Evidence register · Public record
"Ethically sourced", "verified every 5 minutes", and "99.9%+ uptime" are useful only when they can be checked. Follow the evidence from residential consent to research data, network health, and published policy.
Databay states that its residential supply is provided through opt-in partner applications: participants are shown a consent flow describing bandwidth sharing before traffic is enabled. This page is Databay's current sourcing disclosure, not an independent certification of every endpoint. Prospective customers with contractual sourcing requirements should request the applicable diligence materials and terms before purchasing.
We accept residential supply only from partners that represent it is gathered through applications with a visible consent step, never from malware, rootkits, or hidden installers. Because consent is revocable, pool membership fluctuates naturally. That churn is the cost of doing this honestly.
The demand side is policed too. This page is the long-form explanation behind every "ethically sourced" claim elsewhere on the site; these are the enforcement points behind it:
We publish original measurements, not aggregated marketing claims. The centerpiece is our residential IP reputation study: we took 1,000 residential IPs across 25 of the world's largest residential ASNs and ran every one of them through six free IP-reputation databases, plus a Cloudflare-Free cross-verification origin.
The malice-focused feeds (Tor exit list, Spamhaus DROP, ASN-drop) flagged none of the sample; Spamhaus ZEN listings were dominated by PBL policy codes: the "this is a residential line" designation, not a malice verdict. The same open-methodology standard applies to the free proxy list: every entry is re-probed every 5 minutes, and stale entries are dropped rather than left to inflate counts.
Verify our numbers yourself:
Every guide, benchmark and dataset note on this site ships under one collective byline: the Databay Research Team. The engineers behind that byline work with the network they write about - the same probe pipelines, reputation scans and free-proxy dataset that customers query - and every published number links back to the method that produced it. No individual personas, no borrowed credentials: the byline is the team, and the proof is the published methodology.
Engineering & Research
The same group that runs the probe pipelines and the free proxy list dataset researches, writes, and maintains every guide on this site. Where a page states a number, the harness that produced it is documented on this page or linked from the article itself.
The working output is public: the technical guides, the IP-reputation study, and the probe-pipeline methodology. Corrections and factual disputes go to hello@databay.com. We update the article and its modified date rather than quietly editing a number.
What must be executed for real before it ships, how dates behave, and what happens when we get something wrong.
Corrections: factual disputes go to hello@databay.com. When a report holds up, we correct the article and update its modified date rather than quietly editing a number.
We publish real-time network health at status.databay.com: uptime, incident history and per-service availability. The network targets 99.9%+ uptime; the status page is where that figure is continuously kept honest.
Nothing is gated behind a sales call; the complete policy set is published:
Don't take our word for any of this: the third-party surfaces are linked directly, unfiltered.
Setup guides, benchmarks, and research notes on proxies and web data collection.
A 403 Forbidden error means the server understood your request and refused it. What causes it, and the fix for visitors, developers, and site owners.
HTTP 429 means you sent requests faster than the server allows. What triggers it, how long it lasts, and the fix for users, developers, and site owners.
Captive.apple.com is how Apple devices detect Wi-Fi login pages. Why it pops up, what the Success page means, and fixes when portals stick.
Read the policies, download the dataset, and check live status before putting the network itself to the test.
Every register entry above links to its primary source.