The short answer
- 1 Novada Mobile carrier city reach Visit site
- 2 ProxyEmpire Rollover mobile verification Visit site
- 3 Decodo (ex Smartproxy) Sticky city-level sessions Visit site
17 best proxies for ad verification, compared
| # | Provider | Fit | From | Readers | Protocols | Links |
|---|---|---|---|---|---|---|
| 1 | 4.8/5 | $0.50 / GB | 88% | HTTP(S) | ||
|
100M+ residential IPs with city-level targeting and 2M+ mobile IPs across 120+ carriers, plus 1–120 minute residential sessions and mobile from $1.5/GB. Good for this job
Watch out
|
||||||
| 2 | 4.6/5 | $0.35 / GB | 90% | HTTP, HTTPS, SOCKS5, HTTP/2 | ||
|
30M+ residential IPs and 4M+ mobile IPs in 170+ countries, with mobile targeting by country, city and carrier, residential ZIP targeting and rollover bandwidth. Good for this job
Watch out
|
||||||
| 3 | 4.5/5 | $0.08 / GB | 100% | HTTP(S), SOCKS5 | ||
|
115M+ residential IPs across 195+ locations, targeting to city, ZIP and ASN, sticky sessions up to 30 minutes and mobile proxies from $2.25/GB. Good for this job
Watch out
|
||||||
| 4 | 4.4/5 | $2.20 / GB | 88% | SOCKS5, HTTP, HTTPS | ||
|
Residential, mobile and ISP proxies, 190+ countries down to ZIP, sticky sessions up to 24 hours and residential from $2.20/GB; mobile pool is 250K IPs. Good for this job
Watch out
|
||||||
| 5 | 4.2/5 | $2 / GB | 80% | HTTPS, SOCKS5 | ||
|
Mobile and residential proxies on one plan, 100M+ IPs, sticky sessions up to 7 days and Smart Sessions that keep the IP within the same carrier and area; 38+ countries. Good for this job
Watch out
|
||||||
| 6 | 4.1/5 | $0.63 / GB | 86% | SOCKS5, HTTP, HTTPS | ||
|
Rotating residential and mobile proxies with country, city and ASN targeting, mobile tiers at $2.07–$2.36/GB and rotating traffic from $1.00/GB. Good for this job
Watch out
|
||||||
| 7 | 4.0/5 | ₽490 / day | 83% | SOCKS5, HTTP | ||
|
Mobile proxies in 53 countries with 223 operators and 258 cities, IP rotation every 2–7 seconds on unlimited-traffic plans from 490₽/day. Good for this job
Watch out
|
||||||
| 8 | 3.9/5 | $0.30 / GB | 96% | HTTP, HTTPS, SOCKS5 | ||
|
47M+ residential IPs across 220+ locations at $0.30/GB, with mobile proxies in 20+ countries and Vodafone, T-Mobile and Orange networks. Good for this job
Watch out
|
||||||
| 9 | 3.8/5 | $0.40 / GB | 100% | — | ||
|
175M+ residential IPs across 195 locations and mobile proxies from $7.50/GB at Starter; residential starts at $6/GB, so continuous checks cost more. Good for this job
Watch out
|
||||||
| 10 | 3.7/5 | $0.02 / month | 100% | — | ||
|
80M+ residential IPs across 195 countries, targeting by country, state, city, ZIP and ASN, and rotating residential from $3.50/month; no mobile product is listed. Good for this job
Watch out
|
||||||
| 11 | 3.5/5 | $0.90 / GB | 86% | HTTP, HTTPS, SOCKS5 | ||
|
100M+ residential IPs claimed, 195+ countries with state and city targeting in third-party data, unlimited concurrent sessions, but no mobile proxies in the line-up. Good for this job
Watch out
|
||||||
| 12 | 3.4/5 | $0.06 / GB | 100% | HTTP, HTTPS, SOCKS5 | ||
|
86M+ residential IPs across 200+ countries with country, state and city targeting, HTTP/HTTPS/SOCKS5 and residential pricing from $0.5/GB; no mobile product is listed. Good for this job
Watch out
|
||||||
| 13 | 3.3/5 | — | 81% | SOCKS5, HTTP, HTTPS | ||
|
80M+ residential IPs across 195 countries with country, city, ISP and ASN targeting, but no free trial or refunds and mobile is not sold as a separate product. Good for this job
Watch out
|
||||||
| 14 | 3.1/5 | $2 / GB | 88% | SOCKS4, SOCKS5, HTTP, HTTPS | ||
|
20M+ residential IPs in 190 countries with targeting to ZIP and ASN and sticky IPs up to 72 hours, but mobile proxies cover Poland and Ukraine only. Good for this job
Watch out
|
||||||
| 15 | 2.8/5 | $2 | 82% | HTTP/S | ||
|
Rotating residential and mobile proxies with 60-minute and 30-minute sticky sessions and private IP allocation, but geography is country or US-state level and pricing starts at $45/plan. Good for this job
Watch out
|
||||||
| 16 | 2.4/5 | $5 / proxy | 85% | — | ||
|
Mobile, residential and corporate proxies at $15 and $5 per proxy/month with 150+ locations and 7M+ IPs, but no city targeting or session duration is published. Good for this job
Watch out
|
||||||
| 17 | 1.4/5 | $1 | 82% | — | ||
|
Shared and dedicated HTTP/SOCKS5 proxies priced from $1.00, with shared IPs, blocked torrents and PayPal, and no residential pool size, city or carrier targeting published. Good for this job
Watch out
|
||||||
Ordered by how well each provider fits this job, from the figures the providers publish, our review scores and reader votes — see our methodology.
What ad verification actually checks
Ad verification answers one question with a long tail: was the ad that was paid for delivered, to the audience it was bought for, on the page or app it was booked against, and did the click land where the buyer was told it would? Everything below hangs off that sentence.
A single page load produces a set of observations you can then compare with the insertion order:
- Creative match. The slot rendered the right size, the right image or video, the right copy and the right landing URL. Creative substitution and resizing are old tricks and still common.
- Placement. The page or app bundle is the one that was bought. Ad stacking — several creatives piled into one visible slot, with the rest hidden behind it — shows up here, as does a 300×250 creative reported as a 970×250.
- Geo and language. The exit address resolved to the targeted country, region or city, and the site served the matching language and currency variant.
- Device and environment. Desktop, mobile web, in-app, connected TV. A buy for iOS in-app cannot be verified from a desktop browser, and cannot be verified from an Android handset either.
- Viewability and layout. Position in the viewport, how much of the slot was in view, whether it sat below five screens of scroll, whether a sticky footer covered it.
- Consent and data signals. Whether the CMP fired, what the consent string contained, and whether tags passed consent before they passed a user identifier.
- Invalid traffic signals. Datacenter ASN, headless browser fingerprint, missing or duplicated click identifiers, impossible timing between impression and click.
- Landing page and offer. Final URL after redirects, price shown, app store listing, the product actually on sale.
Fraud detection and verification are the same crawl read two ways. The fetch that proves a creative rendered also produces the evidence for an invalid-traffic finding — which is why the address you fetch from decides the quality of both.
Why your office IP cannot do it
An office connection is one address, or a small block, on one network, in one city, behind one NAT. Every check you run from it looks identical: same ASN, same subnet, same browser fingerprint, same consent state, same cookie jar, often the same ten-minute window.
Four things go wrong.
- You see the wrong ad, or none. Ad servers geo-fence, day-part and device-target. From a single city you get that city's campaign, or a house ad, or an empty slot — and an empty slot is not proof of anything.
- You are filtered out before the auction. Datacenter and known-corporate ranges sit on IVT blocklists. Your bid request never reaches the exchange, so the publisher looks clean when the problem is your address.
- Frequency capping fires. Run twenty checks from one IP and the ad stops serving after the third. You then record nineteen non-deliveries that are your own fault.
- State leaks between checks. Shared cookies, cached creatives, an accepted consent banner from last Tuesday. Every check after the first is contaminated by the one before it.
If you only need to confirm that your own tag fires, your own sitemap resolves, or your own pixel returns 200, a plain connection is fine. The moment the question involves a third party's delivery, you need addresses that look like the audience. Confirm what the outside world sees with what is my IP before you record anything.
Which type of IP matches your inventory
Match the address class to the buy, not to the budget.
Datacenter addresses are fast, cheap and honest about what they are. Use them for checks where nobody is pretending to be a consumer: tag firing, tag sequencing, ads.txt and sellers.json retrieval, redirect chain capture at the server level, SSL and certificate checks. Do not use them to judge whether a consumer in Leeds saw a banner, because the ad server will treat them as what they are.
Residential addresses belong to real consumer ISPs and appear as ordinary broadband. They are the right instrument for desktop display, landing page checks, price and offer consistency by city, and brand-safety adjacency. City-level targeting matters more than country-level: a campaign bought for Manchester and verified from London has verified nothing.
Mobile carrier addresses sit behind CGNAT, so thousands of handsets genuinely share one IP. That shared history is the point — it is indistinguishable from a phone, which is the only way to see in-app inventory and carrier-targeted creative as delivered. Mobile is the most expensive class per gigabyte because the traffic is real and metered.
Static ISP or dedicated addresses are for longitudinal monitoring. If you want to watch the same placement twice a day for a month with persistent cookies and a stable identity, a dedicated address in a sticky session beats a fresh one every time.
Whatever you pick, check it before you spend bandwidth: a dead or mislocated proxy produces a clean-looking result that is pure noise. Run candidates through the proxy checker and compare the exit country with the provider's claim. For throwaway tag tests, the free list is enough; do not build a monitoring programme on it.
How to verify a campaign step by step
- Write down the truth first. Pull the insertion order: geos, cities, device split, creative IDs, landing URLs, flight dates, day-parting. Without this you are collecting screenshots with nothing to compare them to.
- Build the sample. One check per city per device per day is a reasonable floor. If a placement matters commercially, check it more often than the reporting interval you are auditing.
- Confirm the exit address through the proxy before every run. Load your IP tool through the tunnel. If it reports the wrong country, stop; the rest of the run is worthless.
- Set the environment to match the buy. Locale, timezone, user agent and screen size should all agree. A German locale on a US timezone in a desktop UA advertises that you are a test.
- Fetch the page, then fetch it again with a clean profile. First-load with no cookies is the honest consumer view; the second pass with a cookie jar shows what returning users get.
- Capture evidence, not impressions. Save the HAR file, a full-page screenshot, the ad request URL, the creative hash and the final landing URL with its status code.
- Compare, then flag. Differences go to a human before they go to the publisher. Every automated alert should carry the run parameters so it can be reproduced.
A server-side fetch with a proxy and a cookie jar looks like this:
curl -x http://user:pass@gate.example.com:8000 -L --max-redirs 10 -c jar.txt -b jar.txt -A 'Mozilla/5.0 (iPhone; CPU iPhone OS 17_0 like Mac OS X)' -w '%{http_code} %{url_effective}\n' -o page.html https://example.com/landing
For anything that renders ads, a browser is required. Chrome takes the proxy at launch, and a throwaway profile keeps state out of your checks:
google-chrome --proxy-server='http://gate.example.com:8000' --user-data-dir=/tmp/vx-de-berlin --lang=de-DE --accept-lang=de-DE,de
For repeatable runs, drive headless Chromium with Playwright or Puppeteer and set the proxy, locale, timezone and geolocation per context, with HAR recording switched on. One context per check, discarded afterwards, is the discipline that keeps results comparable.
Following a redirect chain without breaking it
An ad click is a chain: ad server, DSP, SSP or exchange, verification vendor, tracker, landing page, sometimes an app store. Each hop can set cookies, mint a click ID, append a consent string and add a cache-buster. Break the chain and you have not observed a redirect — you have created a new one.
The rule is simple: one address, one session, one pass. Keep the same sticky session and the same cookie jar for the whole chain. Do not strip the Referer, do not rewrite query parameters, do not decode a tracked URL and re-request it. That last habit is how verification teams accidentally generate billable clicks and, occasionally, invalid-traffic flags against themselves.
Three practical points:
- Rotating mid-chain invalidates the click. The tracker sees a cookie minted on one address presented from another and treats it as fraud. Sticky sessions of ten to thirty minutes are enough for most chains.
- curl -L only follows HTTP redirects. Meta refresh, JavaScript navigation, POST hand-offs and mobile universal links are invisible to it. Those need a real browser, and mobile deep links need a mobile environment.
- Sometimes observe, do not follow. An intercepting proxy such as mitmproxy gives you every hop with headers intact and no alteration to the request:
mitmproxy --mode upstream:http://gate.example.com:8000 --save-stream-file run.flow
Certificate pinning and HTTPS interception will block some hops. Record the blockage as a finding about the chain, not as a failure of the check.
What it costs to verify continuously
Verification is not a one-off purchase, so the unit rate compounds. Three billing shapes matter: per gigabyte, per IP per month, and per port or per day for mobile.
Residential and mobile are billed by bandwidth, and ad pages are heavy. A single load with a video player, several tags and a consent wall can pull a lot of megabytes, and a programme of one check per city per device per hour multiplies that quickly. Mobile ports cost more per unit than residential because the underlying traffic is genuinely metered. Datacenter and static ISP addresses are billed per address, which makes them cheap for high-frequency tag checks and useless for consumer-view checks.
Budget for the parts nobody quotes: engineering time to keep the harness running, storage for screenshots and HAR files, and human review of every flag. An unresourced review queue turns a good detection system into a source of false accusations.
Keep the cost down by matching frequency to risk. Check the placements that carry the most spend most often; check static, low-risk placements daily rather than hourly. Run datacenter checks for tag and redirect integrity, and reserve residential and mobile bandwidth for the checks that need a human-looking address. If you are choosing a paid provider rather than building your own pool, the provider reviews cover pricing structure and session behaviour; judge them on exit consistency and sticky-session length, not on headline pool size.
Mistakes that produce a false fraud report
A false report is worse than no report. It damages a publisher relationship, can breach a contract, and in the wrong wording invites a defamation claim. Most false positives come from the harness, not the publisher.
- Device mismatch. Desktop user agent on mobile inventory returns no ad, which gets logged as non-delivery. Set the user agent, viewport and touch behaviour to match the buy.
- Trusting one geolocation source. IP databases disagree at city level. Check the exit address through an independent tool and record which database you used.
- Rotating mid-chain. A broken landing page is read as cloaking. Keep the session sticky from impression to landing page.
- Ignoring consent state. A fresh profile in the EU hits a consent wall, no ad renders, and the slot is reported empty. Decide whether you are testing the consented or unconsented experience, and say so.
- Ad blockers and tracking protection. One extension in the profile removes the ad request and half the tags.
- Day-parting and timezone drift. A check at 02:00 local is outside the flight window. Convert deliberately and log UTC alongside local time.
- Caching. Browser cache, service workers and CDN edges all serve yesterday's creative and today's report says creative substitution.
- Reading a lost auction as fraud. No bid, no ad. Frequency capping, budget exhaustion and audience mismatch all produce blank slots that have nothing to do with invalid traffic.
- Single-check conclusions. One datapoint is an anecdote. Require repeat observations before anything is escalated.
- Missing evidence. A claim without a screenshot, timestamp, exit IP and HAR file cannot be defended and will not be acted on.
On legality: checking how your own or a client's campaign is delivered is ordinary commercial monitoring. It becomes something else if you forge clicks, bypass consent walls, harvest personal data or breach a site's terms of service. Mobile and residential addresses should come from pools whose device owners consented; ask the provider how the pool is sourced before you route a client's brand through it.
What to look for
- How ordinary the address looks
- Ad fraud checks are themselves looking for non-human traffic. A datacenter address is served a different page, or none.
- City and carrier targeting
- Geo-targeted campaigns are the whole point; verifying from the wrong city verifies nothing.
- Mobile and carrier reach
- A large share of display inventory is mobile, and a carrier address is the only way to see it as delivered.
- Session stickiness
- A creative, its landing page and its redirect chain have to be followed from one address or the chain breaks.
- Cost at verification volume
- Verification is continuous rather than one-off, so the unit rate compounds.
Every provider, in order
1. Novada — Mobile carrier city reach
4.8 / 5 for this job · from $0.50 / GB
Novada sits at the top of this list on the mobile line. The provider publishes 2M+ mobile IPs across 120+ carriers, and city-level targeting across 195+ countries. For ad verification that pairing does most of the work: a mobile display creative bought for a named city can only be checked from an address in that city sitting on a carrier network, because anything else is served a different page or no page at all.
The rest of the record fits the same job without strain. The residential pool is quoted at 100M+ IPs, with country counts published rather than implied — 8,426,658 for the United States, 2,838,522 for the United Kingdom, 3,167,312 for Brazil, 1,257,655 for France and 1,645,341 for Canada. Session windows are stated: 1-120 minutes on residential, and 360-minute sticky sessions on rotating ISP, which is long enough to walk a creative through its landing page and redirect chain on one address. Prices start at $0.50/GB for rotating datacenter, $0.65/GB for rotating ISP, $1.3/GB for residential and $1.5/GB for mobile. Mobile is the dearest bucket and verification volume on mobile is rarely small, so price that line before anything else.
The awkward part is the success figure. The headline is 99.99%, but the provider quotes 99% for rotating datacenter, so the big number does not clearly apply to every pool on the account. Unlimited Proxies, IPv6 Proxies and Datasets carry no published price and go through an account manager. The free trial covers the Scraper API only; the general offer is a $10 credit, which is a different thing.
Key features
- 2M+ mobile IPs across 120+ carriers, per the provider
- City-level targeting across 195+ countries
- Residential sessions 1-120 minutes; rotating ISP sticky sessions 360 minutes
- Rotating datacenter from $0.50/GB; mobile from $1.5/GB
- Published country IP counts for the US, UK, Brazil, France and Canada
- $10 credit without a card; free trial limited to the Scraper API
Best suited for: best suited for campaigns where a mobile creative has to be seen from an address in a named city on a named carrier
2. ProxyEmpire — Rollover mobile verification
4.6 / 5 for this job · from $0.35 / GB
ProxyEmpire takes second because of rollover. Verification is continuous — checks run weekly, sometimes daily — and metered bandwidth that goes unused in a quiet month is money thrown away. This provider's GB-based plans carry rollover, with unused GB that never expire, and that single term changes the arithmetic of a long-running verification budget more than a few cents off the unit rate.
On the mobile side there are 4M+ rotating mobile IPs across 170+ countries, with mobile targeting at country, city and carrier. That is the layer most display inventory needs. Residential targeting goes further — country, region, city, ISP, ASN, OS/TCP and ZIP — and unlimited sessions and concurrent connections are offered for approved use cases. Pricing starts at $4.50/GB on rotating mobile pay as you go, falling to $2.00/GB on the 1000 GB mobile plan, and $3.50/GB on residential pay as you go, down to $1.50/GB at 1000 GB, with custom residential from $0.75/GB.
The catches are specific. Dedicated mobile covers only 5+ countries and 50+ locations, and drops to country, city and carrier rather than finer targeting. Coverage overall is uneven: 62 datacenter countries and 24 static ISP countries against 170+ for residential and mobile. The $1.97, 150 MB trial excludes static ISP and datacenter. Unlimited bandwidth is priced from $200/day with no rollover. The site also quotes its Trustpilot score as both 4.6 and 4.5 from 100+ reviews.
Key features
- Rollover bandwidth on GB-based plans; unused GB never expire
- 4M+ rotating mobile IPs across 170+ countries
- Mobile targeting at country, city and carrier
- Residential targeting down to ZIP, ASN and ISP
- Rotating mobile from $4.50/GB pay as you go; $2.00/GB at 1000 GB
- Dedicated mobile limited to 5+ countries and 50+ locations
Best suited for: best suited for verification programmes that run continuously and cannot predict monthly traffic
3. Decodo (ex Smartproxy) — Sticky city-level sessions
4.5 / 5 for this job · from $0.08 / GB
Smartproxy, now trading as Decodo, earns the sticky city-level line on the strength of its targeting. The residential pool is 115M+ IPs across 195+ locations, and targeting runs down to city, ZIP code and ASN rather than stopping at country. For a campaign bought against a metro area or a postcode, that is the difference between verifying the buy and verifying something adjacent to it.
Sessions are the other half. The provider offers rotating and sticky options, with a sticky IP held for up to 30 minutes — long enough to follow a creative into its landing page and through a redirect chain, short enough that you will not park a long-lived session on it. HTTP(S) and SOCKS5 are both supported, so older clients work. The published figures for residential are a 99.92% success rate and an average response under 0.5s. Residential pricing runs from $2/GB, with monthly tiers from $3.75/GB at 3 GB to $2.75/GB at 100 GB, enterprise at $2.50/GB from 250 GB and $2/GB at 1 TB; mobile starts at $2.25/GB.
The catches need flagging. The 3-day, 100 MB trial and the 14-day money-back option are mutually exclusive, so taking the trial removes refund eligibility. Refunds are first-purchase only, within 14 calendar days, and capped at under 20% of a residential plan's traffic. Pay as you go at $4/GB is wallet-only and cannot be combined with a monthly subscription. The fraud-score comparison the provider publishes does not name the scoring service behind the numbers.
Key features
- 115M+ residential IPs across 195+ locations, per the provider
- Targeting at city, ZIP code and ASN level
- Sticky IPs held for up to 30 minutes; rotating also available
- HTTP(S) and SOCKS5 support
- Residential success rate 99.92%; average response under 0.5s
- Residential from $2/GB; mobile from $2.25/GB
Best suited for: best suited for campaigns verified against a specific city or postcode where the session has to hold through a redirect chain
4. NodeMaven — Long sticky ZIP sessions
4.4 / 5 for this job · from $2.20 / GB
NodeMaven takes fourth, and its argument is session length. Sticky sessions are quoted at up to 24 hours with unlimited concurrent sessions, and geo-targeting is claimed down to ZIP level across 190+ countries and 195+ locations. Where a verification workflow has to keep one address through a sequence of pages — a creative, its landing page, the click tracker in between — a day-long session removes the risk of the address changing halfway and the chain breaking.
The numbers behind that: 30M+ residential IPs and 250K real-device 5G/LTE mobile IPs, with mobile speed quoted under 0.6s and residential speed under 0.2s. The provider states 95% clean IPs, a verified fraud risk under 5/100 on residential and under 10/100 on ISP, and 99.4% business success rates on residential. Pricing starts at $2.20/GB, the lowest monthly plan is 2 GB at $8.5, and traffic rolls over. ISP plans carry unlimited traffic and a 99.9% uptime guarantee.
The problems are cost and access. There is no free trial: the $3.50, 750 MB entry point is a paid demo, and a third-party write-up says the free-trial route is limited to registered companies that pass KYC. A per-GB model at these rates gets expensive quickly once verification runs at volume across several markets. Smaller locations are reported to be uneven in depth, mobile pool size differs between sources — 250K from the provider against 295K+ in third-party records — and refunds are pro-rated rather than full.
Key features
- Sticky sessions up to 24 hours with unlimited concurrent sessions
- 30M+ residential IPs and 250K mobile IPs, per the provider
- Targeting at country and city level, with ZIP claimed
- Traffic rollover on pay-as-you-go and monthly plans
- Pricing from $2.20/GB; lowest monthly plan 2 GB at $8.5
- No free trial; the $3.50, 750 MB option is a paid demo
Best suited for: best suited for workflows that must hold one address across a long sequence of pages in a single city
5. AnyIP.io — Week-long sticky mobile IPs
4.2 / 5 for this job · from $2 / GB
AnyIP.io sits fifth because it sells the one thing most mobile verification setups cannot get: a sticky mobile session measured in days. Sessions run up to 7 days, and Smart Sessions keep the IP inside the same carrier and area for the duration. If a check has to look like one handset browsing over a week rather than a new device every ten minutes, that is the shape of it.
Residential and mobile are on one plan at one price, with only bandwidth metered. There are 100M+ residential and mobile IPs combined, unlimited IP rotation and unlimited concurrent requests, so threads and rotations are not charged separately — useful when a verification sweep fans out across many placements at once. Filtering covers country, ISP, region and precise GPS coordinates. The provider quotes a request success rate of about 98.6% and an average response of about 0.6s. Unused data rolls over each billing cycle. Plans run from Micro at $25 for 5 GB to Enterprise at $1000 for 500 GB, with extra traffic at $5/GB, $4/GB, $3/GB and $2/GB respectively.
The catches matter for verification specifically. Coverage is 38+ countries, which is a short list for multi-market work. The advertised $2/GB is the Enterprise rate, not the entry price. There is no free trial — Micro is the trial and it is paid up front. Crypto payments fall outside the refund policy, which covers card purchases only, and team seats are capped at 2, 5 and 10 before Enterprise.
Key features
- Sticky sessions up to 7 days; Smart Sessions hold the same carrier and area
- Residential and mobile on one plan, metered on bandwidth only
- 100M+ residential and mobile IPs, per the provider
- Unlimited IP rotation and unlimited concurrent requests
- Filtering by country, ISP, region and GPS coordinates
- Plans from $25 for 5 GB to $1000 for 500 GB; extra traffic $5/GB to $2/GB
Best suited for: best suited for mobile checks that need one address held for days rather than minutes
6. Floppydata — Low-cost mobile city checks
4.1 / 5 for this job · from $0.63 / GB
Floppydata is sixth, and the reason it is on the list at all is the unit rate. Rotating traffic starts at $1.00/GB and covers residential, mobile and datacenter on the same meter, with city, ASN and IP-type targeting and 190+ locations. Where a verification job is broad rather than deep — many placements, light pages, short sessions — a dollar a gigabyte is the number that decides whether the programme is affordable.
Mobile at volume has published tiers: 96 GB at $226.56, which works out at $2.36/GB, and 120 GB at $247.80, or $2.07/GB, with custom pricing above 120 GB. The provider states a 95% clean IP rate and a 97% mobile device trust score, which is the figure that matters if mobile creative is what you are checking. There is a 100 MB free trial for each of residential, datacenter and mobile, cards go through Stripe, and 50+ cryptocurrencies are accepted. SOC 2, CCPA and GDPR compliance are published. Static datacenter starts at $0.79 per IP per month.
The awkward parts: there is no phone support, so a live voice escalation is not available. Documentation is thin for advanced setups such as custom rotation logic. The 100 MB per-type trial is a connectivity check, not a load test. The headline performance figures — 99.99% uptime and sub-0.3-second response — are provider claims on a company founded in April 2024, and should be treated as claims rather than as a track record.
Key features
- Rotating traffic from $1.00/GB across residential, mobile and datacenter
- City, ASN and IP-type targeting across 190+ locations
- Mobile at 120 GB for $247.80, or $2.07/GB
- 100 MB free trial per proxy type; 50+ cryptocurrencies accepted
- 95% clean IP rate and 97% mobile device trust score, per the provider
- No phone support; documentation thin for custom rotation logic
Best suited for: best suited for wide, low-cost mobile verification sweeps across many cities rather than deep single-market checks
7. MobileProxy.Space — Carrier-level mobile rotation
4.0 / 5 for this job · from ₽490 / day
MobileProxy.Space sits at seven because the thing it sells is the thing this job has the least substitute for: carrier addresses, rotated on demand. Verifying mobile inventory means the request has to look like a phone on a cellular network, because that is the only way the mobile creative is served at all. Rotation runs on a timer or by hitting an API link, and the provider states an IP change time of 2–7 seconds, quick enough to walk a redirect chain from a fresh address without stalling the run.
Billing is the second argument. Mobile plans are charged by rental period with unlimited traffic, and the minimum rental is one day. For continuous verification that matters more than it sounds: a per-gigabyte meter punishes you for following every redirect and every landing page on every creative, while a fixed period does not. Each mobile proxy takes up to 250 simultaneous connections and backconnect ports take up to 5. Mobile starts at 490₽/day, and dedicated datacenter and residential lines sit under the same account.
The catch is that the numbers do not line up. The third-party listing gives 30+ countries, 130+ mobile operators and a 2M+ pool; the provider's own pages give 53 countries, 223 operators and 267,370,000 addresses, with 700+ against 1,356 servers. Trial length is two hours in one record and three in another. Third-party write-ups flag the prices as relatively high, and there are reports of laggy proxies, late-night maintenance windows and problems with geographical locations. Declared performance is not fast either: average ping 348 ms, response time 3.24 seconds, and up to about 4 Mb/s when the device is remote. Good for confirming which creative served; wrong for anything where timing is part of the measurement.
Key features
- Mobile proxies with IP rotation on a timer or by API link
- Minimum rental of one day; mobile plans bill by period with unlimited traffic
- IP change time stated at 2–7 seconds
- Up to 250 simultaneous connections per mobile proxy, 5 per backconnect port
- Dedicated datacenter and residential lines on the same account
- Mobile from 490₽/day
Best suited for: checking mobile creative and carrier-served inventory where the address has to look like a handset rather than a server
8. Proxy-Seller — Cheap residential verification volume
3.9 / 5 for this job · from $0.30 / GB
Proxy-Seller sits eighth because it is the cheapest way on this page to buy residential volume, and ad verification is a volume business. Checks run continuously rather than once: every creative, every market, every hour, which means the per-gigabyte rate compounds instead of sitting still. Residential starts at $0.30/GB against a pool the provider declares at 47M+ rotating addresses across 220+ locations, and mobile, ISP, IPv4, IPv6 and datacenter lines sit under the same account with no monthly lock-ins stated.
The rest of the record is largely about procurement rather than performance. ISO/IEC 27001:2022 and ISO 9001:2015 certification, GDPR, CCPA and ePrivacy compliance stated, DPA/SCC readiness and an EWDCI trust seal — the paperwork that decides whether a verification vendor gets approved at all. Technically it is HTTP, HTTPS and SOCKS5 with a RESTful API, IP whitelisting and SDK access. The stated figures are a 99.5%+ connection success rate, 99.7% uptime and roughly 0.35s residential response time.
The catch is mobile. Coverage reaches only 20+ countries against 220+ residential locations, with Vodafone, T-Mobile and Orange named and 5G/4G/LTE supported. If the verification plan leans on mobile inventory across more than a handful of markets, that footprint is thin, and mobile is priced from $10/IP rather than by the gigabyte. The second catch is that the uptime, success rate and response time are the provider's own, with no independent test cited alongside them. Treat them as a specification, not a result.
Key features
- Residential from $0.30/GB, mobile from $10/IP, ISP from $0.98/IP
- 47M+ rotating residential IPs across 220+ locations, provider-declared
- Mobile coverage reaches 20+ countries; carriers named are Vodafone, T-Mobile and Orange
- HTTP, HTTPS and SOCKS5 with RESTful API, IP whitelisting and SDK
- ISO/IEC 27001:2022 and ISO 9001:2015 certified; GDPR, CCPA and ePrivacy compliance stated
- Connection success rate stated at 99.5%+ and 99.7% uptime, both provider-declared
Best suited for: running continuous desktop and residential ad checks at high volume where the per-gigabyte rate decides the budget
9. Oxylabs.io — Enterprise mobile reach
3.8 / 5 for this job · from $0.40 / GB
Oxylabs sits ninth on price rather than capability. At Starter, residential is $6/GB and mobile $7.5/GB, which for a job that runs continuously is the most expensive rate in this group. The rates do fall with commitment — residential to $5/GB, then $4/GB, then $2.5/GB at enterprise; mobile to $6.67, $5 and $3.5/GB — but enterprise starts at $2,500/month through sales. So the argument for it here is breadth, and the reader needs to know the entry rate before believing the pool figure.
The breadth is real. The provider claims 177M+ addresses in total, 175M+ of them residential, across 195 locations, and one subscription carries the whole catalogue. Unit rates are published for every product at every tier, so a verification budget can be modelled before anyone speaks to sales. For a process with a compliance gate, the paperwork is the differentiator: ISO/IEC 27001:2022 across Datacenter Proxies, Residential Proxies and Web API; SOC 2 Type II on the Scrape endpoint and Web Unblocker; Technology E&O and cyber insurance across all products; GDPR compliance and EU AI Act alignment stated.
The catches are cost and scope. Five dollars of welcome credits is under a gigabyte of residential traffic at Starter rates, and the cheapest paid plan is $30/month. VAT may apply on Starter, Basic and Flexible, so the headline rate is not always the invoice, and Bitcoin is not accepted. More subtly, the certifications are per product rather than company-wide — the ISO and SOC 2 each name specific products, not everything in the catalogue — so a procurement team checking scope will find gaps where the marketing implies none.
Key features
- 175M+ residential IPs within a stated 177M+ total, across 195 locations
- Mobile proxies from $7.5/GB at Starter down to $3.5/GB at enterprise
- Residential from $6/GB at Starter to $2.5/GB at enterprise; enterprise from $2,500/month
- ISO/IEC 27001:2022 covers three named products; SOC 2 Type II covers two endpoints
- GDPR compliance and EU AI Act alignment stated, with Technology E&O and cyber insurance across all products
- $5 in welcome credits; cheapest paid plan $30/month; no Bitcoin
Best suited for: cross-market mobile and residential verification inside a procurement process that needs certifications and insurance on file
10. Webshare.io — City-level residential checks
3.7 / 5 for this job · from $0.02 / month
Webshare sits tenth because it is the cheapest way here to check one specific city, and city is where geo-targeted verification usually fails. A campaign bought for a single metro checked from the right country and the wrong end of it verifies nothing. Rotating residential targets country, state, city, ZIP code and ASN, on a product starting at $3.50/month, with static residential from $6.00 and proxy servers from $2.99. Per-IP pricing falls from $0.0299 to $0.0179 as volume grows and yearly billing saves a further 30%, so a wide but shallow schedule is affordable.
What makes it practical for this job is the plumbing. HTTP and SOCKS5 endpoints sit on the same credentials, IP authorisation is offered as an alternative to username and password, there is a Chrome extension for people who would rather not script, and an API with documentation for those who would. Published country counts let you judge whether a market has depth before committing: 6,652,539 addresses in the United States, 1,573,702 in the United Kingdom, 1,547,939 in Germany, 1,174,275 in France, 830,343 in Spain and 323,485 in Canada.
The catches are in the small print. Unlimited bandwidth carries an asterisk pointing to a Fair Usage Policy, so read it before buying on the word unlimited. The provider's own pages give both 195 countries and 50+ locations, so the real footprint is hard to pin down. The static residential ISP list names Comcast and AT&T and then says "& more", which means a specific carrier cannot be checked before purchase. The 99.7% success rate and 99.97% uptime are the provider's, with nothing independent alongside. And the permanent free plan of 10 proxies is listed under the datacenter product, so it tests the workflow, not the residential pool.
Key features
- Rotating residential targets country, state, city, ZIP code and ASN
- Residential from $3.50/month, static residential from $6.00, proxy servers from $2.99
- Per-IP pricing falls from $0.0299 to $0.0179 with volume; yearly billing saves 30%
- Permanent free plan of 10 proxies with no card required, listed under the datacenter product
- HTTP and SOCKS5 on the same credentials, with IP authorisation and a Chrome extension
- Published country IP counts, from 6,652,539 in the United States to 323,485 in Canada
Best suited for: checking locally targeted creative in a named city or ZIP code without paying enterprise rates to do it
11. FlyProxy — Unlimited residential concurrency
3.5 / 5 for this job · from $0.90 / GB
FlyProxy sits eleventh because it is cheap only at the very top of its range, and because the thing it handles better than most is concurrency. Unlimited concurrent sessions are stated by both the provider and a third-party review. For verification at volume that is often the binding constraint rather than bandwidth: thread caps mean queueing checks you already have the budget to run, which stretches a schedule that is supposed to be continuous.
Residential pricing runs at $2.50/GB for 20 GB, $2.40/GB for 50 GB and $0.90/GB at the 10,240 GB tier, with targeting down to state and city. The third-party country tests are the useful part of the record: 98.7% in Germany at 0.52s, 98.4% in the United States at 0.65s, 98.1% in Canada at 0.72s, 97.9% in France at 0.75s, 97.2% in Brazil at 0.80s and 96.8% in Mexico at 1.00s. IP lifetime is given as around 24 hours, which suits following a creative, its landing page and its redirect chain from one address better than per-request rotation does. Nine antidetect browser integrations are listed for anyone driving this from a browser profile.
The catches start with mobile: there are no mobile proxies in the line-up, flagged as a weakness by third-party reviewers, and that is a genuine hole when a large share of display inventory is mobile. There are no refunds once products are sold, in the provider's own words. The third-party listing says there is no free trial while the provider advertises 500M of free traffic through sales. Pool size disagrees — over 100 million by the provider, 69 million by the third party — and so does the protocol list. At 20 GB the rate is $2.50/GB, which is more than the enterprise rate elsewhere on this page; the low figure needs the 10,240 GB commitment.
Key features
- Unlimited concurrent sessions, stated by both the provider and a third-party review
- $2.50/GB at 20 GB, $2.40/GB at 50 GB and $0.90/GB at the 10,240 GB tier
- Country, state and city targeting
- Third-party country tests from 96.8% (Mexico) to 98.7% (Germany), with responses of 0.52s to 1.00s
- No mobile proxies in the line-up
- No refunds once products are sold, per the provider
Best suited for: high-concurrency residential checks where session limits rather than bandwidth are what stops the schedule scaling
12. ProxySale.com — Low-cost city targeting
3.4 / 5 for this job · from $0.06 / GB
Proxy-Sale.com sits twelfth because it is the lowest-cost entry here for city-level residential checks, and the record explains why the position is not higher. The provider declares 86M+ residential addresses across 200+ countries, with country, state and city targeting on the residential product and HTTP, HTTPS and SOCKS5 support. Residential is billed by usage from $0.5/GB, so a continuous verification schedule scales with what it consumes rather than with a seat or a licence.
The rest of the range covers the awkward cases. An unlimited residential plan with no cap on IPs or traffic runs at $69.68/day with random and sticky rotation, which suits a burst of checks across many cities on one day. Static residential is $5/IP and static datacenter $3/IP for anything that needs to keep the same address. The Web Scraper API bills only for successfully delivered results, and there is a 30-day zero-setup trial on that product. Email, live chat and 24/7 account managers are listed for support.
The catches are mostly about evidence. The $0.06/GB on the price list belongs to the Video Downloader API, not the residential product, where the entry rate is $0.5/GB. Uptime at 99.9%, a 99.5% average success rate and a sub-0.6s response time are all provider-declared with no third-party audit shown — more than a rounding detail when the job is checking whether traffic is what it claims to be. The free trial covers the Web Scraper API only, not the proxy network. The service is not available in mainland China by the provider's own policy. And the unlimited plan is billed per day, which is awkward for a long-running job. No mobile product appears in the record at all, which leaves a large share of display inventory out of reach.
Key features
- 86M+ residential IPs across 200+ countries, provider-declared
- Country, state and city targeting on the residential product
- Residential billed by usage from $0.5/GB; unlimited residential at $69.68/day
- Static residential $5/IP and static datacenter $3/IP
- HTTP, HTTPS and SOCKS5 on the residential product
- 99.9% uptime, 99.5% success rate and sub-0.6s response time, all provider-declared
Best suited for: cheap city-level residential spot checks where a provider-declared success rate is acceptable evidence
13. Nsocks — Geo-targeted residential testing
3.3 / 5 for this job
Nsocks sits thirteenth because its case rests on city-level targeting rather than on anything specific to advertising. The provider lists country and city targeting, and third-party write-ups report ISP and ASN filters as well; for a campaign bought against a named city or a named network, that filter is what decides whether a check means anything at all. A residential address served from the right city sees the creative the campaign actually delivers.
The product range is split rather than bundled: rotating residential, static residential, datacenter, ISP and unlimited plans are each billed separately, so verification traffic can sit on residential while other work runs elsewhere. Protocols are SOCKS5, HTTP and HTTPS, the dashboard carries traffic statistics, and integrations are listed for most anti-detect browsers alongside Playwright, Selenium and an API. Nine interface languages and 24/7 support by live chat and email are on the record, and 99.95% uptime is the provider's own claim.
The catch is thin evidence and no safety net. There is no free trial, so pool quality cannot be examined before paying, and the provider states it does not support refunds on products sold. Its own home page quotes two different pool sizes, 80M+ and 90M+, which is the kind of detail that makes every other figure on the page worth treating as a claim rather than a measurement. Third-party records put the founding date at 26 August 2024, so the operating history is short.
Key features
- 80M+ real-user residential IPs claimed by the provider
- Country and city targeting, with ISP and ASN targeting reported by third parties
- Rotating, static, datacenter, ISP and unlimited proxies sold as separate products
- SOCKS5, HTTP and HTTPS, with Playwright, Selenium and API integrations listed
- No free trial and no refunds on products sold
- Provider home page quotes both 80M+ and 90M+ pool sizes
Best suited for: best suited for verifying campaigns bought against a specific city or network, where the address has to match the buy.
14. GonzoProxy — Sticky residential ZIP targeting
3.1 / 5 for this job · from $2 / GB
GonzoProxy earns fourteenth on granularity. Targeting reaches ZIP code and ASN as well as country, city, state and continent, across 190+ countries, and a 20M+ pool of real-user addresses. Verification that stops at city level leaves the awkward cases unresolved: a campaign sold against a handful of postcodes, or against a particular network, needs an address from inside that boundary or the impression being checked is not the impression being bought.
Sticky sessions are the other half of the argument. The provider claims sticky IPs held for up to 72 hours, long enough to follow a creative through its landing page and its redirect chain from a single address rather than restarting the chain at every hop. Billing is per gigabyte, from $2/GB at the highest volume tier, traffic never expires and rolls over to the next plan, and concurrent connections are unlimited on every plan. Access is stated as live in about 30 seconds with no KYC, and payment is by card or crypto with 0% fees on crypto per the provider.
What it will not do is mobile inventory outside two markets: mobile proxies cover Poland and Ukraine only. For display work running across several countries that is a real gap, and mobile delivery has to be verified elsewhere. The dashboard lacks detailed per-request logs, which matters when a client asks which request went where. The demo is not free — $1.99 for 0.3 GB, with a separate 1 GB trial at $4.50 — and two published residential ladders quote different per-GB rates for the same volumes, so the rate you plan against depends on the page you read. Published speed and success figures are the provider's own and are not independently verified on its site.
Key features
- Country, city, state, continent, ZIP code and ASN targeting
- Sticky IPs claimed up to 72 hours
- Pay-per-GB residential billing from $2/GB at the highest tier; traffic never expires and rolls over
- Unlimited concurrent connections on every plan
- Mobile proxies cover Poland and Ukraine only
- Two published residential plan ladders quote different per-GB rates for the same volumes
Best suited for: best suited for checking campaigns sold against ZIP codes, states or particular networks, where one address has to hold a redirect chain together.
15. Live Proxies — Private sticky mobile IPs
2.8 / 5 for this job · from $2
Live Proxies is here for the mobile half of the job. Rotating mobile IPs are sold with private allocation — the provider states its IPs are unique and shared only with you — and that matters in verification because a shared address arrives carrying whatever the previous user did with it. The sticky window is 30 minutes on rotating mobile and 60 minutes on rotating residential, enough to hold a creative, its landing page and its redirect chain together on one address.
Concurrency is not capped: the provider states no concurrency limit, unlimited threads and unlimited requests, which suits continuous checking rather than a one-off sweep. Unused bandwidth rolls over to the standby plan, and dashboard access, API access and IP whitelisting are included. The B2B tier is quoted on a 10M+ IP pool with tailored plans, with the enterprise listing at $2K per month. Metered and unmetered billing are both offered, and payment runs through Stripe with cards, Apple Pay and Google Pay.
The catches are commercial and geographic. The home page prices static residential at $65.00 per plan while the pricing page marks it permanently unavailable, and the home page starting prices of $45.00, $65.00 and $55.00 do not reconcile with the pricing page's B2C plans starting from $70. The free trial goes to custom plan users after an agreement; B2C buyers are told to buy a small plan instead. SOCKS5 is listed for the B2B tier only. Geography is described as available countries or state-level options with no count given, and state IPs are US only, so verification against a named city elsewhere is guesswork.
Key features
- Private IP allocation, with IPs stated as unique and shared only with you
- Sticky sessions of 30 minutes on rotating mobile and 60 minutes on rotating residential
- No concurrency limit, unlimited threads and unlimited requests
- Unused bandwidth rolls over to the standby plan
- SOCKS5 listed for the B2B tier only
- State-level IPs are US only, with no country count published
Best suited for: best suited for mobile creative checks that need a private address held for the length of a redirect chain.
16. ASocks.com — Per-proxy mobile budgeting
2.4 / 5 for this job · from $5 / proxy
ASocks is sixteenth because it prices per proxy per month rather than per gigabyte, which changes the arithmetic of continuous verification. Mobile is $15 per proxy per month; residential and corporate are $5. Volume discounts start at 10 proxies with 5%, 50 with 10% and 100 with 15%. If the job is a fixed panel of locations checked on a schedule, a flat per-proxy cost is predictable in a way metered traffic is not: a heavy creative cannot run up a bill, because there is no bill to run up.
The provider claims 150+ locations, 7M+ IP addresses, a 99.7% connection success rate and 100,000+ clients, and says it runs its own proxy infrastructure, which it links to cleaner addresses and fewer bans. Three product lines are sold — mobile, residential and corporate — and support is 24/7 with a stated average response time under 10 minutes.
The awkward part is that every headline figure here is provider-reported and unaudited. The success rate, the location count, the pool size and the client count all come from ASocks itself, with no independent test on the record. The per-proxy model cuts both ways as well: you pay for the proxy whether or not the campaign you are checking has traffic in that city this week, and mobile at $15 costs three times the residential and corporate rate, so a panel built mostly on carrier addresses is the expensive version of this provider.
Key features
- Per-proxy monthly pricing: mobile $15, residential and corporate $5
- Volume discounts of 5% from 10 proxies, 10% from 50 and 15% from 100
- 150+ locations and 7M+ IP addresses, as claimed by the provider
- 99.7% connection success rate, provider-reported and unaudited
- 24/7 support with a stated average response time under 10 minutes
- Own proxy infrastructure, per the provider
Best suited for: best suited for a fixed panel of locations checked on a schedule, where a flat per-proxy cost beats metered traffic.
17. ChangeMyIP.com — Shared datacenter proxy testing
1.4 / 5 for this job · from $1
ChangeMyIP sits last because a datacenter address is the thing ad verification is trying to get away from. Ad servers and fraud checks treat hosting ranges differently, and a shared datacenter IP may be served a different creative, a stripped page or nothing at all, so this is not the tool for confirming what a viewer saw. What it is good for is the plumbing around the job: confirming a tag fires, that a redirect resolves, that a protocol-specific client is configured correctly, before the same check runs from a residential address.
Public per-unit pricing makes that cheap to try. Shared HTTP and shared SOCKS5 are $1.00, dedicated HTTP $1.80, SOCKS5 and ShadowSocks $2.20, DNS proxy $2.00, Trojan $3.00 and Wireguard VPN $4.00, with unlimited bandwidth on the plans it sells, 1Gb/s ports and no contract. Protocols span HTTP, SOCKS5, ShadowSocks, Trojan, DNS proxy and Wireguard, and activation is stated as instant. Crypto is accepted through Zonda alongside cards and PayPal, and the provider says it has been selling HTTP proxies and SOCKS5 for more than twelve years.
The limits are spelled out by the provider itself. Shared proxies put multiple users on the same IP and share bandwidth, and performance can suffer on busy servers. Torrents and paypal.com are blocked at proxy level on every product. A $10 minimum order applies even for a single $1.00 shared proxy, the 30-day money-back guarantee does not apply once the service has been used and covers bugs or slow performance only, and the fully anonymous and undetectable wording has no published independent test behind it.
Key features
- Shared HTTP and shared SOCKS5 proxies at $1.00 per unit
- Dedicated HTTP $1.80; SOCKS5 and ShadowSocks $2.20; DNS proxy $2.00; Trojan $3.00; Wireguard VPN $4.00
- Protocols: HTTP, SOCKS5, ShadowSocks, Trojan, DNS proxy and Wireguard
- Unlimited bandwidth, 1Gb/s ports, instant activation, no contract
- Shared IPs and shared bandwidth; torrents and paypal.com blocked at proxy level
- $10 minimum order, and the 30-day guarantee ends once the service has been used
Best suited for: best suited for pre-flight checks of tags, redirects and protocol configuration rather than for confirming what a real viewer was served.
Ad verification proxies: questions we get
Do I need a proxy to check my own ads?
If the ad is geo-targeted, device-targeted or served through an exchange, yes. From a single office address you will see one city's creative, a house ad, or nothing. A plain connection is only enough for confirming that your own tag or pixel responds.
What is the difference between ad verification and ad fraud detection?
They are two readings of the same observation. Verification compares what rendered against what was booked; fraud detection asks whether the impression could have come from a human at all. A single crawl supplies the creative, the placement, the redirect chain and the traffic signals for both.
Can I use a VPN instead of a proxy?
A VPN gives you one exit at a time and usually a datacenter range, which is exactly the kind of address ad servers filter. It also tends to tunnel the whole machine, so concurrent checks in different geos are impossible. A proxy pool gives you per-session addresses you can label and reproduce.
How many locations do I need to check?
Enough to cover every city where the buy is differentiated. Start with one check per city per device per day and increase frequency on the placements that carry the most spend. Adding countries you did not buy in adds cost without adding evidence.
Why does the ad show in one city and not another?
Either the campaign is geo-fenced more tightly than the insertion order says, or the address you used did not resolve where you thought. Check the exit IP through an independent tool before concluding anything about the publisher.
Are residential proxies legal for ad verification?
Monitoring how a campaign is delivered is ordinary commercial activity in most jurisdictions. The legal risk sits in how the pool was sourced, whether consent was collected from device owners, and whether your activity breaches a site's terms. Ask providers for their sourcing model in writing.
What is a sticky session and how long should it last?
It is a single address held for the duration of a task instead of rotating per request. Ten to thirty minutes covers most ad redirect chains, including a landing page and an app store hop. Rotating mid-chain breaks the click identity and can look like fraud to a tracker.
How do I check ads inside a mobile app?
You need a mobile carrier address and a mobile environment, because in-app inventory is rarely served to desktop requests. Emulation gets you part of the way; for deep links and store hand-offs you need a real handset or a device farm behind a carrier connection.
How do I spot domain spoofing?
Compare the domain that actually served the impression with the one in the insertion order, and cross-check it against the seller records the publisher declares. Spoofing usually shows up as a legitimate-looking subdomain, a parked page or a resold domain that never appears in the seller file.
What should I save as evidence when I find a problem?
The HAR file, a full-page screenshot, the exit IP and its geolocation source, the timestamp in UTC and local time, the ad request URL, the creative hash and the final landing URL with status code. Without those, a finding cannot be reproduced or defended.
Why do I get a consent wall instead of the ad?
A fresh browser profile has no consent state, so the CMP blocks the tags that would have requested the ad. Decide whether you are measuring the consented or unconsented experience, record which, and keep the choice identical across checks.
Can I verify video ads the same way?
The method is the same but the bandwidth is much higher, since a player will pull far more data than a display slot. Use sticky sessions to keep the player and its tracker on one address, and budget mobile and residential traffic accordingly.
Other jobs people buy proxies for
- Web scraping
- SEO and rank tracking
- Price monitoring
- Multi-account management
- Antidetect browsers
Some links on this page are affiliate links: if you buy through them we may earn a commission, at no cost to you. It does not change the order above — see our monetisation policy.