OSE not picking up all internal links?

[Blink-SEO]

Hi! I've been checking a few pages in OSE. For example:

https://www.holidayhypermarket.co.uk/holidays/last-minute

This, according to the tool, has only 6 internal links. However, a Screaming Frog crawl picks up 700+, which is about right seeing as there is a link in the footer. What do you think could be behind the discrepancy?

[William.Lau]

OSE is not an end all be all for links. They only gather a certain number links and data.

You're already using screaming frog to get this data, so that should be accurate. You can also check Google Webmaster Tools, which has a section called "Internal Links".

[RyanPurkey]

OSE also clusters by domain, so for a given link you're not going to get near the 700+ which are internal. As a general practice OSE is great for getting at link diversity, goal setting, and seeing where the competition has gotten links in the past. It should give you plenty of ideas that way. Cheers!

[Blink-SEO]

Hi! I understand that OSE isn't 100% accurate, but I'm wondering if something else is going on. It picks up internal links for competitors fairly accurately. It would be good to know though why it isn't picking up the footer links - could the code be difficult to crawl etc? It may be hinting at a problem that I've missed.

[RyanPurkey]

I wouldn't use OSE as an onsite crawler. That's what a tool like Xenu Link Slueth or Screaming Frog is for.

[Blink-SEO]

Thanks. I guess my point though is if that OSE is struggling to find the links, could the search engines? These aren't buried links - they are easy to see on the page. It would be good to know if there is a glitch or if something else is going on.

[RyanPurkey]

Doubtful. Running a site:holidayhypermarket.co.uk search in Google returns 10,700 results, plus from the help documentation on OSE it says, "The second thing to keep in mind that the exported data will only include the top 25 pages per linking root domain," from: http://moz.com/help/guides/research-tools/open-site-explorer/top-pages-ose

So by design it's not really a tool that's trying to duplicate Google, rather make an actionable scaled down copy of what's going on.