Proxy tier benchmark Targets

Does the Arizona Department of Real Estate site block datacenter proxies?

Professional licensing boards · minimum viable tier: no proxy needed · measured 2026-08-18

Unproxied requests were served normally and proxied ones were not: 100% direct against 40% on anti-detect with a pinned exit, 60% on a rotating datacenter exit and 90% residential. Read this result with its URL attached — what was measured is azre.gov/, the department homepage, not the public database search. Measured 2026-08-18.

What was measured

URL https://azre.gov/
Content assertion Department of Real Estate
Arms 6
Rounds per arm 10
Requests 60
Measured 2026-08-18, 15:38–18:09 UTC
Benchmark host DigitalOcean droplet, New York (nyc3)

A request counts as a success only if the expected content is present. An HTTP 200 carrying a challenge page, an empty body or the wrong page is a failure, because it is not a page you can use.

This measures whether a tier works, not long-run stability under sustained load. Ten requests per arm, at a two-second floor, never approach a per-IP rate limit — which is the thing a rotating pool exists to defeat.

Results by arm

Arm Success p50 p95 Bytes per success Outcomes
direct (no proxy) 100% 132 ms 173 ms 50,972 success 10
datacenter, one pinned exit 80% 370 ms 1,073 ms 50,972 success 8, blocked 2
datacenter, rotating exit 60% 385 ms 1,041 ms 50,972 success 6, blocked 4
anti-detect, one pinned exit 40% 459 ms 1,123 ms 50,972 success 4, blocked 6
anti-detect, rotating exit 50% 450 ms 1,137 ms 50,972 success 5, blocked 5
residential, rotating exit 90% 1,479 ms 5,319 ms 50,972 success 9, server error 1

What protects this target

HTTP 403, applied to proxied traffic only. Seventeen of 60 requests were refused and not one of them was unproxied.

Two observations from the study bear on the mechanism, and they do not fully agree. During compliance gating, azre.gov served its robots.txt normally to a US datacenter IP and returned HTTP 403 to a European residential IP -- recorded in DIMENSIONS.md as evidence that exit geography, not fingerprint, was the binding constraint on this host. But in the run itself the plain datacenter arms did better than the anti-detect arms (80% and 60% against 40% and 50%), which geography alone does not explain, and the US residential arm scored 90%. This run does not identify the rule, and the honest summary is that something about proxied requests draws a 403 here while direct requests do not.

The URL caveat is not a footnote. The intended target was azre.gov/public-database-search; the reconnaissance probe found it returns HTTP 404, so the homepage was substituted. A department homepage is static, cacheable and undefended by comparison with a licence search. The census that preceded this run exists because the earlier dry run made exactly this mistake across four of five targets and concluded no proxy was needed. This page reports what was measured: the homepage.

robots.txt status

The URL measured is allowed by the host's robots.txt.

Status allowed
File https://azre.gov/robots.txt
Crawl-delay not declared
Fetched as the benchmark's own User-Agent
Note robots.txt 1706 bytes via benchmark_ua

Limits of this result

  • The URL measured is the department homepage, not the licence lookup. The lookup path recorded in the target list returns HTTP 404 and no working search URL was measured. Nothing here describes the search.
  • Ten rounds per arm in a single three-hour window on 2026-08-18. Anti-bot behaviour drifts; a result older than a few months is a historical record, not a current one.
  • One exit geography. Every request left from a US datacenter host or a US exit, so nothing here describes how this target answers traffic from elsewhere.
  • The unproxied arm ran from a clean, previously unused cloud IP with no accumulated reputation. It is a best case a production scraper will not sustain.

Run date: 2026-08-18 · Full benchmark report · Harness and raw data · All 32 targets

The 60 rows behind this page are in results/six-arm/raw.csv; filter them with grep '^[^,]*,us_az_realestate,' raw.csv.


Test your own target

The result above describes one URL on one day. Paste your own URL into the live tester and see the real response before creating an account.

Test your target See plans