Welcome to the latest edition of The Web Scraping Insider, by the ScrapeOps.io team. 

With this newsletter, get a zero-fluff roundup of the most interesting content and news from the world of professional web scraping straight to your inbox every month.

In this month’s issue, we look at:

We hope you enjoy it, and we’re always open to your feedback and suggestions.

When it comes to proxy providers, the data says otherwise.

💰 Price has almost no relationship with performance.

On the same domain, a $19 provider can outperform one costing $300 or more. Paying more does not reliably buy you a higher success rate or lower latency.

🌍 Change the domain and the winner changes.

The best provider for Amazon might struggle on Etsy, Booking.com, or LinkedIn. No provider consistently performs best across every website.

📈 Change your request volume and it changes again.

A provider offering the best value at 10,000 requests can become far more expensive at 100,000 or one million requests. The rankings shift depending on the pricing plan and scale of your workload.

👀 Insider Take

There is no universally "best" proxy provider.

There is only the best provider for:

🎯 Your target websites
📊 Your request volume
⚡ Your performance requirements

Generic rankings, brand reputation, and expensive pricing plans cannot tell you which provider will work best for your use case.

The only reliable answer is to test providers against the websites you actually scrape, at the volume you actually require.

🧪 Test first. Measure the success rate, latency, and real cost per successful request. Then choose. Test proxy providers against your own target with the ScrapeOps Proxy Tester

👉 Read the full proxy provider analysis

July brought a major legal win, a $130M vote of confidence, and two clear warnings that the residential proxy market's Wild West era is ending.

⚖️ Google's DMCA case against SerpApi was dismissed.

Google argued that bypassing its anti-bot systems amounted to circumventing copyright protections. The court rejected its case, although some claims involving copyrighted content could still be amended.

👀 Insider Take: This is a very positive sign for the web scraping landscape. It pushes back against attempts to use copyright law to control access to publicly available web pages.

💰 Oxylabs secured its first outside investment.

After continuing to grow over the last few years (reported $350M in ARR) Oxylabs received a $130M investment from Warburg Pincus, valuing the company at $3.6B. The investment will support product development, network expansion, partnerships, and acquisitions.

👀 Insider Take: This is a huge vote of confidence in the web scraping industry and its future growth. We are still in the early stages of the AI-driven demand for fresh web data, and investors increasingly recognize scraping infrastructure as a critical part of the AI stack.

🚨 The FBI seized domains connected to NetNut and its sourcing network.

The seized domains included NetNut, ProxyJet, and DiViNetworks, amid allegations that the residential network was being used for fraudulent activity.

That does not establish that NetNut knowingly participated. However, it highlights how difficult it can be to control customer activity across large networks of users and proxy resellers.

👉 Check out The Web Scraping Clubs analysis for a deeper take.

👀 Insider Take: Proxy providers will place far more scrutiny on both where their IPs come from and how customers use them.

Expect more KYC checks, tighter restrictions on residential proxy access, and greater scrutiny of resellers. Providers will increasingly want direct control over their networks and the customers using them.

📺 LG is removing residential proxy functionality from smart TV apps.

Researchers examined 6,038 smart TV apps and found residential proxy SDKs in 42.5% of LG webOS apps and 26.5% of Samsung Tizen apps.

LG has now told developers to remove the functionality or face suspension. It also plans to strengthen how it vets apps containing residential proxy SDKs.

👀 Insider Take: This is likely a sign of what is to come. Device producers and platforms will apply greater scrutiny to how residential IPs are sourced, potentially constricting supply across the market.

That will reward providers that have built their own legitimate networks, can prove meaningful user consent, and maintain control over how their IPs are acquired. With AI increasing demand while scrutiny limits supply, high-quality residential IPs could become scarcer and more expensive.

The residential proxy market's Wild West phase is coming to an end. The providers that can prove where their IPs come from and control how they are used will be best positioned for what comes next.

We hosted a Reddit AMA with the team behind CloakBrowser to dig into where browser detection is heading.

The biggest shift: anti-bot systems are moving from checking individual fingerprint values to judging the entire session.

Older systems looked for obvious leaks like navigator.webdriver, suspicious canvas values or mismatched browser properties.

Modern systems combine hundreds of signals and ask: does this browser look normal as a whole?

You can pass every individual check and still fail because the browser, OS, hardware, network and automation layer don't tell the same story.

Behaviour matters too. Sites can watch how you arrived, whether text was typed or inserted instantly, and whether your actions contain human-like pauses.

A perfect fingerprint behaving like a robot is still a bot.

Other insights:

The new stealth stack is:

Coherent browser + clean IP + plausible navigation + human-like behaviour.

Before rebuilding your browser stack, run the boring test first: can stock Chrome access the target through the same proxy?

👉 Read the full CloakBrowser AMA.

That's it for this #9 newsletter.

Ian from ScrapeOps.