Zenrows
A web access API that returns clean page content from sites that block ordinary scrapers
Zenrows is a web data access API that fetches pages your own code cannot reach. It rotates residential and datacenter proxies, renders JavaScript in a real browser, and works past anti-bot systems such as Cloudflare, DataDome, and PerimeterX, returning the page as Markdown, HTML, JSON, or a screenshot. It is sold on a shared credit balance across four products (Fetch, Extract, Browser Sessions, and Batch), starts with a permanent free tier of 5,000 credits a month, and is used by go-to-market and data teams that need live web content for enrichment, monitoring, and AI agents.
Overview
Every go-to-market team eventually hits the same wall. The enrichment vendor does not have the field you need, so someone writes a script to pull it off the company's own website, and the script works for a week before the target site starts returning a challenge page instead of HTML. Zenrows sells the layer that makes that script keep working: a single API endpoint that handles proxy rotation, browser fingerprinting, TLS characteristics, JavaScript execution, and retry logic, so the caller gets page content back rather than a 403.
The company restructured its product names in 2026 around what an operator actually wants to do. Fetch (formerly the Universal Scraper API) retrieves a URL and returns it as clean Markdown, HTML, or JSON. Extract, still in beta, turns a page into structured fields without you writing or maintaining CSS selectors. Browser Sessions (formerly the Scraping Browser) hands you a cloud-hosted real browser over CDP for logins, form fills, and multi-step flows. Batch submits thousands of URLs as one asynchronous job. All four draw from the same credit balance, which is the pricing decision that makes Zenrows easier to budget than vendors that meter each product separately.
The 2026 positioning is explicitly aimed at AI agents rather than only at scraper developers. Zenrows ships an official MCP server that installs into Claude Code, Cursor, VS Code, Windsurf, Zed, and JetBrains IDEs with one command, so an agent can request a live page in natural language and get back Markdown it can reason over. For a small go-to-market team, that is often the real purchase: not a scraping pipeline, but a reliable way for the automation they already run in Clay, n8n, or a Python script to read the open web without maintaining a headless Chrome fleet.
Zenrows is a small, independent Spanish company, founded in 2021 in Getxo and seed-funded once in 2022. It is not an enterprise-only vendor: the free tier requires no card, the first paid step is under twenty dollars a month, and self-serve checkout covers everything up to the Scale tier. The honest constraints are that credits are consumed unevenly (a protected page costs twenty-five times a plain one), that the dashboard assumes you are comfortable reading API responses, and that success rates on the hardest targets are high but never one hundred percent.
Best for
Technical go-to-market operators, data teams, and AI agent builders who need live content from sites that block ordinary requests, and who would rather pay per successful page than run and maintain their own proxy and headless browser infrastructure.
Not the right fit for
- Non-technical users who want a point-and-click scraper; Zenrows is an API first, and the dashboard is a usage console rather than a visual builder.
- Teams that need a scheduler, storage, and a workflow runtime around their scraping; Zenrows returns pages, it does not orchestrate or host the pipeline.
- Anyone who only needs a handful of well-known B2B fields; a contact and company enrichment vendor is cheaper and legally simpler than building it from raw pages.
- High-volume crawls of plain, unprotected HTML, where the premium anti-bot layer you are paying for adds nothing over a self-hosted crawler.
- Use cases that require scraping behind a login you do not own, or data whose collection your counsel has not signed off on; the tool makes access easy, not lawful.
How it works
- 1
You sign up, take an API key, and call a single endpoint with the target URL plus flags for what the page needs: JavaScript rendering, premium (residential) proxies, a country code for geotargeting, an output format, and optionally a set of browser instructions such as click, wait, scroll, or fill. Zenrows chooses the proxy, spins the browser if required, solves the fingerprint problem, retries on failure, and returns the page.
- 2
Credits are the meter, and the multiplier is the thing to understand before you budget. A plain request costs one credit. Adding JavaScript rendering costs five. Premium proxies cost ten. A protected page needing both costs twenty-five. Residential proxy traffic meters at 25,000 credits per gigabyte, and Browser Sessions charge five credits per minute plus bandwidth. Failed requests are not charged, which matters because failure is a normal part of scraping hard targets.
- 3
Extract is the answer to selector maintenance. Instead of writing CSS or XPath rules that break when the target redesigns, you ask for structured output and Zenrows returns JSON fields. It is in beta and coverage is not universal, so the practical pattern is to try Extract first and fall back to parsing the Markdown or HTML that Fetch returns when the site is not supported.
- 4
For anything stateful, Browser Sessions gives you a remote Chrome instance you drive with Playwright, Puppeteer, or Selenium over the standard protocol. That is how you handle a login wall, a multi-step search form, or a flow that only reveals data after three clicks. Batch covers the opposite case: a long list of URLs submitted once, processed asynchronously, and collected when the job finishes, so you are not managing concurrency yourself.
- 5
Concurrency, not just credits, is the second gate. The free tier allows five simultaneous requests, Build twenty, Launch fifty, Growth one hundred, and Scale two hundred. Exceeding your limit returns a 429 rather than queueing, and repeatedly hammering the API with failing calls can get your key temporarily blocked, so backoff logic belongs in your client from the first day.
Feature breakdown
23 features in 4 modulesFetch: page retrieval
The core endpoint, and the reason most accounts exist.- Anti-bot bypass
- Handles Cloudflare, DataDome, PerimeterX, and similar defenses through fingerprint, TLS, and header management rather than a single proxy trick, which is the difference between a page returning and a challenge screen.
- JavaScript rendering
- Runs the page in a real browser so single-page apps, lazy-loaded lists, and infinite scroll return their actual content rather than an empty shell.
- Markdown output
- Returns the page as clean Markdown with navigation and boilerplate stripped, which is the format an LLM can consume directly without a parsing step.
- HTML, JSON, and screenshot formats
- The same request can return raw HTML for your own parser, structured JSON where supported, or a full-page or viewport screenshot for visual monitoring.
- Browser instructions
- A JSON instruction set for click, wait, wait_for selector, scroll, and fill lets one stateless request perform a short interaction before the content is captured.
- CSS and XPath extractors
- Pass selectors with the request and get only the fields you asked for back, which keeps payloads small when you are pulling one price off a large page.
- Session handling
- A session identifier keeps the same IP across a series of requests for up to ten minutes, which is what multi-step flows on the same host require.
Proxies and geography
The network underneath, exposed as request parameters rather than a separate product to configure.- Residential proxy network
- Premium residential IPs for targets that reject datacenter ranges outright, billed at ten credits per request or 25,000 credits per gigabyte when used directly.
- Country geotargeting
- An ISO country code on the request routes through that market, which is how you see localized pricing, regional inventory, or a competitor's country-specific landing page.
- Automatic rotation and retries
- Failed attempts are retried across different IPs and fingerprints inside a single billed request, so your client code sees one call rather than a retry loop.
- Direct proxy endpoint
- The network is also usable as a standard proxy with username and password credentials, for tools that expect a proxy string rather than an API call.
- Success-only billing
- Credits are consumed on successful responses; failures are not charged, though 404 and 410 responses count as successful retrievals.
Extract, Browser Sessions, and Batch
The three products beyond simple retrieval, all drawing on the same credit balance.- Schema-free extraction (beta)
- Extract returns structured JSON from supported pages without you defining a schema, removing the selector maintenance that kills most in-house scrapers.
- Cloud browser over CDP
- Browser Sessions exposes a remote Chrome you drive with Playwright, Puppeteer, or Selenium, so existing automation code connects with a changed endpoint rather than a rewrite.
- Persistent logged-in sessions
- Sessions can hold cookies and authentication state across steps, which is what dashboards, portals, and account pages you legitimately control require.
- Asynchronous batch jobs (beta)
- Submit thousands of URLs as one job and collect results when it completes, instead of managing your own queue and concurrency ceiling.
- Per-minute browser metering
- Browser Sessions bill five credits per minute plus bandwidth, so a long idle session is expensive and short scripted runs are the economical pattern.
Developer surface and agent integration
How the API reaches the code and the agents that actually call it.- Official MCP server
- One command installs Zenrows into Claude Code, Cursor, VS Code, Windsurf, Zed, and JetBrains IDEs, letting an agent request live pages as a tool call.
- SDKs and CLI
- First-party Python, Node.js, and Go SDKs plus a command line client, so a proof of concept is a few lines rather than a proxy configuration exercise.
- No-code platform connectors
- Works from Make, n8n, and Zapier as an HTTP step, which is how most go-to-market teams wire it into an enrichment workflow without a developer.
- LangChain and LlamaIndex loaders
- Document loaders for the common AI frameworks feed retrieved pages straight into a retrieval pipeline or agent context.
- Usage dashboard and concurrency headers
- Response headers report remaining concurrency and the console shows credit burn by product, which is how you catch a runaway job before it eats the month.
- Top-up credits
- Paid plans allow up to four top-ups per billing period and unused top-up credits roll forward, so a one-off spike does not force a permanent tier upgrade.
Use cases
4 documentedGTM engineer building a custom enrichment column in Clay
The target account list needs a field no enrichment provider sells: whether each company's pricing page mentions a specific integration. Direct HTTP calls from the workflow get blocked on roughly a third of the domains.
A Zenrows Fetch call with JavaScript rendering replaces the raw HTTP step and returns Markdown for nearly every domain, so the column fills reliably and the list can be segmented on a signal no competitor has.
Ecommerce operator monitoring competitor prices
Three competitor sites sit behind anti-bot protection and serve different prices by country, and the existing scraper broke the last two times a target redesigned.
Requests run daily with premium proxies and a country code per market, Extract returns price and availability as JSON, and the team gets a per-market price sheet without a developer rewriting selectors each quarter.
Founder building an AI research agent
The agent needs to read live vendor documentation and pricing pages, but the built-in fetch tool returns challenge pages and stale cached copies for the sites that matter most.
The Zenrows MCP server is installed into the agent's toolset, pages come back as clean Markdown, and research output stops citing content the agent could not actually load.
Data team feeding a lead-scoring model
Scoring depends on technology and hiring signals scraped from tens of thousands of company sites each month, and the in-house proxy setup costs more engineering hours than the model does.
Batch jobs submit the URL list overnight on the Growth plan, the proxy layer becomes a line item rather than a maintenance burden, and the engineer who was babysitting headless Chrome goes back to model work.
Pricing
from Free (5,000 credits per month); paid plans from $19 per month, or about $16 per month billed annuallyCredit-based subscription with one shared balance across Fetch, Extract, Browser Sessions, and Batch. Credit cost varies by difficulty: one credit for a plain request, five with JavaScript rendering, ten with premium proxies, twenty-five for a protected page needing both, and 25,000 credits per gigabyte of residential proxy traffic. Plans also step up concurrency, which is often the binding constraint before credits are.
| Plan | Price | Includes |
|---|---|---|
| Free | $0 per month |
5,000 credits is roughly 5,000 plain pages or only 200 fully protected ones, so it is a genuine evaluation budget rather than a production allowance. |
| Build | From $19 per month (from about $16 billed annually) |
The realistic starting point for a small team running a recurring enrichment or monitoring job. |
| Launch | From $69 per month (from about $57 billed annually) |
|
| Growth | From $199 per month (from about $165 billed annually) |
|
| Scale | From $549 per month (from about $456 billed annually) |
An Enterprise tier above this is quoted, covering 12.5 million credits and up, 400 to 1,000 or more concurrent requests, an SLA, and a dedicated account manager. |
Billing notes
- The headline credit number is not the number of pages you get; a protected page costs twenty-five credits, so a 45,000 credit plan is 1,800 hard pages, not 45,000.
- Failed requests are not billed, but 404 and 410 responses are treated as successful retrievals and do consume credits.
- Annual billing is pay for ten months and get twelve, roughly a seventeen percent discount, applied on every paid tier.
- Unused credits roll over for one billing period on Build through Scale; the free plan resets with no rollover.
- Up to four top-ups are allowed per billing period and top-up credits roll forward, which is the intended relief valve for a one-off spike.
- Each named tier contains several credit steps, so the price you actually pay is usually above the tier's floor; budget from the step that matches your volume, not the marketing number.
- Concurrency is bundled with the tier and cannot be bought separately, so a job that needs parallelism may force an upgrade even when credits are sufficient.
Value assessment: Priced against the alternative, which is running your own residential proxy contract plus a headless browser fleet plus the engineer who maintains both, Zenrows is straightforwardly cheap at the low end. A team spending nineteen dollars a month to keep an enrichment workflow alive is making an obvious trade. The economics get harder to defend in two places: high-volume crawling of plain HTML, where you are paying an anti-bot premium you do not need, and heavy protected-page work, where the twenty-five credit multiplier compounds fast enough that Scale tier bills arrive sooner than the credit headline suggests. The single shared balance across all four products is a real advantage over vendors that meter each separately, and the permanent free tier makes it possible to measure your own credit burn on your own targets before committing to anything.
Strengths & limitations
Strengths
- Genuinely high success rates on Cloudflare, DataDome, and PerimeterX protected pages, which is the one thing this category is bought for.
- One shared credit balance across retrieval, extraction, browser sessions, and batch jobs, so you are not forecasting four separate meters.
- A permanent free tier with every feature unlocked, which lets you measure real credit consumption against your actual targets before paying.
- Markdown output and an official MCP server make it usable by AI agents without a parsing or integration layer of your own.
- Failed requests are not billed, which is the correct alignment of incentives in a category where failure is routine.
- Self-serve from free to roughly $549 a month with no sales call, unusual among vendors serving this technical a use case.
- Existing Playwright, Puppeteer, and Selenium code connects to Browser Sessions by changing an endpoint rather than being rewritten.
Limitations
- Credit multipliers make budgeting genuinely difficult: the same 45,000 credit plan is 45,000 plain pages or 1,800 protected ones, and most people discover the difference after their first month.
- Extract is still in beta and only covers supported site patterns; on anything else you fall back to parsing HTML yourself, which is the maintenance burden the feature exists to remove.
- Premium and JavaScript-rendered calls do not automatically fall back to a cheaper mode when they fail, so your client has to implement the escalation logic itself.
- Sending too many failing requests in a short window can get your API key temporarily blocked by the platform's own abuse protection, which surprises teams that ship without backoff.
- The dashboard is a usage and key console, not a scraping workbench; the product is marketed as low-code but is unambiguously developer-facing in practice.
- No scheduling, storage, or workflow layer; Zenrows returns pages and expects something else to own the pipeline, unlike platform competitors that include a runtime.
- Concurrency is fixed to the tier, so parallelism needs can force an upgrade independently of credit consumption.
- It is a small independent company with a single seed round, which is a supplier-concentration consideration if a core revenue process depends on it.
Head-to-head comparisons
6 alternativesZenrows vs Apify
from Free with $5 of monthly platform credit; paid plans from $29 per month ($26 billed annually)Different layers of the same job. Apify is a platform: a marketplace of prebuilt scrapers, a scheduler, storage, and a runtime you deploy code into. Zenrows is the access layer only, one endpoint that returns a page. If you want a working LinkedIn or Google Maps scraper today with somewhere to put the results, Apify is the shorter path. If you already have a pipeline and the only failing part is getting the HTML back, Zenrows is the cheaper and more focused fix, and plenty of teams run an Apify actor that calls Zenrows underneath.
Full Zenrows vs Apify comparisonZenrows vs Firecrawl
from Free (1,000 credits per month); paid plans from about $16 per month billed annually (about $19 billed monthly)The closest direct comparison for AI use cases, since both return clean Markdown built for LLM consumption. Firecrawl leans toward crawling and site-wide ingestion for retrieval pipelines, with strong crawl and map primitives. Zenrows leans toward reaching pages that actively resist being read, with a deeper proxy and anti-bot layer and a real cloud browser. Choose Firecrawl for documentation and content ingestion at breadth; choose Zenrows when the targets fight back.
Full Zenrows vs Firecrawl comparisonZenrows vs Browse AI
from Free (50 credits per month, 2 websites); paid plans from $19 per month billed annuallyOpposite ends of the technical spectrum. Browse AI is a point-and-click recorder that a non-developer can train on a page in a few minutes, with scheduling and change monitoring included. Zenrows expects you to write the request. For a marketer tracking twenty competitor pages, Browse AI is obviously right. For anything programmatic, high-volume, or aggressively protected, Browse AI's abstraction becomes the ceiling and Zenrows becomes the answer.
Full Zenrows vs Browse AI comparisonZenrows vs PhantomBuster
from $69 per month (Start), or $56 per month billed annuallyNot really competitors, but frequently confused. PhantomBuster automates specific platforms, mostly LinkedIn, with prebuilt actions and account-session handling. Zenrows is target-agnostic infrastructure with no knowledge of any particular site's semantics. A go-to-market team usually ends up with PhantomBuster for social platform actions and Zenrows for everything on the open web that blocks a plain request.
Full Zenrows vs PhantomBuster comparisonZenrows vs Clay
from Free plan; paid from $149/moComplementary far more often than competitive, and the pairing is common enough to be a pattern. Clay is the workflow and enrichment surface where the list lives; Zenrows is what Clay calls when a needed field only exists on a page that blocks HTTP requests. Clay's own scraping steps handle easy targets well, so the honest rule is to add Zenrows only when Clay's built-in retrieval starts returning empty or challenged pages.
Full Zenrows vs Clay comparisonZenrows vs Common Room
from About $2,500 per month billed annually (roughly $30,000 per year) on the Essential planBoth feed go-to-market signal, from opposite directions. Common Room ingests and unifies signals from communities, social platforms, and product usage through maintained integrations, then scores and routes them. Zenrows gives you raw access to any page and leaves interpretation entirely to you. Common Room is the buy decision for a team that wants signal without engineering; Zenrows is the build decision for a team whose differentiating signal is not available from any vendor.
Full Zenrows vs Common Room comparisonImplementation & onboarding
- Setup time
- Minutes for a first successful request: sign up, copy the API key, paste the sample call. Getting a specific hard target working reliably takes longer, usually a session of testing which combination of JavaScript rendering, premium proxies, country, and wait conditions that particular site requires.
- Learning curve
- Low for anyone comfortable with an HTTP client, meaningful for anyone who is not. The concepts that take longest are not the API but the domain: why a request needs residential IPs, what a fingerprint check is doing, and how to write backoff logic that does not trip the abuse protection.
- Onboarding
- Entirely self-serve. Documentation includes copy-paste examples per language, a request builder in the dashboard, and per-target guidance. Support is community-only on Free, email on Build, chat on Launch, and priority above that.
- Migration notes
- Moving from a self-hosted scraper generally means replacing the request layer and leaving parsing untouched, which is a small change. Moving from another scraping API means retranslating parameter names and, more importantly, rebenchmarking cost, since credit models across vendors are not comparable without running your own URLs through both. Run the free tier against your real targets first; success rates quoted in marketing material are not transferable to your site list.
Platform, API & security
- Platforms
- REST APIPython, Node.js, and Go SDKsCommand line clientProxy endpoint (username and password)Chrome DevTools Protocol for Browser SessionsMCP server for Claude Code, Cursor, VS Code, Windsurf, Zed, and JetBrains
- API
- The API is the product. A single Fetch endpoint takes a URL plus parameters for JavaScript rendering, premium proxies, country, output format, CSS extractors, session identifier, and a JSON browser instruction set. Browser Sessions are driven over CDP by Playwright, Puppeteer, or Selenium; Batch accepts asynchronous URL lists. Response headers report remaining concurrency.
- Compliance
- GDPRSOC 2ISO 27001
- Data residency
- Operated from Spain under EU data protection law; the proxy network itself spans 250 or more countries by design.
- SSO
- Available on Enterprise agreements.
- Security notes
- SOC 2 and ISO 27001 attestations are presented as Enterprise-tier commitments alongside the SLA rather than as blanket claims on self-serve plans; confirm scope in writing if you need them contractually. The platform enforces its own abuse protection and will temporarily block an API key that sends a high volume of failing requests. Zenrows is an access tool, so responsibility for what you retrieve, whether it is personal data, and whether the target's terms permit it remains entirely with the customer.
Support & resources
- Channels
- Community support (Free plan)Email support (Build and above)Live chat (Launch and above)Priority support (Growth and above)Dedicated account manager (Enterprise)
- Documentation
- Detailed developer documentation with per-language examples, a parameter reference, and a large public library of scraping guides that ranks well enough to serve as the company's main acquisition channel.
- Community
- Reviewers consistently describe support as responsive and technically competent, which is one of the more reliable positives in an otherwise mixed review corpus. There is no large public user forum; the company's own blog and documentation carry most of the knowledge.
Company
- Founded
- 2021
- Headquarters
- Getxo, Spain
- Ownership
- Independent, venture-backed
- Founders
- Aurken Bilbao, Ander Rodriguez
- Employees
- Small team, headcount not publicly disclosed
- Funding
- One disclosed seed round of about €1.1 million in 2022, led by 4Founders Capital with Athos Capital and Bynd Venture Capital.
Funding history
| Round | Amount | Year | Notes |
|---|---|---|---|
| Seed | About €1.1 million | 2022 | Led by 4Founders Capital, with Athos Capital and Bynd Venture Capital participating. |
Timeline
- 2021Founded in Getxo, Spain by Aurken Bilbao and Ander Rodriguez, initially bootstrapped.
- 2022Raises a seed round of about €1.1 million led by 4Founders Capital.
- 2023Residential proxy network and the Scraping Browser are added alongside the Scraper API, making it a full stack rather than a single endpoint.
- 2024Anti-bot coverage broadens to the major commercial defenses and the free tier becomes permanent rather than a trial.
- 2025Markdown output and AI framework loaders arrive, repositioning the API toward LLM and agent consumption.
- 2026Products renamed and restructured into Fetch, Extract, Browser Sessions, and Batch on one shared credit balance, with an official MCP server for coding agents and IDEs.
Integrations
- Model Context Protocol (Claude Code, Cursor, VS Code, Windsurf, Zed, JetBrains)
- Playwright
- Puppeteer
- Selenium
- LangChain
- LlamaIndex
- n8n
- Make
- Zapier
- Clay
- Python SDK
- Node.js SDK
- Go SDK
Frequently asked questions
12 questionsWhat is Zenrows used for?
Fetching web pages that ordinary code cannot reach. It rotates residential and datacenter proxies, renders JavaScript in a real browser, and works past anti-bot systems such as Cloudflare, DataDome, and PerimeterX, returning the page as Markdown, HTML, JSON, or a screenshot. Typical uses are enriching account lists with data no vendor sells, monitoring competitor prices, and giving AI agents reliable access to live pages.
How much does Zenrows cost?
There is a permanent free plan with 5,000 credits a month and five concurrent requests. Paid plans start at $19 a month (about $16 billed annually) for the Build tier with 45,000 credits, then Launch from $69, Growth from $199, and Scale from $549, with an Enterprise tier quoted above that. Each named tier contains several credit steps, so the price you pay is usually above the tier floor.
How do Zenrows credits work?
Every product draws from one shared monthly balance, and the cost per request depends on difficulty. A plain request costs one credit, JavaScript rendering costs five, premium proxies cost ten, and a protected page needing both costs twenty-five. Residential proxy traffic meters at 25,000 credits per gigabyte, and Browser Sessions cost five credits per minute plus bandwidth. This is why a 45,000 credit plan can mean 45,000 pages or 1,800 pages depending entirely on your targets.
Is Zenrows free?
There is a free plan rather than a free trial, and it does not expire or require a card. It gives 5,000 credits a month, five concurrent requests, and access to every feature, with community support only and no credit rollover. It is designed for measuring your own credit burn against your own target sites before you pay.
Does Zenrows charge for failed requests?
No. Credits are consumed only on successful responses. The one thing to note is that 404 and 410 responses count as successful retrievals and do consume credits, since the page was reached and the server answered.
Can Zenrows get past Cloudflare?
That is its core capability, and independent reviewers put its success rate on protected targets among the highest in the category. It handles fingerprinting, TLS characteristics, headers, and proxy selection rather than relying on any single technique. No vendor achieves one hundred percent, and premium calls do not automatically retry as cheaper standard calls when they fail, so your client should implement its own escalation logic.
Zenrows vs Firecrawl: which should I pick?
Both return clean Markdown for LLM use. Firecrawl is stronger at crawling entire sites and building retrieval corpora at breadth. Zenrows is stronger at reaching individual pages that actively block access, with a deeper proxy layer and a real cloud browser. Pick Firecrawl for documentation and content ingestion; pick Zenrows when your targets fight back.
Does Zenrows work with AI agents and MCP?
Yes. There is an official MCP server that installs into Claude Code, Cursor, VS Code, Windsurf, Zed, and JetBrains IDEs with one command, letting an agent request live pages as a tool call and receive Markdown it can read directly. There are also LangChain and LlamaIndex document loaders for retrieval pipelines.
Do I need to be a developer to use Zenrows?
Effectively yes. It is marketed as low-code and it can be called as an HTTP step from Make, n8n, or Zapier, which puts it within reach of a technical operator. But there is no visual scraper builder, the dashboard is a usage and key console, and diagnosing why a specific target fails requires reading responses. Non-technical users are better served by a point-and-click tool.
Was Zenrows renamed or acquired?
Not acquired. It remains an independent Spanish company founded in 2021 with a single disclosed seed round. Its products were renamed in 2026: the Universal Scraper API became Fetch, the Scraping Browser became Browser Sessions, and Extract and Batch were added. Older reviews and comparison articles still use the previous names and, more importantly, still quote the older Developer and Startup pricing tiers that no longer exist.
Is web scraping with Zenrows legal?
The tool does not change the legal question. Collecting publicly available data is broadly permitted in many jurisdictions, but a target's terms of service, copyright, database rights, and data protection law all still apply, and personal data collected in the EU or UK brings GDPR obligations onto you as controller. Zenrows is a processor of the request, not a shield; treat the legal review as your own work.
What is the concurrency limit on Zenrows?
Five simultaneous requests on Free, twenty on Build, fifty on Launch, one hundred on Growth, two hundred on Scale, and four hundred or more on Enterprise. Exceeding your limit returns a 429 rather than queueing the request, and sustained failing traffic can trigger a temporary block on your API key, so backoff belongs in your client from the start.
Editorial verdict
Zenrows does one thing that is hard and does it well: it gets the page back. For a go-to-market team whose differentiating data lives on sites that reject plain HTTP requests, that is worth far more than the nineteen dollars the first paid tier costs, and the permanent free plan means you can prove it against your own targets before spending anything. The 2026 restructuring into Fetch, Extract, Browser Sessions, and Batch on a single shared credit balance is a real improvement in both clarity and budgeting, and the MCP server makes it one of the more practical ways to give an AI agent honest access to the live web. The reservations are specific rather than generic. Credit multipliers mean the headline allowance tells you almost nothing until you know your target mix, Extract is still beta and falls back to hand parsing on unsupported sites, and there is no scheduling or storage layer, so Zenrows is a component and not a pipeline. Buy it as infrastructure, benchmark it on the free tier against the sites you actually care about, and keep a fallback plan for the targets where nothing reaches one hundred percent.
Written by the SaaSTracker editorial team. Awards, when shown, are judged against the published criteria in our methodology.