The short answer
- 1 Webshare.io Best postcode-level targeting Visit site
- 2 Decodo (ex Smartproxy) Best all-round retail pool Visit site
- 3 Novada Best cheap city-level pool Visit site
17 best proxies for price monitoring, compared
| # | Provider | Fit | From | Readers | Protocols | Links |
|---|---|---|---|---|---|---|
| 1 | 4.7/5 | $0.02 / month | 100% | — | ||
|
Rotating residential targets country, state, city, ZIP code and ASN, with 80M+ IPs across 195 countries and a permanent free plan of 10 proxies. Good for this job
Watch out
|
||||||
| 2 | 4.6/5 | $0.08 / GB | 100% | HTTP(S), SOCKS5 | ||
|
115M+ residential IPs across 195+ locations, targeting down to city, ZIP and ASN, sticky sessions up to 30 minutes and a declared 99.92% residential success rate. Good for this job
Watch out
|
||||||
| 3 | 4.5/5 | $0.50 / GB | 88% | HTTP(S) | ||
|
100M+ residential IPs with 195+ countries and city-level precision, residential from $1.3/GB and rotating ISP from $0.65/GB, plus 6-hour sticky ISP sessions. Good for this job
Watch out
|
||||||
| 4 | 4.4/5 | $2.20 / GB | 88% | SOCKS5, HTTP, HTTPS | ||
|
30M+ residential IPs with ZIP-level targeting claimed, sticky sessions up to 24 hours, traffic rollover and residential from $2.20/GB. Good for this job
Watch out
|
||||||
| 5 | 4.3/5 | $0.40 / GB | 100% | — | ||
|
Market price monitoring is a listed use, with 175M+ residential IPs across 195 locations; residential is $6/GB on Starter and Web Unblocker $7.2/GB. Good for this job
Watch out
|
||||||
| 6 | 4.2/5 | $0.35 / GB | 90% | HTTP, HTTPS, SOCKS5, HTTP/2 | ||
|
30M+ residential IPs across 170+ countries with ZIP targeting and rollover bandwidth on GB plans; Pay As You Go residential is $3.50/GB (list $7). Good for this job
Watch out
|
||||||
| 7 | 4.1/5 | $0.63 / GB | 86% | SOCKS5, HTTP, HTTPS | ||
|
Rotating residential from $1.00/GB with country, city and ASN targeting across 190+ locations, plus 20% off at 96 GB and 30% off at 120 GB. Good for this job
Watch out
|
||||||
| 8 | 4.0/5 | $0.30 / GB | 96% | HTTP, HTTPS, SOCKS5 | ||
|
Residential from $0.30/GB across 47M+ rotating IPs in 220+ locations, with HTTP/HTTPS/SOCKS5 and price monitoring listed among its stated uses. Good for this job
Watch out
|
||||||
| 9 | 3.8/5 | $0.90 / GB | 86% | HTTP, HTTPS, SOCKS5 | ||
|
100M+ residential IPs across 195+ countries with state and city targeting, unlimited concurrent sessions and a $0.90/GB rate at the 10,240 GB tier. Good for this job
Watch out
|
||||||
| 10 | 3.6/5 | $0.06 / GB | 100% | HTTP, HTTPS, SOCKS5 | ||
|
86M+ residential IPs across 200+ countries with country, state and city targeting, residential billing from $0.5/GB. Good for this job
Watch out
|
||||||
| 11 | 3.4/5 | — | 81% | SOCKS5, HTTP, HTTPS | ||
|
80M+ residential IPs across 195 countries with country and city targeting, but no free trial and no refunds, and no published per-GB prices in the record. Good for this job
Watch out
|
||||||
| 12 | 3.2/5 | $2 / GB | 80% | HTTPS, SOCKS5 | ||
|
100M+ residential and mobile IPs with sticky sessions up to 7 days and unused data rollover, but coverage is only 38+ countries. Good for this job
Watch out
|
||||||
| 13 | 3.1/5 | $2 / GB | 88% | SOCKS4, SOCKS5, HTTP, HTTPS | ||
|
20M+ residential IPs across 190 countries with ZIP and ASN targeting, traffic that never expires and sticky IPs claimed up to 72 hours. Good for this job
Watch out
|
||||||
| 14 | 3.0/5 | $2 | 82% | HTTP/S | ||
|
Private IP allocation with no concurrency limit, unlimited threads and 60-minute sticky sessions on rotating residential; geography is state-level and US-only for state IPs. Good for this job
Watch out
|
||||||
| 15 | 2.7/5 | $5 / proxy | 85% | — | ||
|
Residential is sold at $5 per proxy per month, not per GB, across 7M+ IPs in 150+ locations, with volume discounts from 10 proxies. Good for this job
Watch out
|
||||||
| 16 | 2.2/5 | ₽490 / day | 83% | SOCKS5, HTTP | ||
|
Mobile plans bill by rental period with unlimited traffic across 53 countries on the provider's figure, but there is no explicit city or postcode residential targeting. Good for this job
Watch out
|
||||||
| 17 | 1.8/5 | $1 | 82% | — | ||
|
Shared proxies put multiple users on the same IP, performance can suffer on busy servers, and a $10 minimum order applies; torrents and paypal.com are blocked. 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.
Why prices differ by address
A retail price is not a property of a product. It is a function of the address the site believes you are at, the session you are carrying and the moment you ask. Change any of the three and the number changes with it. That is why a spreadsheet filled from one office connection disagrees with what a customer in another region actually pays.
Four mechanisms do most of the work. A GeoIP lookup decides which country storefront, catalogue and currency you are served. A delivery postcode decides which warehouse fulfils the order, which sets delivery cost, the delivery estimate and often the price itself. Local tax rules change the tax-inclusive figure without touching the shelf price. And the retailer's own tests — A/B pricing, member-only offers, marketplace seller rotation in the buy box — change what appears even when address and session stay fixed.
None of that is visible from a single connection. One IP address means one region's price, one currency, one warehouse's stock, and a rate ceiling that arrives after a few hundred requests. The result is a dataset that looks complete and is not: it is a record of what one address saw.
What a monitoring job needs is the ability to present many addresses, at known locations, repeatedly, without being blocked. That is the whole job of the proxy layer. Everything else — parsing, storage, alerting — is ordinary engineering.
Which type of IP a retailer trusts
Large retailers sit behind bot management that is stricter than almost anything outside search and social. The checks are layered: the ASN and reputation of the exit address, the TLS and HTTP/2 fingerprint of the client, header order, cookie history, request pace and whether a JavaScript challenge was solved. A proxy only fixes the first layer. If the rest of the request looks like a script, the address will not save you.
- Datacentre addresses are the cheapest per gigabyte and the easiest to identify. Large retailers challenge them quickly, sometimes on the first request. They remain useful for catalogue discovery, sitemap walks and smaller shops with lighter defences.
- Residential addresses belong to consumer ISPs, so a request from one looks like a shopper at home. They cost more, usually priced per gigabyte, and hold up far better on the big sites.
- Mobile addresses sit behind carrier-grade NAT, where thousands of real people share one address. Blocking them is expensive for the retailer, so they survive longest. They are also the most expensive per gigabyte.
- Static residential or ISP-registered addresses sit between the two: hosted in a data centre but registered to an ISP. That compromise is useful when you need the same address for days.
Rotation policy matters as much as type. Catalogue pages can rotate freely. A basket, a login or a postcode selection cannot: the moment the address changes, the session, the basket and the delivery estimate go with it. A serious setup needs both a rotating pool and sticky sessions that hold one address for the length of a checkout-shaped flow.
The awkward part: residential and mobile pools are assembled from real users' devices, and sourcing standards are not uniform across the market. Ask how a provider recruits and consents its peers before you build a dependency on them. Separately, reading a public price is usually lawful; breaching a site's terms of service is a contract matter, not a criminal one, and the two get confused constantly. Neither is a licence to hammer a site, ignore the paths its robots file asks you to leave alone, or republish a competitor's content wholesale.
Postcode targeting and why it decides the number
Targeting granularity is where most price-monitoring projects quietly fail. Providers advertise different levels — country, city or region, sometimes postcode, sometimes ASN — and the labels do not mean the same thing between them. A "city" target may be a genuine ISP allocation inside the city boundary, or a block registered to the metro area twenty miles out. If delivery cost depends on the postcode, that gap shows up as a wrong number in your table.
Postcode-level targeting is rare and the pools at that level are thin, so do not build a plan that assumes it. The reliable pattern has two legs:
- Use the proxy to reach the right catalogue. Target the country, or the city if the retailer runs regional pricing. This gets you past geo-redirects and into the right currency and storefront.
- Use the site's own postcode control to set the address. Retailers expose this openly — a "deliver to" field, a store selector, a location banner. Set it inside the session, then read the price, delivery cost and stock that come back.
The second leg only works if the session holds. Sticky sessions keep the same exit address for a defined window, and the postcode you set is usually stored against that session or a cookie. Rotate mid-flow and the site resets you to its default warehouse. That is the most common source of a plausible-looking but wrong delivery cost.
One practical check: before a run, confirm where the address actually resolves. A lookup through what-is-my-ip through the proxy tells you the country and city the retailer will see, and the proxy checker tells you whether the address is alive and what it leaks. Do this once per region per run, not once per project.
How to set up a daily price sweep
The shape of a sweep is simple: a fixed list of products, a fixed list of addresses, one pass at the same hour every day. The failures come from concurrency, session handling and what you do with a blocked response.
- Define the grid first. Products × regions. A thousand SKUs across four regions is four thousand fetches per run, and that number drives everything else — pool size, concurrency, bandwidth, cost.
- Decide what you extract. Price, currency, tax treatment, seller, delivery cost, delivery estimate, stock state. Nothing else needs to come off the page.
- Pin the address per region for the whole run. Rotating per request will get the catalogue page fine and wreck every basket-dependent field. Use one sticky session per region per pass unless the site is lightly defended.
- Schedule with jitter. Same local hour, staggered across regions, so you are not launching thousands of requests in one second. A sweep that takes forty minutes is not worse than one that takes four.
- Classify failures, never coerce them. A 403, a 429, a captcha page or a challenge script is no data. It is not a price of zero and not out of stock.
- Store the raw response before parsing. Keep the HTML or the JSON payload on disk so a parser fix does not mean re-fetching forty thousand pages.
A minimal single fetch through an authenticated HTTP proxy looks like this:
curl -sS --proxy http://user:pass@gate.example.com:8080 -H 'Accept-Language: en-GB' -H 'Accept: text/html' -o raw.html 'https://www.example.com/product/12345'
If the page needs JavaScript, drive a real browser with the proxy attached and the same session held for the run:
browser = playwright.chromium.launch(proxy={'server': 'http://gate.example.com:8080', 'username': 'user', 'password': 'pass'})
In a Scrapy project the equivalent is a proxy middleware plus conservative pacing: CONCURRENT_REQUESTS = 8, DOWNLOAD_DELAY = 0.5, AUTOTHROTTLE_ENABLED = True, and a retry policy that backs off rather than retrying immediately. Whatever the stack, log the exit address used for each result. Without it you cannot tell a genuine price change from a change of address.
Getting delivery cost and stock, not just the price
A price on its own is a poor signal. The landed cost is what a competitor is undercutting you on: item price plus delivery plus tax plus any membership or threshold effect. On a £20 item, free delivery over £25 versus £3.95 below it flips the comparison entirely, and that boundary is often the thing that changed overnight.
Stock matters just as much. "In stock" is not one state. There is stock at the retailer, stock at a third-party seller, dispatch in three days, backorder with a date, and collection-only at a store forty miles away. A monitor that records only in-stock or out-of-stock will report a competitor as unavailable when they simply moved to a marketplace seller — and you will price against a gap that does not exist.
Most of these fields only appear after interaction:
- Delivery cost usually needs a postcode or a basket, so it lives behind a sticky session.
- The buy box rotates between sellers, so record which seller and whether the offer is fulfilled by the retailer or a third party.
- Trade or account pricing needs a login, which means a persistent session and a different address from your public-price sweep.
- Stock and delivery estimates are frequently rendered client-side, so a plain HTML fetch returns the shell and not the data.
Where the page publishes structured data — an offer block with price, currency, availability and seller — read that before writing a selector against the layout. It survives redesigns far better than CSS classes do, and it gives you the currency and availability vocabulary the retailer itself uses.
What continuous monitoring costs
Two things drive the bill, and they are not the same thing. The first is bandwidth. A modern product page with images, fonts, CSS and analytics is heavy; stripped to the HTML or the structured payload you actually need, it is a fraction of that. Since residential and mobile pools are priced per gigabyte, what you choose not to download is a cost decision. Block images, fonts, media and third-party analytics in the browser context and the same run can cost a tenth as much.
Run the arithmetic before choosing anything. Two thousand SKUs across five regions is ten thousand fetches a day, roughly three hundred thousand a month. At a megabyte and a half of full page each, that is around 450 GB. At two hundred kilobytes of trimmed HTML, it is around 60 GB. The request count is identical; the bill is not.
The second driver is burst capacity. Monitoring is not a steady trickle — it is thousands of pages demanded in the same hour, every day. A pool that looks large on paper may have only a few hundred addresses active in your target country at 07:00, and that is the number that decides whether the sweep finishes. Ask what concurrency is realistic per country, not how many addresses exist globally.
The awkward part: the cheapest monitoring is the monitoring you do not do. Fetch every product daily when the retailer changes prices weekly and you are paying for noise. Tier the list — hourly for the twenty products that matter, daily for the few hundred that move, weekly for the long tail. Use conditional requests and caching where the site supports them. Split the work by proxy type: datacentre addresses for catalogue discovery and sitemap walking, residential or mobile for the price and basket reads on defended sites. The provider reviews section covers pricing structure alongside pool claims, and the free proxy list is fine for testing a parser against a lenient site but not for a daily sweep of a major retailer.
Mistakes that record the wrong price
Most bad price data is not caused by a proxy failing. It is caused by a plausible wrong number being written down without a check.
- Not verifying the exit location. A proxy labelled "Germany" may resolve to a data centre in another country, or fail over to a default region. Check the address actually used, per run, and store it with the result.
- Rotating mid-session. The basket empties, the postcode resets, and the delivery cost you record is the default warehouse's. Pin the address for any basket-shaped flow.
- Treating a block as data. Challenge pages, soft 404s and empty search results get parsed into "price unavailable" or "out of stock". Detect them: check status, page size, the presence of the expected product identifier, and whether the price element exists at all.
- Ignoring currency and tax treatment. A currency cookie left over from the previous region converts at the wrong rate, or records a tax-inclusive figure beside a tax-exclusive one. Record the currency code, not just the symbol.
- Reading the wrong seller. Buy-box rotation means the first offer on the page is not always the offer the customer sees. Capture the seller and fulfilment type.
- Comparing across time zones. Promotions start and end at the retailer's local midnight. A sweep at 02:00 UTC catches some regions mid-change and others not.
- Losing the timestamp. A price without the time and the address it was seen at is an anecdote, not a data point. Store SKU, region, postcode, exit address, currency, timestamp and a reference to the raw payload together.
- Retrying hard on a block. Hammering a 403 gets the address banned and eventually the subnet. Back off, rotate, slow down.
The check that catches most of this is boring: once a day, fetch one known product from each region through a browser and compare it against what the pipeline recorded. If they disagree, the pipeline is wrong, not the shop.
What to look for
- Country, city and postcode
- Retail price, stock and delivery cost all move with the address. The postcode is often the variable that matters.
- How the pool holds up on retail sites
- Large retailers run the strictest bot defences outside search, and a blocked request looks like a price change you missed.
- Cost per gigabyte
- Product pages are heavy. The unit rate and what you choose not to download decide the bill together.
- Capacity for a daily sweep
- Monitoring is a burst of thousands of pages at the same hour, every day.
- Sessions for carts and logins
- Delivery cost and trade pricing often sit behind a basket or an account, which needs one address throughout.
Every provider, in order
1. Webshare.io — Best postcode-level targeting
4.7 / 5 for this job · from $0.02 / month
Webshare sits at the top of this list because postcode is usually the variable that decides a price-monitoring job, and its rotating residential pool states targeting at country, state, city, ZIP code and ASN. That is the level at which delivery cost and local stock actually change. The pool is quoted at 80 million-plus addresses, with country counts listed separately for the larger markets.
Pricing is per IP rather than per gigabyte on the datacenter product, which suits a daily sweep: 400K+ IPs at $2.99/month, with the per-proxy rate falling from $0.0299 to $0.0179 as volume grows and a further 30% off on yearly billing. Three pools sit under one account, so a sweep can be split — datacenter threads for catalogue pages that do not fight back, static residential for baskets and account-level trade pricing, rotating residential where the retailer's defences are strict. HTTP and SOCKS5 run on the same credentials, and IP authorisation is offered instead of a username and password. Thread limits run from 500 to 3,000 and bandwidth tiers from 250GB upward.
The catch is in the wording. Unlimited bandwidth carries an asterisk pointing at a Fair Usage Policy, so the word unlimited is not the whole story. The provider's own pages give both 195 countries and 50+ locations, which makes the real footprint hard to check before buying. The static residential list names Comcast and AT&T and then stops at "& more", so a specific carrier cannot be confirmed in advance. The 99.7% success rate and 99.97% uptime are the provider's own figures with nothing independent alongside them.
Key features
- Rotating residential targets country, state, city, ZIP code and ASN
- Free plan: 10 proxies, no credit card
- Per-IP pricing falls from $0.0299 to $0.0179 per proxy as volume grows
- Three pools: datacenter, static residential (ISP), rotating residential
- Bandwidth 250GB to unlimited; threads 500 to 3,000
- HTTP and SOCKS5 on the same credentials; IP authorisation available
Best suited for: teams that need to vary the delivery postcode as well as the country on a daily retail sweep
2. Decodo (ex Smartproxy) — Best all-round retail pool
4.6 / 5 for this job · from $0.08 / GB
Smartproxy, now trading as Decodo, takes second because its residential pool is the broadest general-purpose one here for retail work: 115M+ IPs across 195+ locations, with targeting down to city, ZIP code and ASN rather than country alone. Sticky sessions hold an address for up to 30 minutes, which covers a product page, a basket and a delivery estimate in one visit without changing identity halfway through.
The billing is bandwidth-based, and product pages are heavy, so the unit rate matters more than the headline. The published tiers step down as the block grows, from $3.75/GB at 3 GB to $2.75/GB at 100 GB, with pay as you go at $4/GB and enterprise rates from $2.50/GB at 250 GB. Residential proxies are listed from $2/GB. There is also a Web Data API from $0.15 per 1K requests and a Site Unblocker from $0.95 per 1K requests, which is worth pricing against raw proxy traffic if your parser only needs a few fields from each page. HTTP(S) and SOCKS5 both work, so older SOCKS5-only clients still connect.
The awkward parts are commercial. The 3-day 100 MB trial and the 14-day money-back option are mutually exclusive — taking the trial removes refund eligibility. Refunds apply to a first purchase only, must be requested within 14 calendar days, and are capped at under 20% of a residential plan's traffic. Pay as you go is wallet-only, sold a gigabyte at a time, and cannot be combined with a monthly subscription. The fraud-score comparison on the site does not name the scoring service behind the numbers. High-Speed Proxies carry no published price. Buy from decodo.com: smartproxy.org is named in the record as an impersonator.
Key features
- Residential pool of 115M+ IPs across 195+ locations
- Targeting at continent, country, state, city, ZIP code and ASN
- Sticky sessions held for up to 30 minutes; rotating per request
- Residential tiers from $3.75/GB at 3 GB down to $2.75/GB at 100 GB
- Web Data API from $0.15 per 1K requests; Site Unblocker from $0.95 per 1K requests
- 3-day, 100 MB trial, or a 14-day money-back option on a first purchase
Best suited for: mixed retail sweeps that need one pool covering many countries at city and ZIP level
3. Novada — Best cheap city-level pool
4.5 / 5 for this job · from $0.50 / GB
Novada is third because it is the cheapest way here to get city-level targeting across a wide spread of countries, and price monitoring rewards putting each page on the cheapest pool that will actually return it. Rotating datacenter starts from $0.50/GB across 190+ countries with a stated 99% success rate; rotating ISP starts from $0.65/GB with 360-minute sticky sessions across 100+ countries; residential starts from $1.3/GB at city-level precision over 195+ countries. A daily sweep can therefore put catalogue and category pages on datacenter, logged-in or basket-dependent checks on ISP sessions that hold for six hours, and only the hardest product pages on residential.
The residential pool is quoted at 100M+ IPs, with published country counts including 8,426,658 for the United States and 2,838,522 for the United Kingdom. Sessions run 1–120 minutes on residential. Mobile inventory is listed separately at 2M+ IPs and 120+ carriers. Entry is a $10 credit with no card required, plus a 10GB new-user residential pack for $10. Scraping APIs draw on a single credit balance, so a SERP check and a page fetch come out of the same top-up.
Two things to weigh. The protocol support listed is HTTP(S) only — no SOCKS5 — which rules out clients that only speak SOCKS5. And the 99.99% success figure is quoted overall while rotating datacenter is stated at 99%, so the headline does not apply evenly across the pools. Unlimited Proxies, IPv6 Proxies and Datasets have no published price and are quoted through an account manager, and the free trial covers the Scraper API rather than the proxy pools.
Key features
- Residential from $1.3/GB; rotating datacenter from $0.50/GB; rotating ISP from $0.65/GB
- 100M+ residential IPs across 195+ countries with city-level targeting
- 360-minute sticky sessions on rotating ISP; 1–120 minute sessions on residential
- Published country counts: 8,426,658 US IPs, 2,838,522 UK IPs
- $10 credit with no card, plus a 10GB new-user residential pack for $10
- HTTP(S) support only; no SOCKS5 listed
Best suited for: budget-conscious sweeps that want city-level targeting and can route easy pages onto datacenter traffic
4. NodeMaven — Best day-long sticky sessions
4.4 / 5 for this job · from $2.20 / GB
NodeMaven is fourth, and it earns the position on one feature: sticky sessions quoted at up to 24 hours with unlimited concurrent sessions. For price monitoring that matters at the basket stage. Delivery cost, trade pricing and stock by postcode often sit behind a login or a cart, and a session that survives a working day means one address through the whole flow instead of a fresh login every half hour. The residential pool is 30M+ IPs, with country, city and ISP targeting and ZIP level claimed by the provider.
ISP plans carry unlimited traffic, a 99.9% uptime guarantee and UDP support, which suits a monitored checkout that has to keep returning the same address. Traffic rolls over on both pay-as-you-go and monthly plans, so an idle month is not lost spend. There is a quality guarantee of $1 in bonus traffic for each confirmed blacklisted proxy, and up to 10% cashback on traffic on standard plans. Residential starts at $2.20/GB, with the lowest monthly plan at 2 GB for $8.5. HTTP, HTTPS and SOCKS5 are all supported, and the API is RESTful with SDKs and documentation.
The catch is cost at scale. Monitoring is thousands of heavy pages at the same hour every day, and a per-GB traffic model gets expensive quickly at that volume. There is no free trial: the $3.50, 750 MB entry point is a paid demo, and third-party notes say the free route is limited to registered companies that pass KYC. Smaller locations are reported to be uneven in depth, refunds are pro-rated rather than full, and the mobile pool is quoted at 250K by the provider against 295K+ in third-party records.
Key features
- Sticky sessions up to 24 hours with unlimited concurrent sessions
- 30M+ residential IPs; 250K mobile IPs
- ISP plans with unlimited traffic, 99.9% uptime guarantee and UDP
- Traffic rollover on pay-as-you-go and monthly plans
- Residential from $2.20/GB; 2 GB monthly plan at $8.5
- $1 in bonus traffic for each confirmed blacklisted proxy
Best suited for: monitoring flows that need to hold one address through a login or basket for hours at a time
5. Oxylabs.io — Best unblocking for retailers
4.3 / 5 for this job · from $0.40 / GB
Oxylabs is fifth, and it takes the unblocking line because it sells the part that most often fails on a large retailer: a managed endpoint rather than a raw proxy. Web Unblocker is priced per gigabyte, from $7.2/GB on the $30/month Starter plan down to $3/GB at enterprise level. Residential proxies run from $6/GB on Starter to $2.5/GB at enterprise. That is not cheap traffic, but a blocked request that returns a captcha page instead of a price is worse than an expensive one that returns the price.
The pool is stated at 177M+ IPs, 175M+ of them residential, across 195 locations. Every product and tier carries a published unit rate, so a sweep can be costed before anyone talks to sales — useful when a monitoring contract needs sign-off. Compliance is itemised rather than implied: ISO/IEC 27001:2022 covers three named products, SOC 2 Type II covers two named endpoints, and technology errors and omissions plus cyber insurance apply across the catalogue. GDPR compliance and EU AI Act alignment are stated. A single subscription carries the whole catalogue, with card, wire in USD or EUR, AliPay and PayPal accepted.
The entry cost is the weak point. The only free option is $5 in welcome credits, which is under a gigabyte of residential traffic at Starter rates. The cheapest paid plan is $30/month and enterprise starts at $2,500/month through sales. VAT may apply on Starter, Basic and Flexible, so the headline rate is not always the invoice. Bitcoin is not accepted. Certification is per product, not company-wide, so check that the product you intend to buy is inside the stated scope.
Key features
- Web Unblocker from $7.2/GB on Starter to $3/GB at enterprise
- Residential proxies from $6/GB on Starter to $2.5/GB at enterprise
- 177M+ IPs total, 175M+ residential, across 195 locations
- ISO/IEC 27001:2022 on three named products; SOC 2 Type II on two endpoints
- Technology E&O and cyber insurance across all products
- Starter at $30/month; enterprise from $2,500/month; $5 welcome credits
Best suited for: retail monitoring where the target sites are hard to unblock and procurement needs certified, insured handling
6. ProxyEmpire — Best rollover for sweeps
4.2 / 5 for this job · from $0.35 / GB
ProxyEmpire closes the list because rollover is the right answer to how a daily sweep actually behaves. Monitoring is a burst at the same hour every day, and volume moves with the catalogue and the season. On GB-based plans here, unused bandwidth rolls over and never expires, so a quiet week is not money burnt. Plans are prepaid with no contracts and no monthly minimums.
The residential pool is 30M+ IPs across 170+ countries, with targeting at country, region, city, ISP, ASN, OS/TCP and ZIP — the last of those being what postcode-level delivery pricing needs. Rotating residential starts at $3.50/GB pay as you go, falling to $1.50/GB on the 1000 GB plan, with the EMPIRE50 code taking half off eligible GB-based plans. Unlimited sessions and concurrent connections are available for approved use cases, and a $1.97 trial of 150 MB does not auto-convert into a subscription. Sessions can be time-based, session-based or a new IP on request.
The scope narrows outside residential. Coverage is uneven by product: 62 datacenter countries and 24 static ISP countries against 170+ for residential and mobile, and dedicated mobile reaches only 5+ countries and 50+ locations with city-level targeting dropped to country, city and carrier. Unlimited bandwidth is priced per day or week from $200/day with no rollover, so it is a different purchase entirely. The trial excludes static ISP and datacenter, EMPIRE50 excludes flat-rate and static plans, and the site 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
- 30M+ residential IPs across 170+ countries
- Residential targeting by country, region, city, ISP, ASN, OS/TCP and ZIP
- Rotating residential from $3.50/GB pay as you go; $1.50/GB at 1000 GB
- $1.97 trial of 150 MB, no auto-conversion
- Static ISP covers 24 countries; dedicated mobile 5+ countries
Best suited for: sweeps with uneven daily volume, where unused gigabytes need to survive to the next burst
7. Floppydata — Best $1/GB residential
4.1 / 5 for this job · from $0.63 / GB
Seventh, and the reason it is here is narrow: one rotating product at $1.00/GB that spans residential, mobile and datacenter. For price monitoring that matters when a sweep needs residential addresses for the catalogue and a handful of carrier addresses for the pages that reject everything else, without a second account and a second invoice. Targeting sits underneath — 190+ locations with country, city, ASN and IP-type filters — which is the layer that decides which regional price comes back.
The bulk tiers are open rather than negotiated: 20% off at 96 GB and 30% off at 120 GB. Read the per-gigabyte numbers by product, because the headline rate is not the residential rate. At the 120 GB tier residential works out at $2.07/GB and mobile at $2.07/GB; $0.63/GB is the datacenter tier at 120 GB. Where a fixed address is needed for a basket or a trade-pricing login, static datacenter IPs are $0.79/IP/month in bundles of 20, 50 or 100 and static residential is $1.79/IP/month. Payment by card through Stripe or 50+ cryptocurrencies including USDT, BTC and ETH.
The catch is support and documentation. There is no phone line, so a live escalation during a broken nightly run is not available, and custom rotation logic is thinly documented — expect to work it out from the API rather than a guide. The 100 MB trial per proxy type is a connectivity check, not a load test, and will not tell you how the pool behaves at 6 a.m. against a strict retailer. The 99.99% uptime and sub-0.3-second response figures are provider claims.
Key features
- Rotating traffic from $1.00/GB across residential, mobile and datacenter
- 190+ locations; country, city, ASN and IP-type targeting
- 20% off at 96 GB, 30% off at 120 GB
- Static datacenter IPs from $0.79/IP/month in bundles of 20, 50 or 100
- 100 MB free trial for each of residential, datacenter and mobile
- Cards via Stripe plus 50+ cryptocurrencies
Best suited for: teams that want residential, mobile and datacenter rotation on one account at a flat $1.00/GB entry rate
8. Proxy-Seller — Cheapest residential per GB
4.0 / 5 for this job · from $0.30 / GB
Eighth, because $0.30/GB is the lowest residential rate published among these providers and price monitoring is one of the jobs this provider names itself. The pool is 47M+ rotating residential IPs across 220+ locations, both provider figures. For a nightly sweep that is ample address space, and the no-monthly-lock-in terms suit a quarterly price audit that only runs in the months it is needed.
The detail that decides procurement reviews: ISO/IEC 27001:2022 and ISO 9001:2015 certification, GDPR, CCPA and ePrivacy statements, and DPA/SCC readiness. HTTP, HTTPS and SOCKS5 with a RESTful API, IP whitelisting and SDK access. Six proxy types under one account, which means residential for catalogue pages and static IPv4 or ISP addresses for the basket step — IPv4 at $0.49/IP, IPv6 at $0.02/IP, ISP from $0.98/IP. A 1 Gbps IPv4 channel across 600+ networks is the provider's specification.
The catch is mobile. Mobile coverage reaches 20+ countries against 220+ residential locations, so if a retailer only relents to carrier addresses, this is a thin option outside the main markets. Uptime at 99.7%, connection success at 99.5%+ and response time around 0.35s are self-declared with no independent test cited — treat $0.30/GB as the published fact and the performance numbers as something to check against your own targets before scaling.
Key features
- 47M+ rotating residential IPs across 220+ locations (provider figures)
- Residential from $0.30/GB; IPv4 $0.49/IP; IPv6 $0.02/IP; ISP from $0.98/IP
- Mobile coverage limited to 20+ countries
- ISO/IEC 27001:2022 and ISO 9001:2015 certified; DPA/SCC ready
- HTTP, HTTPS and SOCKS5 with RESTful API and IP whitelisting
- No monthly lock-ins stated
Best suited for: high-volume residential sweeps where $0.30/GB and 220+ locations matter more than mobile coverage
9. FlyProxy — Best bulk-sweep rate
3.8 / 5 for this job · from $0.90 / GB
Ninth, and the bulk-sweep angle comes with a condition attached. $0.90/GB is the 10,240 GB tier — that is ten terabytes of product pages before the rate applies. At 20 GB the price is $2.50/GB and at 50 GB it is $2.40/GB, so the cheap number describes a very large operation, not a modest daily sweep. Where the volume is genuinely there, unlimited concurrent sessions is the feature that matters: a monitoring run is thousands of requests in the same hour, and thread counts that are not capped let the scheduler work at the size of the job.
Third-party country tests reported in the record give success rates from 96.8% in Mexico and 97.2% in Brazil up to 98.7% in Germany, with response times from 0.52s to 1.00s and the United States at 98.4%. That shape of data is the useful part — the softer markets are the ones where a failed fetch looks like a missing price. Coverage is 195+ countries with country, state and city targeting, and IP lifetime is around 24 hours, so session planning should assume roughly a day of stickiness at the top end rather than a week.
The catch is the rest of the line-up. There are no mobile proxies, so basket and login steps needing carrier addresses are out. The provider states no refunds once products are sold, the free 500 MB offer runs through sales rather than self-serve, and a third-party listing says no trial at all. Pool size, protocol list and dashboard availability disagree between the provider's own pages and the third-party listing, which also records no rotation and no dashboard. Confirm all of that on the account before committing to ten terabytes.
Key features
- $0.90/GB only at the 10,240 GB tier; $2.50/GB at 20 GB, $2.40/GB at 50 GB
- Unlimited concurrent sessions, stated by provider and third party
- State and city targeting across 195+ countries and regions
- Third-party country success rates from 96.8% (Mexico) to 98.7% (Germany)
- IP lifetime around 24 hours
- No mobile proxies; no refunds once products are sold
Best suited for: operations already planning ten-terabyte monthly sweeps across many countries at unlimited concurrency
10. ProxySale.com — Budget city-level targeting
3.6 / 5 for this job · from $0.06 / GB
Tenth, and the city-targeting angle needs one correction first. The $0.06/GB in the price column belongs to the Video Downloader API, not the residential pool. Residential starts at $0.5/GB on usage-based billing, which is still low, and what sits alongside it is country, state and city targeting across 86M+ IPs in 200+ countries, by the provider's own count. Postcode granularity is not listed; city is, and city is where most regional delivery pricing actually diverges.
Two products matter for retail work. The Web Scraper API bills only for successfully delivered results, which changes the arithmetic of a sweep: a blocked request costs nothing, so the risk moves from wasted bandwidth to gaps in the dataset. There is a 30-day zero-setup trial on that API, and it is the only free access in the line-up. Static IPs are $5/IP residential and $3/IP datacenter for carts and trade-pricing logins, and the unlimited residential plan does random and sticky rotation with no IP or traffic cap.
The catch is billing shape. Unlimited Residential is $69.68 per day, which is workable for a release-day watch and awkward for a standing nightly sweep — the cost scales with the calendar, not with the work. The service is not available in mainland China by the provider's own policy. Uptime at 99.9%, a 99.5% residential success rate and sub-0.6s responses are provider-declared with no third-party audit shown.
Key features
- 86M+ residential IPs across 200+ countries (provider count)
- Country, state and city targeting on residential proxies
- Residential from $0.5/GB, billed on usage
- Unlimited Residential at $69.68/day, no IP or traffic cap
- Web Scraper API bills only for delivered results; 30-day trial on it
- Not available in mainland China
Best suited for: city-level price collection on a usage-billed budget, with a scraper API that only charges for delivered results
11. Nsocks — Best broad city coverage
3.4 / 5 for this job
Eleventh, for breadth of city coverage rather than polish. Country and city targeting across 190+ country and city-level regions is the claim that counts for this job: retail pricing moves by metro, and a list that stops at country level will not separate one city from the next. Rotating, static, datacenter, ISP and unlimited proxies are sold as separate products with separate billing, so a nightly sweep can run on rotating residential while cart and delivery-cost checks sit on static addresses. HTTP(S) and SOCKS5 are supported, with listed integrations for Playwright, Selenium and an API, plus a dashboard showing traffic statistics.
The awkward parts are not small. There is no free trial, so pool quality cannot be checked before paying, and the provider states it does not refund products sold. The pool count is quoted as 80M+ in one place on its own home page and 90M+ in another, so neither is a planning number. Third-party records put the founding at 26 August 2024, which is a short history for a supplier carrying a nightly job. Residential pricing is tiered from 10 GB up to 500 GB, but no figures are published against those rows, so the per-gigabyte comparison has to be done on a quote rather than a price list.
Support is 24/7 by live chat and email, and 99.95% uptime is the provider's own figure.
Key features
- 80M+ residential IPs claimed; home page also says 90M+
- Country and city targeting across 190+ country and city-level regions
- Rotating, static, datacenter, ISP and unlimited products billed separately
- Playwright, Selenium and API integrations; dashboard with traffic statistics
- No free trial and no refunds on products sold
- Founded 26 August 2024, per third-party records
Best suited for: multi-market price collection that needs city-level granularity and separate static IPs for cart steps
12. AnyIP.io — Best week-long sticky sessions
3.2 / 5 for this job · from $2 / GB
Twelfth, and the reason is a single feature: sticky sessions up to 7 days. Price work that goes past the catalogue into a basket, a delivery-cost step or a logged-in trade price needs one address held across the whole flow, and a session measured in days rather than minutes is the difference between a clean checkout read and a challenge mid-flow. Smart Sessions keep the address inside the same carrier and area, which is exactly what a delivery estimate depends on.
It is one plan covering residential and mobile with only bandwidth metered, so carrier addresses for the stubborn retailer do not need a second contract. Rotation and concurrent requests are both unlimited. Filtering runs to country, ISP, region and precise GPS coordinates, finer than most, and useful where a retailer prices by locality rather than by city boundary. Unused data rolls over to the next billing cycle, quota control and alerting are included, and there is no contract.
The catch is price and geography. The advertised $2/GB is the Enterprise extra-gigabyte rate on a $1000/500 GB plan; the entry point is $25 for 5 GB, and extra traffic is $5/GB on Micro and $4/GB on Basic. Product pages are heavy, so at those rates a careless sweep is expensive — block images, cap redirects and fetch only the part of the page the price lives in. 38+ countries is a short list for multi-market work. There is no free trial: Micro is the trial and it is paid up front. Crypto orders fall outside the 14-day money-back guarantee, which covers card purchases.
Key features
- Sticky sessions up to 7 days; Smart Sessions stay within one carrier and area
- Residential and mobile on one plan, bandwidth-only metering
- Unlimited IP rotation and unlimited concurrent requests
- Country, ISP, region and precise GPS coordinate filtering
- $25/5 GB Micro up to $1000/500 GB Enterprise; $2/GB is the Enterprise rate
- 38+ countries; unused data rolls over to the next cycle
Best suited for: basket, delivery-cost and logged-in price checks that need one address held for days
13. GonzoProxy — Best no-expiry ZIP traffic
3.1 / 5 for this job · from $2 / GB
GonzoProxy sits at thirteen because the mechanic behind its placing — postcode-level targeting on traffic you buy by the gigabyte — is the one most directly aimed at retail pricing, and because the expiry terms suit a sweep whose daily volume nobody can predict. Retail prices, stock and delivery charges move with the address, and often with nothing finer than the postcode. GonzoProxy targets country, city, state, continent, ZIP code and ASN. The pool is 20M+ addresses across 190+ countries on third-party records, with HTTP and SOCKS5 published by the provider and SOCKS4 also listed by third parties.
Two things matter for a daily sweep. Traffic never expires and unused gigabytes roll over to the next plan, so a day when half the catalogue returns a challenge page does not waste the month's budget. Concurrency is unlimited on every plan, which is what a burst of thousands of pages at the same hour needs. Sticky sessions hold an address for up to 72 hours, long enough to work through a basket or a trade-pricing login on one identity. No KYC, access claimed live in about 30 seconds, and payment by card or crypto.
The catch is on the price sheet. Product pages are heavy, so the meter runs quickly, and the ladder starts at $6.5/GB for 1 GB, falling to $3.9/GB at 100 GB, $2.75/GB at 500 GB and $2/GB on the largest volume band. Two published residential ladders quote different rates for the same volume — 50 GB is $225 on one and $175 on the other — so check which one you are buying. The demo is not free ($1.99 for 0.3 GB), the dashboard lacks detailed per-request logs, which is awkward when you need to know whether a page came back empty or was blocked, and mobile coverage is Poland and Ukraine only. Published speed and success figures are the provider's own and not independently verified.
Key features
- ZIP, city, state, country, continent and ASN targeting
- Traffic never expires; unused GB rolls over to the next plan
- $6.5/GB at 1 GB, $3.9/GB at 100 GB, $2/GB at 5,000+ GB
- Unlimited concurrent connections on every plan
- Sticky sessions up to 72 hours
- 20M+ IPs across 190+ countries (third-party)
Best suited for: postcode-level retail checks where daily volume is unpredictable and unused gigabytes must keep
14. Live Proxies — Best private IP allocation
3.0 / 5 for this job · from $2
Live Proxies sits fourteenth on the strength of private IP allocation: the provider says your addresses are unique and shared only with you. For retail monitoring that is worth something specific. Large shops score an address partly on how many other people have already hammered it, and a private address is not carrying someone else's scraper traffic when your sweep arrives.
The rest of the package suits the awkward half of the job — carts, accounts and trade pricing. Rotating residential sessions hold one address for 60 minutes, rotating mobile for 30, which is enough for a basket flow where the delivery cost only appears after the first page. There is no concurrency limit and no cap on threads or requests, as stated, and unused bandwidth rolls over to the standby plan. Dashboard access, API access and IP whitelisting are included, with metered and unmetered billing. Entry is not cheap: the smallest B2C row is $70 for 4 GB over 30 days, and 50 GB over 60 days is $500. A $2,000 a month B2B enterprise tier sits on the 10M+ pool.
The catch is the postcode. Geography is described only as available countries or state-level options, with no published country count, and state IPs are US only. On a job where a delivery charge can turn on a single postcode, that is a gap rather than a detail. Pricing is also unsettled: the home page quotes $45.00, $65.00 and $55.00 as starting prices while the pricing page says B2C plans start from $70, and static residential is listed at $65.00/plan on one page and marked permanently unavailable on the other. SOCKS5 is B2B only. The free trial goes to custom plan users after an agreement; everyone else is told to buy the smallest plan.
Key features
- Private IP allocation: addresses unique and shared only with you
- No concurrency limit, unlimited threads and unlimited requests
- 60-minute sticky sessions on rotating residential, 30 on mobile
- Unused bandwidth rolls over to the standby plan
- Metered and unmetered billing options
- 10M+ IP pool on the B2B enterprise tier
Best suited for: basket and account checks that need one private address held for the whole flow
15. ASocks.com — Best flat per-proxy billing
2.7 / 5 for this job · from $5 / proxy
ASocks.com sits fifteenth because it prices the other way round: $5 per proxy per month for residential and corporate lines, $15 for mobile, rather than a rate per gigabyte. That changes behaviour on this job. Product pages are heavy — images, review blocks, scripts, recommendation carousels — and on a metered plan the temptation is to strip them out and hope the price still parses. A flat per-proxy bill decouples cost from page weight, so a daily sweep of the same catalogue costs the same whether you pull the whole page or a fragment of it.
Volume pricing is published: 5% off from 10 proxies, 10% from 50 and 15% from 100. The provider claims 150+ locations, 7M+ IP addresses, a 99.7% connection success rate and 100,000+ clients, runs its own infrastructure and links that to cleaner addresses and fewer bans, and states 24/7 support with an average response under ten minutes. Those are the provider's numbers, unaudited.
The catch is what the record leaves out, because for this job some of it is the point. There is no published session detail, no concurrency or traffic allowance, and no geo granularity below "150+ locations" — no country, city or postcode list. If your sweep depends on quoting a Leeds postcode against a London one, that absence is the whole question. Every headline figure is provider-reported, and mobile costs three times the residential rate per proxy.
Key features
- $5 per proxy/month residential and corporate; $15 mobile
- 5% off from 10 proxies, 10% from 50, 15% from 100
- 150+ locations and 7M+ IPs (provider-reported)
- 99.7% connection success and 100,000+ clients (unaudited)
- Own proxy infrastructure
- 24/7 support, stated average response under 10 minutes
Best suited for: fixed-budget daily sweeps where a set number of proxies beats metering gigabytes
16. MobileProxy.Space — Best unlimited mobile traffic
2.2 / 5 for this job · from ₽490 / day
MobileProxy.Space sits sixteenth because it sells mobile addresses by the calendar instead of the gigabyte. Mobile rental starts at 490₽/day with unlimited traffic, so the bill for a sweep depends on how long you rent, not on how much of each page you download. On retail sites that pairing is hard to beat: mobile addresses survive bot defences longer than datacentre ranges, because they look like phones on a carrier network rather than a rack in a hosting hall, and unlimited traffic means the parser can fetch images, stock badges and delivery estimates in full.
The minimum rental is one day, which lets you run a real catalogue and see the hit rate before committing. Rotation runs on a timer or by hitting an API link, addresses change in 2–7 seconds per the provider, and each mobile proxy takes up to 250 simultaneous connections. Dead proxies are replaced automatically in the provider's wording, refunds cover unused time, and there is an API plus an MCP server for Claude Code, Cursor, Python, Node.js, PHP and cURL. The dashboard carries a proxy checker, port checks, a speed test and a browser proxy manager extension.
The catches are coverage and consistency. Third-party records and the provider's own sheet disagree on nearly every count — 30+ countries against 53, 130+ mobile operators against 223, 700+ servers against 1,356 — and some customers have reported problems with geographical locations, laggy proxies and late-night maintenance windows. Location granularity is not stated finely enough to plan postcode-level checks. Third-party write-ups also flag the prices as relatively high, and because the meter is time, a permanent daily sweep pays for every day whether the sweep needed it or not.
Key features
- Mobile plans bill by rental period with unlimited traffic
- Minimum rental of one day
- From 490₽/day for mobile proxies
- Up to 250 simultaneous connections per mobile proxy
- IP change in 2–7 seconds; 99.9% uptime claimed
- API plus MCP server for Claude Code, Cursor, Python and Node.js
Best suited for: mobile-address sweeps where page weight is unknown and gigabyte metering would throttle the parser
17. ChangeMyIP.com — Best cheap shared test
1.8 / 5 for this job · from $1
ChangeMyIP.com sits seventeenth. At $1.00 for a shared HTTP proxy, $1.80 for a dedicated one and $2.20 for SOCKS5, it is the cheapest way to prove that a scraper's proxy handling, rotation loop and parser work at all before any residential traffic is paid for. Activation is instant, bandwidth is unlimited on the plans it sells, ports run at 1Gb/s and there is no contract. The protocol spread is wider than most: HTTP, SOCKS5, ShadowSocks, Trojan, a DNS proxy and a Wireguard VPN line, paid by card, PayPal or crypto through Zonda.
For retail price monitoring, though, it is the wrong instrument. Shared means multiple users on the same IP with bandwidth divided between them — the provider says performance can suffer on busy servers — and an address other people are already scraping from is the first one a large shop challenges. Nothing in the record gives country, city or postcode selection, which removes the variable that decides price and delivery cost. Paypal.com is blocked at proxy level on every product, which matters if the figure you need sits behind a checkout.
Two more practical notes. A $10 minimum order applies even when you want a single $1.00 shared proxy, so the entry ticket is ten times the headline. And the 30-day money-back guarantee does not apply once the service has been used, with refunds limited to bugs or slow performance. The "fully anonymous and undetectable" line is marketing with no published independent test behind it. The provider has been selling proxies for more than twelve years, which is a fair signal of survival rather than of fit.
Key features
- $1.00 shared HTTP proxy; $1.80 dedicated HTTP; $2.20 SOCKS5
- Unlimited bandwidth on the plans it sells
- Multiple IP subnets with 1Gb/s ports
- HTTP, SOCKS5, ShadowSocks, Trojan, DNS proxy and Wireguard options
- $10 minimum order
- 30-day money-back guarantee, void once the service is used
Best suited for: proving a scraper's proxy plumbing works before spending on residential traffic
Price monitoring proxies: questions we get
Is price monitoring with proxies legal?
Reading prices that are visible to the public is generally lawful in the UK and EU, though contract terms, database rights and how you reuse the data are separate questions. Breaching a site's terms of service is a contract issue rather than a criminal one, but it can still get your access withdrawn. Do not use proxies to reach accounts or areas you have no right to, do not overload a site, and take advice where the commercial stake is high.
Do I need residential proxies to track marketplace prices at scale?
Not always to start. Small batches often succeed on datacentre addresses, which is the cheapest way to find out whether a site will tolerate you. The larger marketplaces challenge datacentre ranges quickly, so sustained daily coverage usually needs residential or mobile addresses. Test with a few hundred requests on each address type before committing to a per-gigabyte bill.
How do I see prices in another country?
Two things have to match: the country the site thinks you are in, and the site's own locale, currency and delivery settings. Set both, then confirm the exit address actually resolves where it claims — a mislabelled proxy gives you a plausible price in the wrong currency. If the price depends on a delivery postcode, set that inside the same sticky session.
Why does the price in my dataset differ from what a shopper sees?
Usually one of five things: a different exit address, a different session, a different time of day, a different seller in the buy box, or a different tax treatment. Any one of those is enough to move the number. The fix is not more proxies; it is recording address, session, seller, currency and timestamp alongside every price.
What is a sticky session and why does it matter for price monitoring?
A sticky session holds one exit address for a defined window instead of rotating on every request. Baskets, postcode selections, delivery estimates and logins are all tied to the address that set them. Rotate in the middle of that flow and the site quietly resets you to its default warehouse, which is where most wrong delivery costs come from.
Why am I getting a 403 or a captcha page instead of a price?
Bot management decided your request did not look like a shopper. The usual causes are the address type, the request pace, a mismatched TLS or header fingerprint, or missing cookies from an earlier page. The proxy only addresses the first of those. Slow down, keep a realistic session, and treat the block as missing data rather than a price.
Do I need a headless browser for price monitoring?
Only where the fields you need are rendered client-side, which is common for stock and delivery estimates. A real browser costs far more bandwidth and CPU per page, so it is worth checking whether the page publishes an offer block in HTML or structured data first. A hybrid works well: plain fetches for the bulk, a browser for the pages that will not give up the numbers otherwise.
Can I use free proxies for price monitoring?
For testing a parser against a lenient site, yes — it costs nothing to find out whether your selectors work. For a daily sweep of a major retailer, no. Free pools are shared, short-lived and heavily blocked, and you cannot tell who else has been using the same address before you arrived. Treat them as a development tool, not infrastructure.
How often should I check competitor prices?
As often as they actually change, which is usually less often than people assume. Start with a daily sweep across your whole list, then look at the change history and tier it: hourly or a few times a day for the handful that move constantly, weekly for the long tail. Frequent fetching costs bandwidth and raises your block rate for no extra insight.
What should I store alongside each recorded price?
At minimum: the SKU, the region, the delivery postcode, the exit IP address used, the currency code, whether the figure includes tax, the seller and fulfilment type, the stock state, and a timestamp in the retailer's local time. Keep a reference to the raw payload as well, so you can re-parse later without re-fetching.
How many concurrent requests can a monitoring setup push?
Fewer than the pool size suggests. What matters is how many addresses are active in your target country during your sweep window, and how much traffic the site will take before it starts challenging you. Start with a low concurrency, raise it gradually, and watch the block rate rather than the throughput. A sweep that finishes in forty minutes with clean data beats one that finishes in four with half the pages missing.
Other jobs people buy proxies for
- Web scraping
- SEO and rank tracking
- Ad verification
- 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.