A post goes out, a link inside it points at a page, and nobody looks at that link again. Two years later the site has been redesigned, a product line has been retired, and the post is still collecting clicks that land on an error page. A free desktop crawler can find those dead destinations, but the one most teams reach for stops at 500 URLs per crawl, and a brand archive often holds more than that. Figures below were read from the vendor’s own pages on 2026-09-30.
Why old social posts keep sending people to dead pages
A social post is published once. The link in it is fixed at that moment, while the site behind it keeps changing: pages get renamed, folders get restructured, old campaign pages get pruned.
There are two places a dead link can live, and a crawler sees only one of them. Inside the site’s own pages, a crawler can follow links and report what each one returns. Inside the published posts themselves, on the social platforms, it cannot see anything. A crawler audits the destination side. The list of URLs that old posts point at has to come from your own post export or scheduler history, and you then hand that list to the crawler.
This is a broken landing page problem, and the 500 URL cap is the constraint you plan around.
What the free version of Screaming Frog actually gives you
The vendor’s pricing page says the SEO Spider “is free, however you can purchase a licence to crawl more than 500 URLs and have access to advanced features.” Its FAQ says that without a licence the tool is “limited to crawling a maximum of 500 URLs each crawl, crawls can’t be saved, and advanced features and the configuration are restricted.”
| Item | Free version | Source |
|---|---|---|
| Crawl size | Up to 500 URLs each crawl | FAQ, pricing page |
| Saving crawls | Not possible | FAQ |
| Configuration options | Restricted | FAQ |
| Advanced features | Restricted; the product page names JavaScript rendering, custom extraction and API integrations | Product page, FAQ |
| Finding broken links, errors and redirects | Listed as included in the free column | Product page comparison table |
For an audit, the consequence of no saved crawls is practical: you cannot keep a baseline and reopen it next quarter to see what changed. You would export the results to a spreadsheet yourself.
The FAQ also covers lapse. Licences are annual, and it states that “the SEO Spider returns to the restricted free version upon expiry of the licence,” with configuration unavailable, the 500 URL maximum back in force, and previously saved crawls unable to be opened. A team treating a licence as a one-off cleanup should export what it needs before expiry.
Where 500 URLs runs out on a normal brand site
The cap counts URLs, and a crawl meets more than pages. The vendor’s general user guide advises “only crawling internal HTML, by unticking images, CSS, JavaScript, SWF and external links” in the configuration, which implies those resource types are otherwise crawled. Whether they count toward the 500 is not stated in the pages read for this piece. No crawl was run for this piece, so that point stays unconfirmed.
Here is the catch. That guide’s suggested fix lives in the configuration, and the FAQ says configuration is restricted in the free version. Those two statements pull against each other, and the vendor pages do not reconcile them. What the guide does describe is two ways to narrow a crawl: entering a subfolder and choosing the subfolder option, or using list mode to paste or upload specific URLs. The guide does not say whether either is available without a licence. That is unconfirmed here, so test both on a small crawl before relying on them.
The arithmetic below uses hypothetical numbers only, not data from any real site. Picture a site of 120 pages. If each page also loads 3 images and 2 scripts that a crawl records as separate URLs, and if those count, the total is 120 + (120 x 3) + (120 x 2) = 720. That already exceeds 500 while covering only 120 pages. If resources do not count, the same site uses 120 of 500. The answer changes the plan, so check it with a real crawl first.
A routine that spends the cap on the pages that matter
- Export the destination URLs from your old posts, using your platform exports or scheduler history.
- Rank them by how often each was shared. A page linked from forty posts matters more than one linked from one.
- Crawl the top of that ranking. Use list mode or a subfolder crawl if your version permits it; if not, crawl in batches under the cap.
- Open the response codes report and filter for client errors (4XX) and redirects (3XX). The vendor’s broken link page describes filtering the Response Codes tab by Client Error (4XX) and using the Inlinks tab to see which URLs link to a broken one.
- Decide what to do with each result.
Two result classes, defined
A 404 means the page is gone and the visitor arrives at an error. A redirect, a 3XX response, sends the visitor on to another URL, so the link still delivers them. For this audit, a 404 is a loss and a redirect is a maintenance item.
What to do with each
For a 404, redirect it to the closest live page, restore the page if it still has value, or accept the loss if the old page has no fair equivalent and few posts pointed at it. For a redirect, check that it lands somewhere sensible and leave it alone if it does. This is practice advice, not a claim about platform behaviour.
No crawl was run for this piece, so it makes no first-hand claim about results or timings.
What the licence adds, and what it costs
The vendor shows its price in more than one currency, and the pages read on 2026-09-30 do not agree in shape, so each figure is quoted with its page. The user guide says “For £199 a year you can buy a licence.” The product page shows a licence at €245 per year. These are two figures from two pages and are not converted here. No price is taken from the pricing page, whose figures were not readable in its static HTML. That page does say licences last 1 year, are per user, and that discounts apply to bulk purchases of 5 or more licences. It gives no amount in the text read.
Per the FAQ, a licence removes the 500 URL limit, allows crawls to be saved and opened, and opens the configuration and advanced features. The product page names JavaScript rendering, custom extraction and API integrations among those advanced features.
The reading that follows is this post’s judgement, not a vendor claim. For a broken link audit, two things matter: removal of the cap, and saved crawls, which give you a before and after comparison. Most of the rest is aimed at other jobs.
One other crawler’s free limit, for scale
Brokenlinkcheck.com is an online tool, not a desktop crawler, so this is a different kind of comparison. Its page states that the free version has a 3000-page limit, with no limits on the number of hyperlinks within those webpages. The same page’s free-version message states that subfolders and URLs with slashes are not supported in the free version. Both were read on 2026-09-30.
This post does not rank or recommend between the two. There are two limits and two constraints, and the reader decides which one fits their archive.
What a crawler will not tell you about old social posts
A clean crawl proves that the destination pages respond. It does not prove that the link inside a published post is the one you intended, that its tracking string survived, or that a short link still resolves. For the tracking side, see UTM conventions that survive a handoff. For links that pass through a custom domain, see what breaks on a branded short domain.
A one-time crawl of many links is also a different job from watching one page continuously. For the second job, read monitoring the one page a campaign sends traffic to.
followedapp is published by the team behind RecurPost. Prices verified 2026-09-30 from the vendors’ own pages. This piece is not Watchdog data and is not on the re-verification schedule.
Next step
Start with the export, not the crawler. List the destination URLs from your posts, rank them by share count, and check the top of that list first.
FAQ
Is the Screaming Frog SEO Spider really free?
Yes for crawls up to 500 URLs, per the vendor’s FAQ, which says no email capture or sign-up is needed. Crawls cannot be saved and configuration and advanced features are restricted, so the free version suits a bounded audit and not an ongoing one.
Can the crawler find the broken link inside the social post itself?
No. It checks the URLs it is given or can reach on a site. The text of a post lives on the platform, so the links inside it have to be exported from your own post history and then supplied to the crawler as a list.
Sources
- Screaming Frog SEO Spider FAQ, fetched 2026-09-30
- Screaming Frog SEO Spider pricing, fetched 2026-09-30
- Screaming Frog SEO Spider user guide, General, fetched 2026-09-30
- Screaming Frog SEO Spider product page, fetched 2026-09-30
- How To Find Broken Links (404s), Screaming Frog, fetched 2026-09-30
- Brokenlinkcheck.com, fetched 2026-09-30