webscraper.io Review: Pricing, Limits, and When to Switch

Web Scraper is a point-and-click scraping tool from webscraper.io that runs as a free Chrome or Firefox extension, with a paid Web Scraper Cloud for unattended jobs. You build a sitemap: start URLs plus a tree of selectors that click, follow links, paginate, and extract text, images, tables, or attributes into rows.

Think of a sitemap as a recipe: the start URLs are where it begins, and each selector is a step that either moves the scraper or collects data.

Navigation selectors do the moving. Link selectors open detail pages, pagination selectors walk through numbered pages or "next" buttons, and element click selectors press "load more" buttons or scroll to reveal more content.

Data extraction selectors (text, image, table, HTML, and element attribute) do the collecting. Anything nested under a link selector runs on the page that link opens, which is how one sitemap turns a category listing into one row per product.

You click elements and the extension writes CSS selectors for you, in CSS and jQuery-style syntax rather than XPath. When a generated selector matches too much or too little, you edit it by hand, so keep a CSS selectors cheat sheet open while you build.

A typical listing-to-detail sitemap looks like this:

flowchart TD
    A["Start URL<br/>/category/books"] --> B["Pagination selector<br/>next page / page 2, 3, ..."]
    B --> C["Link selector<br/>each product card"]
    B -.->|"repeats on every listing page"| C
    C --> D["Product detail page"]
    D --> E["Text selector<br/>title"]
    D --> F["Text selector<br/>price"]
    D --> G["Element attribute selector<br/>image URL"]
    D --> H["Table selector<br/>specifications"]
    E & F & G & H --> I["One row per product<br/>export to CSV, XLSX, or JSON"]

Figure 1. A listing-to-detail sitemap: pagination walks the listing, a link selector opens each product page, and four data selectors produce one row per product.

Results export as CSV, XLSX, or JSON. In Cloud, parsers such as regex match, replace text, strip HTML, and virtual columns clean or derive values before export.

The extension runs inside your own browser, is free, keeps only the latest job's data in local storage, and stops when you close the browser. Web Scraper Cloud runs the same sitemaps on remote servers with scheduling, an API, webhooks, proxies, data-quality checks, and stored results, which is what recurring work usually needs.

Both use the same sitemap format, so you build and test in the extension, then import the sitemap into Cloud when the job needs to run on its own.

Table 1. The extension is where you build, and Cloud is where recurring jobs run.

Two rows explain why results can arrive in a different order. The extension works last-in, first-out, finishing one branch of the site before the next, while Cloud visits discovered URLs in pseudo-random order. Same data, different row order. Cloud also skips images, so test any page whose values depend on them.

The extension is free. As listed on webscraper.io in October 2026, Web Scraper Cloud has a 7-day free trial, then Project at $50 a month for 5,000 URL credits, Professional at $100 for 20,000, Scale from $200 with unlimited credits priced per concurrent scraper, and custom Enterprise plans.

Table 2. Web Scraper Cloud plans as listed in October 2026. Credits and concurrency, not features, separate the plans.

Prices and limits from webscraper.io's pricing page, October 2026. Check the vendor's site for current terms.

The plans share the scheduler, API, parsers, data-quality monitoring, and exports. What changes is monthly page volume, concurrency, and retention. Scale includes a free US datacenter proxy. Residential traffic was listed at $2.50 per GB in October 2026, which matters if your targets block datacenter IPs.

If price is the main constraint, see our comparison of free web scraping tools.

One URL credit is one page load, whatever the page contains. A sitemap that opens 40 category pages and 1,200 product pages uses about 1,240 credits per run, so a weekly schedule needs about 5,000 a month. Automatic retries of empty or failed pages and test scrapes up to 500 links are free.

To estimate a job, count every page the sitemap opens, multiply by runs per month, and compare the result with the plan allowance.

Table 3. What a job costs in URL credits. Detail pages drive the total.

Pages, not records, drive cost. If the listing already shows the fields you need, such as name, price, and stock, skip the detail pages. For the weekly catalog, that drops each run from 1,240 credits to 40.

Watch the margins, too. The weekly catalog in Table 3 leaves only 40 spare credits on Project, so a few extra category pages, or a month with five scheduled runs (6,200 credits), pushes it over. Size the plan for the largest month, not the average one.

Web Scraper is at its best on ordinary public sites with a clear listing-to-detail structure. It's free to build and test locally, handles pagination, clicks, and infinite scroll without code, exports to common formats, and lets a non-developer move the same sitemap into Cloud when the job needs scheduling.

Its strengths in practice:

• Zero setup. Install the extension and start building. Local runs need no account or server.
• Visual selection with live preview. You see what each selector matches before you scrape.
• Multi-level sitemaps. Listings, pagination, and detail pages live in one selector tree.
• Portable sitemaps. JSON import and export lets you share sitemaps, version them, and move them to Cloud unchanged.
• Built-in parsers. Cleanup happens in the tool, not in a spreadsheet afterward.
• Free test scrapes. Cloud tests up to 500 links before spending credits.
• Integration hooks. Cloud's API and webhooks hand finished jobs to other systems.

For a marketer tracking a few dozen competitor product pages, or an analyst pulling a directory once a quarter, that combination is hard to beat for the price.

Web Scraper strains when jobs get big, protected, or fragile. Local runs tie up your browser, Cloud caps a sitemap at 100 selectors and 10 minutes of continuous JavaScript actions per URL, social platforms including LinkedIn aren't supported, and every site redesign means someone fixing selectors by hand.

• Local runs need your browser. The browser and computer must stay open for the whole job, and only the latest job's data is kept.
• Cloud execution limits. A sitemap can have up to 20,000 start URLs and 100 selectors, and JavaScript actions such as clicking and scrolling can run for up to 10 continuous minutes per URL.
• Retention. Cloud keeps job data for 30 days on Project and Professional and 60 days on Scale. If you need history, schedule exports to your own storage.
• Concurrency. Project runs 2 scrapers at once and Professional runs 3, so busy schedules queue below Scale.
• Protected sites. webscraper.io itself says compatibility with sites that use advanced anti-bot protection varies, including those behind Cloudflare Turnstile. Scale includes only a free US datacenter proxy, and residential traffic is billed per GB.
• Social platforms. Cloud doesn't support data extraction from social media platforms, including LinkedIn.
• Logins. The extension uses your browser session, which raises account and terms-of-service questions (see the legal section).
• Selector maintenance. When a site changes its markup, someone fixes the sitemap by hand. It's the cost missing from the pricing page, and it grows with every site.
• Limited code-level control. You can't write custom retry logic, set arbitrary headers, or run your own parsing code.

Keep Web Scraper when a few people need structured data from a handful of ordinary public sites, refreshed daily or weekly, and someone is happy to maintain selectors. Look elsewhere when you need many sites, protected or JavaScript-heavy targets, guaranteed freshness, code-level control, or data nobody on your team wants to maintain.

The deciding question is usually not "can Web Scraper scrape this site?" but "who will keep this job working?" Find the row in the table that matches your situation.

Table 4. Choose a tool by job. The right tool depends on who maintains it.

If the first row describes you, Web Scraper is a sound choice and the rest of this guide is background. Already decided to move? Skip to the migration section below.

Web scraper tools fall into five groups: browser extensions like Web Scraper, no-code desktop or cloud apps such as Octoparse and ParseHub, AI extractors that read pages from a prompt, code libraries and frameworks like Scrapy and Playwright, and scraping APIs or ready-made data services that take over fetching, rendering, and maintenance.

Browser extensions. Web Scraper is the best-known example. They're cheap and visual, and their ceiling is your browser unless a cloud service sits behind them.

No-code desktop and cloud apps. Tools such as Octoparse and ParseHub offer visual workflow builders, site templates, and hosted runs, with the same maintenance question when sites change.

AI extractors. These return fields from a plain-language prompt instead of selectors. They help on messy layouts and need spot checks. Our roundup of AI web scraping tools covers the category in depth.

Code libraries and frameworks. Scrapy, Playwright, and similar libraries give developers full control, and full ownership of every fix.

Scraping APIs and data services. These take over proxies, rendering, retries, or parser upkeep, and return pages for your code or finished records on a schedule.

For the lightest possible option, Google Sheets' IMPORTXML function pulls values from simple static pages into a spreadsheet with no install.

If you want specific products compared and ranked within each group, see our ranked list of web scraping tools.

Write code when the scraping logic is part of a product, when you need tests, version control, and custom retries, or when sites change often enough that you want failures to raise alerts. Scrapy suits large crawls of static pages, and Playwright suits sites that need a real browser to render or interact.

Code makes a sitemap's rules testable: an assertion that every row has a price can alert someone before bad data reaches a dashboard.

Scrapy is a Python framework for fast crawls of static pages. Playwright drives real browsers for pages that build content with JavaScript. The web.dev rendering on the web explainer shows why such pages look empty to a plain HTTP request.

Code moves the work rather than removing it: selectors still break, but you find out faster.

Use a scraping API when your code should own parsing and storage but not proxies, browser rendering, retries, or blocks. You send a URL, optionally with rendering, location, or page actions, and get back HTML, JSON, or a screenshot, so the pipeline stays in your repo without running browser fleets.

It's the usual next step for teams that spend more time keeping requests working than parsing data. You can compare web scraping APIs by what they return, from a single page to a whole-site crawl.

If your sitemaps rely on element click selectors, scrolling, or waiting for content to appear, a browser API for clicks, waits, and screenshots is the closest equivalent: you describe the actions, and a hosted browser runs them and returns HTML, JSON, fields, or a screenshot.

Buy or subscribe to data when the sources are popular and the fields are standard, like products, prices, reviews, listings, or jobs, and nobody wants to own selectors. A structured-data API or a scheduled feed returns clean records, and the provider absorbs site changes, which is often cheaper than a person fixing sitemaps.

This is the most practical alternative to web scraping itself. A structured data API for 116 sources returns records with consistent fields, while WebScrapingAPI handles source access, extraction, parser maintenance, and quality monitoring.

If what you really pay Cloud for is the schedule, scheduled data feeds deliver recurring structured data with no sitemap to maintain.

Compare a feed's cost with your Cloud plan plus the hours spent fixing selectors. For one or two stable sites, Web Scraper usually wins. For a dozen changing sites, it often doesn't.

Export every sitemap as JSON first, because it records your start URLs, selectors, and page structure. Then write down the fields, the refresh schedule, and a sample of known-good rows. Rebuild that spec in the new tool, run both side by side for a cycle, and compare the outputs before switching off Cloud.

The sample rows tell you whether a difference comes from the new tool or the site. Download Cloud job history too, before retention expires. For the detailed process, follow our step-by-step guide to moving a Web Scraper sitemap to another tool.

Scraping public pages is generally lawful in many places, but legality depends on the data, the site's terms, and your jurisdiction. Respect robots.txt, avoid personal data you can't justify, don't scrape behind logins you aren't permitted to automate, and review what a browser extension can read before installing it.

Our guide on whether web scraping is legal goes deeper. Three points apply to Web Scraper in particular.

First, robots.txt. The Robots Exclusion Protocol (RFC 9309) is how sites tell crawlers which paths to avoid. It isn't access control, but honoring it is the baseline.

Second, logins. The extension can reach anything your account can see, and many sites' terms prohibit automated account access. Get permission first.

Third, the extension itself. Chrome's guide to extension permission warnings explains what a scraping extension can read and change. Review them, especially on a work browser.

This isn't legal advice. Talk to counsel about your use case.

Yes, the Chrome and Firefox extension is free for scraping inside your own browser. You pay only for Web Scraper Cloud, which adds scheduling, an API, proxies, and stored results. Cloud plans offered a 7-day free trial in October 2026, and test scrapes of up to 500 links don't use credits.

A URL credit is one page load in Web Scraper Cloud, no matter how many records the page holds. A listing page with 100 products costs one credit, while opening those 100 product pages costs 100. Automatic retries of empty or failed pages don't use extra credits.

Yes. Pagination and link selectors follow "next" buttons and numbered pages, and element click selectors press "load more" buttons or scroll to load more items. Build the selector tree so pagination sits above the data selectors, then preview a few pages before running the whole sitemap.

Usually, because it runs in a real browser that executes page scripts. In Web Scraper Cloud, a sitemap can run JavaScript actions such as clicking and scrolling for up to 10 continuous minutes per URL. Sites with aggressive bot protection may still fail, so test a target before committing.

The extension runs in your own browser session, so it can see pages you're logged into. That doesn't make it allowed: many sites forbid automated access to accounts in their terms. Web Scraper Cloud also doesn't support social media platforms, including LinkedIn, so check the target first.

Not as a dedicated feature. Cloud jobs use proxy rotation, retries, and failover to reduce interruptions, and webscraper.io says compatibility with protected sites varies by site. If CAPTCHAs block your target regularly, a scraping API or managed data service that handles access for you is usually more dependable.

It can scale for structured public sites, especially on the Scale plan with unlimited URL credits priced per concurrent scraper. Large programs across many changing or protected sites usually outgrow it, because selector upkeep, concurrency, and proxy costs grow faster than the plan price suggests.

It depends on the job. Other no-code apps suit visual workflows, AI extractors cut selector work, Scrapy or Playwright suit teams that want code, scraping APIs handle blocks and rendering, and data APIs or feeds remove maintenance entirely. Match the tool to who will maintain it.

Done maintaining sitemaps? Tell us the sources, fields, and schedule, and Managed Data runs collection, quality checks, and delivery for you.

Prefer to keep scraping in your own code? Scraper API starts with 5,000 free requests, JavaScript rendering and geotargeting cost no extra credits, and failed requests aren't billed.