Screaming Frog broken link building: find dead links to replace
Short answer Collect pages that link out to many sites, such as local resource lists. Paste them into Screaming Frog in list mode, remove the crawl depth limit and check only external links. Every 4xx or No Response link is a gap your page might fill. That setup needs the $279-a-year licence; the free version can check one page at a time.
Want help applying this to your business? Talk it through on a free 30-minute call.
Book a callThe method in one table
You don't crawl your own site. You crawl other people's pages that link out a lot, and look for their dead links. A resource page is any page that lists useful links: a chamber of commerce member list, a city's "community resources" page, an association's "find a pro" page. Businesses close and move, so these lists collect broken links.
| Step | What you do | Where in Screaming Frog (November 2026) |
|---|---|---|
| 1. Collect | 20–200 resource pages in your town or trade | Outside the tool: Google searches |
| 2. Set up | Check only the outside links on those pages | Mode > List; Configuration > Spider > Limits and > Crawl |
| 3. Crawl | Paste the list and run | Upload > Paste |
| 4. Read | Find 4xx and No Response links, and which page holds each | Response Codes tab; Inlinks; Bulk Export |
| 5. Pick and pitch | Keep dead links your page can truly replace | A spreadsheet and your email |
Steps 2 to 4 follow Screaming Frog's own broken link building tutorial. The why behind the tactic, rebuilding a dead page and the pitch itself are in broken link building, step by step. This page is the crawler setup.
Do you need the paid licence?
For the list-mode method, yes. Screaming Frog's pricing table marks "Crawl Configuration" as paid only, and its FAQ says that without a licence "crawls can't be saved, and advanced features and the configuration are restricted." The method depends on changing two settings.
| Free | Paid licence | |
|---|---|---|
| Price | $0 | $279 a year per user (US-dollar price), or £199 |
| URLs per crawl | 500 | No set cap; depends on your computer's memory and storage |
| List mode with depth limit removed | No (needs configuration) | Yes |
| Check one resource page's outside links | Yes, on default settings | Yes |
Prices are from Screaming Frog's pricing page in November 2026. It shows US dollars, pounds and euros, with no Canadian-dollar price. A licence lasts one year and covers one person.
If you'll check a dozen pages once, the free version is enough. If you'll check a hundred pages, or repeat this every quarter, the licence saves days.
Step 1: build a list of resource pages
Search Google for pages in your area or trade that link out. The `site:` operator limits results to one domain or ending; Google says it requests "results from the particular domain, URL, or URL prefix specified".
Searches that tend to surface resource pages:
- `"[your city]" "community resources"`
- `"[your trade]" "recommended contractors"` or `"preferred vendors"`
- `"[your city]" chamber "member directory"`
- `site:.gov "[your county]" "helpful links"`
- `site:.edu "[your trade]" resources`
Keep pages whose readers could be your customers. A homeowners' association vendor list matters to a roofer; a national "top 100 tools" list usually doesn't. Paste each page's full address into a spreadsheet, one per row.
Step 2: set up list mode to check only outside links
Screaming Frog's tutorial uses four settings. Menu names below are as of November 2026.
- Switch to list mode. Mode > List. List mode crawls only the addresses you give it.
- Remove the depth limit. In list mode, the list mode tutorial says "The 'Limit Crawl Depth' configuration is automatically enabled and set to '0'", so links on your pages are never checked. Untick it under Configuration > Spider > Limits.
- Check outside links only. Under Configuration > Spider > Crawl, keep External Links ticked and untick every other page-link and resource type (internal links, images, CSS, JavaScript). Otherwise the crawler wanders into each site's own pages.
- Paste and run. Upload > Paste, then let it reach 100%.
Optional: under Configuration > Speed, slow the crawl. The default is 5 requests at once, which small association sites sometimes block.
Free version: one page at a time
Leave Spider mode on and enter one resource page's address. Default settings check outside links' response codes. The crawler stays in the folder you start from, so starting at `chamber.org/resources/` covers that section, not the whole site. A large site can hit the 500-URL cap before you get useful results; start as deep in the site as you can.
Step 3: read the results
Two filters hold your prospects.
- Response Codes > Client Error (4xx). Mostly 404, page not found. Click one, then the Inlinks tab at the bottom to see which of your resource pages holds it.
- Response Codes > No Response. Screaming Frog defines this as when "the URL does not send a response". Often the whole domain has lapsed, as when a business closed. These are often the best prospects.
To work in a spreadsheet, use Bulk Export > All External Links. Per the tutorial, source pages land in column B and dead destinations in column C; filter the status column for 4xx.
Before you pitch, rule out false alarms:
| Code | What it usually means | What to do |
|---|---|---|
| 403 | The site's security refused the crawler, per the Screaming Frog FAQ | Open it in a browser; often it works |
| 429 | The crawler sent "too many HTTP requests" too fast | Slow the crawl and recheck |
| 301 then 404 | The link redirects to a dead page | Recrawl the destinations with "Always follow redirects" ticked |
| 404 | Page gone | Click it yourself to confirm |
Screaming Frog's tutorial suggests cleaning out social media and login links first; they redirect or refuse crawlers and waste time.
Step 4: pick the dead links worth chasing
Keep only links where your page is a real replacement. Sort your shortlist by these, top first:
- Same customers. The resource page serves your area or trade.
- A clear swap. The dead link was a business like yours, or a guide you already have or can write.
- A live, edited page. Recent dates or working links elsewhere mean someone still maintains it.
- Several pages share one dead link. One replacement page can be pitched to all of them.
Step 5: replace and pitch
Write to the page's editor: name the page, the dead link, and your page as one option. Offer nothing for the link. Google's spam policies list "Exchanging money for links, or posts that contain links" and "Exchanging goods or services for links" as link spam. How to rebuild a dead page from the Wayback Machine, and the US and Canadian email rules, are in broken link building, step by step. For writing the emails at scale, see AI outreach.
Track each pitch in your spreadsheet: sent, replied, swapped. A month later, recrawl the won pages to confirm your link is live.
Example
Illustrative cases with invented numbers.
A plumber with the free version (US dollars)
He finds 12 local resource pages and crawls each in Spider mode, one at a time. Three link to a plumbing company that closed; its domain shows No Response. He emails all three with his service page. Two swap the link. Cost: $0 and an evening.
A bookkeeper checking 150 pages (Canadian dollars)
She buys a licence, billed $279 US since there is no Canadian price. In list mode she checks 150 small business resource pages across Ontario and finds 40 dead links. After removing 403s that open fine in a browser, 11 point to dead tax guides she can replace with her own. She pitches 11 editors.
A landscaper hitting the cap
His city's parks site has thousands of pages. Started at the homepage, the free crawl stops at 500 URLs with nothing useful. Restarted at the "community gardens resources" folder, it finishes and shows two dead nursery links.
FAQ
Can Screaming Frog find who links to a dead page on another site?
No. It checks links on pages you give it. Finding every site linking to a dead page needs a backlink database like Ahrefs or Semrush.
Does crawling someone else's site cause problems?
It's normal; be gentle. Screaming Frog says the crawler "is robots.txt compliant", so it obeys a site's crawl rules. A slow speed avoids tripping security.
Why do some broken links in my crawl work in my browser?
The site refused the crawler (403) or rate-limited it (429), not you. Click every link before you pitch.
Is the free version enough for broken link building?
For a handful of local pages, yes. For list mode across many pages, no: it needs settings only the licence unlocks.
Different for you?
- You want the full tactic, Ahrefs route and email rules: broken link building, step by step.
- You're new to Screaming Frog: Screaming Frog for owners.
- Links from partners and suppliers would come easier: backlinks for a local business.
- You need help writing many pitches: AI outreach.
- You'd rather have prospecting and pitching done for you: SEO & GEO Agent.