Screaming Frog Free vs Paid: What the 500 URL Limit Actually Blocks

Screaming Frog's free version crawls 500 URLs per crawl. That number gets quoted everywhere and it is correct, but it is the smaller half of the restriction. The license also gates crawl configuration, saving and reopening crawls, JavaScript rendering, custom extraction, and every API integration. Here is exactly what the free tier does, what a license costs, why paying is often the right answer, and what to do when the budget says no.

Before anything else: Screaming Frog SEO Spider is an excellent tool. It has been the default technical SEO crawler for well over a decade, it is fast, it is accurate, and the team behind it has spent years fixing the kind of edge cases that only turn up after millions of real crawls. We build a competing crawler and we still say that without hesitation. This post is about the shape of the free tier, not a case against the product.

We checked the figures below on screamingfrog.co.uk on September 29, 2026. Prices and feature lists change, so verify on their site before you buy.

What the Free Version Actually Includes

The free version downloads and runs without a sign-up or an email capture, which is more generous than most vendors manage. On the current pricing page, these are the features marked as available without a license:

• Find Broken Links, Errors & Redirects
• Analyse Page Titles & Meta Data
• Review Meta Robots & Directives
• Audit hreflang Attributes
• Discover Exact Duplicate Pages
• Generate XML Sitemaps
• Site Visualisations

That is a real technical audit. Broken links, redirect chains, title and description problems, noindex tags where they should not be, hreflang mismatches, exact duplicate pages: for a small site, those seven items cover most of what an audit would find anyway. If you have a 200 page brochure site and you want to know what is broken, the free version will tell you, today, for nothing.

The 500 URL Limit in Practice

The cap is 500 URLs per crawl, and it applies to each crawl individually rather than to your month or your account. You can run as many 500 URL crawls as you like.

Sites that fit comfortably inside 500 URLs: a local business site, a consultancy site, a documentation site with a few dozen pages, a blog with a couple of hundred posts. Sites that will hit the wall on the first run: any ecommerce catalog, anything with faceted navigation, anything with paginated archives, most news publishers, and any site with a large tag or category tree. A WordPress blog with 300 posts can easily produce 2,000 crawlable URLs once you count author archives, date archives, tag pages and paginated series.

There is a second-order effect that catches people out. The SEO Spider's default configuration crawls images, CSS and JavaScript files, and checks external links. On a paid license you would turn those off to focus the crawl on HTML. Without a license you cannot, because the configuration options are the part that is locked. So the free crawl is both capped and unfocused: you get the default scope, whatever that means for your site, and you cannot narrow it.

The Restrictions Nobody Mentions

Search results for "screaming frog free version limit" almost always stop at 500 URLs. The pricing page's own comparison table is blunter than that. Here is what carries a cross in the free column:

• Crawl Configuration. Include and exclude rules, crawl depth limits, speed and thread control, user agent switching, custom robots.txt, and the toggles for images, CSS, JS and external links.
• Save & Open Crawls. Close the window and the crawl is gone.
• JavaScript Rendering. Free crawls see the raw HTML only.
• Scheduling. No automated or recurring crawls.
• Crawl Comparison. No before-and-after diffing between two crawls.
• Custom Source Code Search, Custom Extraction and Custom JavaScript. No XPath or regex scraping, no custom snippets.
• Google Analytics, Search Console and PageSpeed Insights integrations. No traffic or Core Web Vitals data alongside your crawl.
• Crawl with OpenAI & Gemini. The LLM integrations are license-only.
• Near Duplicate Content, Structured Data validation, Spelling & Grammar Checks, Segmentation. All license-only.

Screaming Frog's own documentation says the same thing in plain text: without a license you are "limited to crawling a maximum of 500 URLs each crawl, crawls can't be saved, and advanced features and the configuration are restricted."

Which Restriction Hurts Most

In our experience it is saving crawls, not the 500 URL cap.

A 500 URL crawl finishes in under a minute on a small site, so re-running it is cheap. Losing it is a different problem. Every real audit involves going back to the data: a developer asks which pages had the redirect chain, a client questions a finding, you want to check whether the fix worked. With saved crawls you reopen the file. Without them, every question means a fresh crawl, and if the site changed in between you are comparing two different things without knowing it.

Crawl Comparison being license-only compounds this. The standard technical SEO workflow is crawl, fix, crawl again, diff. The free version can do the first and third steps but nothing to connect them.

The second sharpest restriction is JavaScript rendering. If your site is React, Vue, Angular, or any of the modern meta-frameworks that render client-side, a raw HTML crawl can report a nearly empty site: no internal links, no titles, no content. That is a rendering artifact rather than a site problem, and on the free tier there is no way to tell the two apart from inside the tool.

What a License Costs

As of September 2026, a Screaming Frog SEO Spider license is $279 per year, which is £199 in sterling and €245 in euros. Licenses are per user: their site states plainly that "2 users will require 2 individual licences."

Bulk pricing brings the per-seat cost down:

Licenses	USD per user / year	GBP per user / year
1 to 4	$279	£199
5 to 9	$265	£189
10 to 19	$249	£179
20 or more	$235	£169

One detail worth checking against your own situation: the paid crawl limit is described as unlimited with a caveat that the real ceiling depends on the memory and storage you allocate. That is honest of them, and it is true of every desktop crawler. A 5 million URL crawl needs database storage mode and a machine with the RAM to match.

Why Paying Is Often the Right Call

For a working SEO consultant, $279 a year is roughly $23 a month. One recovered client issue pays for it. The specific things you get for the money:

• A decade and a half of edge cases already handled. Odd redirect loops, malformed canonicals, hreflang return-tag logic, weird server behavior under load. This is the part that is genuinely hard to replicate and the reason the tool is trusted.
• The wider ecosystem. The Log File Analyser is a separate product and remains one of the few practical ways to see what search engine bots actually requested from your server. Pairing crawl data with log data answers questions no crawler alone can.
• Real support. A team that answers, plus thorough documentation and tutorials. When a crawl behaves strangely the day before a client call, that has value.
• Windows and Mac desktop convenience. Download, install, crawl. No server, no dependencies, no maintenance.
• Everything in one interface. Search Console and Analytics data sitting in the same table as your crawl saves a lot of spreadsheet work.

If you audit client sites for a living, buy the license. We would tell you that directly.

About Searches for a "Free Screaming Frog License"

A share of the traffic for these queries is people looking for a cracked license key or a patched build. Let us be straightforward about it.

Running cracked software while handling client data is a poor trade. A patched crawler is an unsigned binary from an anonymous source, running on the machine that holds your client credentials, your Search Console access and your export files. Crawlers are an especially bad category to take that risk with, because their normal behavior is making thousands of outbound network requests, which makes anything extra very hard to spot. Add to that the licensing exposure if you are billing clients for work done on unlicensed software, and no working update path, since a patched build cannot safely take vendor updates.

The honest options are three: use the free version within its limits, buy the license, or use a tool that is free because its author intended it to be. The third option exists, which is what the rest of this post is about.

Where LibreCrawl Fits

LibreCrawl is a crawler we build and release under the MIT license. It is free because that is the license, rather than as a trial or a tier. You host it yourself, it crawls unlimited URLs, it renders JavaScript through Playwright, and you drive it from a browser.

For the specific problem this post describes, a site too big for 500 URLs and a budget that will not stretch to $279, it addresses the parts that the free Screaming Frog tier holds back:

• No URL cap. Crawl size is limited by your hardware rather than a license tier.
• JavaScript rendering included. Playwright with a choice of Chromium, Firefox or WebKit, so client-rendered sites crawl properly.
• Crawls persist. Data is written to a database as the crawl runs, which gives you crash recovery on long crawls and a history dashboard afterwards.
• Full configuration. Include and exclude rules, depth and speed control, and scope settings are all available, because there is no tier to gate them behind.
• PageSpeed Insights integration for Core Web Vitals next to crawl data.
• A REST API (documented at /api/docs/) so crawls can run from cron and results can feed a dashboard.
• A plugin system for adding your own analysis tabs in JavaScript.
• Self-hosted data. Crawl data stays on machines you control, which matters for staging sites, intranets and client NDAs.

The honest costs: you need somewhere to run it (a modest VPS or a spare desktop is plenty), you will run pip install and read a README, support is GitHub issues rather than a support desk, and it has years less accumulated edge-case handling than Screaming Frog does. It also does not produce client-ready PDF reports or explain issues to you the way a mature commercial tool does.

Free Screaming Frog vs Paid vs LibreCrawl

Capability	Screaming Frog (free)	Screaming Frog (licensed)	LibreCrawl
Price	Free	$279 / £199 / €245 per user per year	Free (MIT license)
URLs per crawl	500	Unlimited (hardware bound)	Unlimited (hardware bound)
Crawl configuration	Locked	Full	Full
Save and reopen crawls	No	Yes	Yes, database persistence plus history
JavaScript rendering	No	Yes (Chrome based)	Yes (Playwright: Chromium, Firefox, WebKit)
Crawl comparison	No	Yes	Via crawl history
Custom extraction (XPath, regex)	No	Yes	Via plugin system
Search Console / Analytics data	No	Yes	No
PageSpeed / Core Web Vitals	No	Yes	Yes (PageSpeed Insights)
Scheduling	No	Yes	Via API and cron
Structured data validation	No	Yes	No
Log file analysis	No	Separate product	No
Client-ready PDF reports	No	Export options	No
REST API	No	Limited	Full, at /api/docs/
Runs on	Windows, Mac, Ubuntu desktop	Windows, Mac, Ubuntu desktop	Self-hosted web app (Python Flask)
Support	Docs and community	Dedicated team	Community (GitHub)

The licensed column wins on breadth, and the table shows why the license sells. The free column and the LibreCrawl column are the two options for people who are not buying, and they fail in different places.

Working Around the 500 URL Limit Without Paying

A few approaches people use, with the honest limitations of each:

Crawl in sections. Point separate free crawls at separate subfolders. This half-works. You cannot set include or exclude rules on the free tier, so you are relying on starting URLs and whatever the default crawl scope does, and you will re-crawl shared navigation on every pass. You also end up with several unsaved crawls that you cannot merge or diff.

Crawl your XML sitemap in list mode. Useful for checking status codes on your canonical set of URLs, still capped at 500, and it tells you nothing about internal linking or orphan pages.

Sample a section and extrapolate. Fine for spotting template-level problems, since a broken title tag pattern shows up in the first 50 URLs of a template. Useless for anything that requires completeness: broken link counts, orphan page discovery, duplicate detection at scale.

Use an unlimited free crawler. Beyond LibreCrawl there are other options with different trade-offs, and we have written up the field in our roundup of the best Screaming Frog alternatives and a more direct Screaming Frog alternative comparison. Some are cloud SaaS with their own free tiers and their own caps; some are command line tools. Read the caps before you commit an afternoon.

Deciding What to Do Next

• Site under 500 URLs, static HTML, one-off audit. Use the free version of Screaming Frog. It is the fastest route to an answer and it costs nothing.
• Paid client work, any site size. Buy the license. At $279 a year, the saved crawls, configuration and integrations pay for themselves inside a month of billable work.
• Site over 500 URLs, budget of zero. Run LibreCrawl. Unlimited URLs and JavaScript rendering without a license key, at the cost of hosting it yourself.
• Crawl data feeding other systems. An API-first self-hosted crawler fits a pipeline better than a desktop app does, whichever tool you use for interactive auditing.
• Client data on your machine and a crack in your browser tabs. Take the free tier or the free open source tool instead. Neither one carries the risk.

Verify the pricing yourself at screamingfrog.co.uk before buying, since these numbers were checked in September 2026 and vendors change them.