Solutions › SEO & SERP rank tracking

Scraping search results for rank tracking

Rank tracking is a high-volume, highly repetitive fetch of pages that are personalised by location and metered per IP address. What it needs from a proxy is a large rotating pool, precise control of the exit country, a browser-like TLS fingerprint, and session handling that keeps queries spread across addresses rather than piled onto one. Datacenter IPs do this at a cost per query that residential pricing cannot approach.

What counts as a SERP target

  • Organic rank tracking — positions by keyword, device and locale, on a daily or weekly cadence.
  • SERP feature monitoring — featured snippets, people-also-ask, local packs, shopping and news blocks, AI answer panels.
  • Paid-listing observation — which advertisers appear for which terms in which markets.
  • Share of voice and competitor reporting across a keyword portfolio.
  • Marketplace and app-store search, which behaves the same way and has the same constraints.

The shape of the workload is what makes it distinctive: tens of thousands of near-identical requests, all to one host, all needing to be comparable with yesterday's run.

Location decides the result set, not just access

A search result page is a function of where the request came from. The same keyword returns different ordering, different local packs and sometimes a different index depending on country and region, so the exit IP is part of the measurement rather than a detail of transport. A rank reported from the wrong country is not a slightly noisy rank, it is a different number.

Two rules follow. Pin the exit explicitly in the proxy login (pxu1000-0+US, or a state such as us-ny), and keep it stable across runs — a tracker that silently drifts from a US to a European exit will report a ranking change that never happened. All locations are included on every plan, so tracking several markets is a login change, not an upgrade.

Geography is also checked before anything else: in our probing of US public-data sites, two servers answered a US datacenter IP normally and returned 403 to a European one. Search engines are not those sites, but the lesson generalises — match the exit to the market before you debug anything more exotic.

Query budgets are counted per IP

Search engines meter how many queries one address may issue in a window. That is a budget, not a ban, and it is the mechanism that dominates rank-tracking throughput. Our production traffic shows the same effect across the board: measured on 268 targets, 28 show failure rates rising sharply as the same volume is concentrated onto fewer exit IPs. Throughput therefore scales with how many distinct addresses your queries are spread across and how evenly they are paced — not with how many threads you start.

Rotation for rank trackers

Every request through BotProxy belongs to a session, and a session holds one exit IP. Supply no session id and a single default session is used, pinning all your queries to one address — which for a SERP workload burns the budget almost immediately. The session is part of the proxy login:

login + location + SESSIONID

pxu1000-0+US+kw-00042      # keyword batch 42, US market
pxu1000-0+DE+kw-00042      # same batch, German market

Change the session id to change the IP. A session is dropped after 5 minutes idle, and limit_session_age caps its total lifetime — 60 seconds on most credentials, 0 to remove the cap.

Patterns that work for trackers:

  • A fresh session per keyword, or per small batch. Queries are independent, so there is nothing to keep sticky.
  • Hold the session across pagination when you need results beyond the first page, then drop it.
  • Rotate on the first sign of throttling rather than retrying through the same address.
  • Spread a keyword portfolio over time. Bursting the whole list in five minutes concentrates load no matter how many sessions you use.

Anti-Detect Mode

Search engines fingerprint the TLS and HTTP/2 handshake, and a default HTTP client is recognisable as automation before its headers are read. Anti-Detect Mode replaces the handshake with a real browser's; it is the mechanism that gets past fingerprint-based checks in front of WAF-protected targets generally. It is enabled by default. Because it works as a man-in-the-middle your client must accept the proxy's certificate, and the impersonated profile supplies the User-Agent, overriding your client's — which is worth knowing for a tracker, since mobile-versus-desktop results depend on that header. Select the browser profile to select the device you are measuring.

Detecting a degraded SERP

The dangerous failure in rank tracking is not an error, it is a result page that arrives with HTTP 200 and fewer results than a real one — a consent wall, a "verify you are human" body, or a trimmed set. Counted by status code it is a success; parsed naively it silently reports a ranking drop. Validate the shape of the page: expected number of organic results, presence of the query echo, plausible result count. Anything that fails those checks is a fetch error, not a ranking change.

Where an API is the better answer

If what you need is your own site's performance, Google Search Console and Bing Webmaster Tools give you impressions, clicks and average position directly, with no crawling and no sampling error. Use those for first-party reporting. Scraping is the right tool for the parts they cannot answer: competitor positions, SERP feature layout, and the result page as a searcher in a given market actually sees it.

Where this will not work

Anything requiring a signed-in search session, and engines that have moved the surface you want behind an interactive challenge, are outside what rotating datacenter IPs can do. Check the target's terms of service before you scale, and keep your query rate defensible — a tracker that is polite per address is also the one that keeps working.


Try it on the target you actually need

Paste your URL into the live tester and see the real response before you create an account. Every plan includes all locations, Anti-Detect Mode and the Browser API, and is billed on traffic volume alone.

Test your target See plans Read the documentation

Other things people scrape with BotProxy

Public records & government registries

Licensing boards, Secretary of State business registries, county assessors and recorders. Roughly 1,450 separate government domains, each running its own stack and its own WAF.

B2B data & business directories

Company and professional profiles, directories and firmographic enrichment — long-running pipelines that re-check the same records on a schedule.

E-commerce price & catalog monitoring

Marketplace and retailer product pages, price history, stock and assortment tracking across many sites at once.

SEO & SERP rank tracking

Search result pages by keyword and locale, rank monitoring and share-of-voice reporting for agencies and in-house teams.

Travel & airline fare monitoring

Airline and OTA fare and availability checks, where the answer depends on the country the request comes from.

Property & real-estate data

Listing portals plus the county assessor, recorder and land-record sources that carry the authoritative ownership and tax data.