connection success rates or session persistence without committing to at least the entry-level paid plan. check soax.com/pricing for current rates because they do adjust plans periodically.

for budget reference: running 100 GB/month of residential traffic through SOAX costs roughly $550 at mid-tier pricing. at that same spend you could get more raw bandwidth from providers like Oxylabs or Smartproxy on their volume plans, though the targeting granularity may differ. the calculation depends heavily on how much of your bandwidth you are actually converting versus wasting on failed sessions.

what works

city and ISP-level targeting is genuinely functional. on most residential proxy providers, city-level filtering is approximate at best. in my testing, SOAX’s filtering held city accuracy at a level that was usable for platform operations where account location matching matters. ISP filtering works as advertised, which is useful when you need to appear as a specific carrier rather than a generic residential block.

mobile proxies use real carrier IPs. the 3G/4G/5G mobile pool is not just datacenter IPs with a label on them. carrier ASNs check out, and the IPs rotate through real mobile ranges. for platforms that use device fingerprinting signals in combination with IP type detection, having a real mobile carrier IP is the baseline requirement.

sticky sessions up to 120 minutes are reliable. session persistence is one of those features that providers advertise but often implement badly. SOAX’s sticky sessions actually hold across requests for the configured duration. for multi-account ops where you need a consistent IP across a login sequence, session activity, and logout, 120 minutes is enough headroom for most workflows.

the proxy format integrates with standard tools. no proprietary client means you can use SOAX IPs directly in Playwright, Puppeteer, Selenium, or any antidetect browser that accepts proxy strings. this is table stakes but some providers still try to funnel you through a custom app.

bandwidth rollover on monthly plans. unused bandwidth rolls over to the next billing month, which matters for ops with uneven monthly traffic patterns. this is not universal across providers and it reduces the pressure to burn through your allocation before reset.

what doesn’t

pricing per GB punishes high-volume ops. the model is fine for targeted, low-waste scraping. it is not fine if you are running large-scale account registration flows or continuous monitoring tasks where bandwidth consumption is high and conversion per GB is low. at $5-6/GB for residential traffic, a 1 TB monthly operation would cost $5,000-6,000 in bandwidth alone.

support response times are inconsistent. the live chat widget on the dashboard is responsive during business hours for basic questions. for technical issues that require actual investigation, expect a wait. i have seen reports across operator communities of 24-48 hour resolution times on technical tickets. for a service priced at the premium end, that gap is noticeable.

the IP pool has quality variance. 191 million IPs sounds large but not all IPs in any residential pool are equally clean. some will be flagged on Cloudflare, Akamai, or platform-specific blocklists. SOAX does not publish granular data on what percentage of the pool is clean against specific targets. you will need to run your own success rate testing against your actual targets before committing.

no built-in proxy health checking. the dashboard shows your usage but does not surface per-IP quality data. if you are building a system that needs to validate IP quality before assignment, you have to implement that testing layer yourself. some competitors have started building basic health indicators into their dashboards.

geo coverage thins outside major markets. tier-1 countries (US, UK, DE, FR, etc.) have deep pools. coverage in Southeast Asian markets, Africa, and parts of Latin America is thinner. for Singapore-based operations targeting regional platforms, you may find the local pool smaller than you need for scale. singaporemobileproxy.com is worth looking at if Singapore and Malaysia carrier IPs are your primary use case.

who should buy

targeted account operators running moderate volume. if your workflow is 50-500 accounts per platform with careful session management, the per-GB cost is manageable and the targeting quality gives you real advantages. this is the use case SOAX is built for.

agencies running geo-verified ad verification. checking that ads are actually serving in specific cities on specific carriers is one of the strongest use cases for SOAX’s filtering depth.

anyone needing real mobile carrier IPs. if you are building automation that specifically needs to look like mobile carrier traffic rather than residential broadband, SOAX’s mobile pool is one of the better options available at a commercial API level.

who should skip

high-volume scrapers running millions of requests per month. the math does not work. you need either a provider with flat-rate or significantly cheaper per-GB pricing, or a self-managed residential pool. see the proxy scraping resources at proxyscraping.org if you are evaluating the build-vs-buy question for large-scale scraping.

operators primarily targeting Southeast Asian local platforms. pool density in markets like Thailand, Vietnam, and Indonesia is not deep enough for sustained high-volume ops. you will hit IP reuse rates that create account linkage risk.

anyone on a tight budget testing a new platform. the entry price point is high enough that it is not the right tool for initial validation. start cheaper, validate your workflow, then upgrade infrastructure.

alternatives to consider

Oxylabs. larger claimed pool, similar pricing tier, stronger enterprise support SLAs. better option if you need dedicated account management or have compliance requirements around vendor contracts. see our residential proxy comparison at /blog/residential-proxy-providers-compared/ for a side-by-side breakdown.

Bright Data. the largest residential network available commercially, with more sophisticated session management features and a proxy browser product. significantly more expensive, with a steeper learning curve on the dashboard. worth it at enterprise scale.

IPRoyal. cheaper per-GB residential pricing with a smaller but usable pool. no mobile proxy product at the same quality level, but if you are doing residential-only ops and price sensitivity is high, IPRoyal is worth testing.

verdict

SOAX is a solid mid-tier residential and mobile proxy provider. the targeting system is the strongest thing they offer, and the mobile proxy infrastructure is legitimate. the per-GB pricing model is the ceiling that limits how far you can push it without the costs becoming a problem. for targeted multi-account ops where quality per session matters more than raw volume, it is worth the trial. for high-volume commodity scraping, look elsewhere.

check the multiaccountops.com blog for more proxy and tooling reviews as we add them through 2026.

Written by Xavier Fok

disclosure: this article may contain affiliate links. if you buy through them we may earn a commission at no extra cost to you. verdicts are independent of payouts. last reviewed by Xavier Fok on 2026-05-19.