Discrepancy between OSE internal links and Webmaster Tools
-
Hi,
We have been attempting to increase our rankings for our key term for a number of months now, and we are seeing an unusual discrepancy between OSE and Webmaster Tools that we believe may be causing us harm.
OSE shows our homepage, www.unionroom.com, as having only 27 internal links (only three of these aren't 301's too), whereas Webmaster Tools is showing just over 1,500 internal links to the homepage.
The three pages that OSE are displaying as linking back to the homepage are the homepage, our blog and our accessibility page. These pages are all ran off the same template and have the same header and footer (with three home links spread across header and footer). You can see our confusion!
Is this a case of OSE's stats being incorrect (the website has been live for ~6 months) or is there an issue with our in-linking structure that we are missing?
Thanks!
-
Yep that is correct, which can be 50/50.
-
Thanks for the link David, we will take a look through that.
I've quickly looked at some competitor websites and their home links are all href="/", so would I be correct in assuming that it if this affected our website it should also affect others in OSE?
-
To provide more detail there is a relative link back to the homepage though href="/" which our crawler doesn't work well with so having the full absolute path is recommended.
Here is a great resource that goes more in-depth about the differences: http://www.coffeecup.com/help/articles/absolute-vs-relative-pathslinks/
-
I ran some inner pages through OSE (/work and /contact) and both look to be indexed.
It's weird that you say 'For example we do have /blog/ indexed but there isn't a link going back to the homepage as "www.unionroom.com"' as this is one of only two pages (blog and accessibility) showing that DO link back to unionroom.com.
In summary:
- All inner pages are crawled.
- All inner pages (that we have tested) are indexed.
- All of our pages use the same template.
- OSE is stating that only two inner pages link back to the homepage (blog and accessibility).
-
It's possible we may not have indexed those pages or they have a link back to www.unionroom.com within the source code. You can verify by putting those pages into OSE directly to see if they have been indexed and also verify the link is in the source code.
For example we do have /blog/ indexed but there isn't a link going back to the homepage as "www.unionroom.com"
-
The crawl looks to be working correctly, as in it is finding all of the pages and posts that we would expect (approximately 300). It looks to be solely an issue with the number of internal links that OSE is saying the page www.unionroom.com has (27).
All but two of our inner pages, aside from the blog and accessibility, aren't listed on our internal link report for the homepage. This includes top level pages like services, about, contact etc, all missing.
The templates that we use for the blog and accessibility pages are identical to to what we use for other inner pages, so it's unusual that the others aren't listed.
-
Anytime!
Website structure can play a part but also time timing of when sites are indexed. There can be various networking issues that can prevent our crawler and others from reaching certain pages. You can always run a crawl test to look for any possible errors which you will have time to correct before the next update: https://moz.com/researchtools/crawl-test
Also some pages may be blocked intentionally from robots.txt and no-follow tags.
-
Hi David,
Thanks for your reply.
I'm concerned that on the competitor websites that we are evaluating using OSE, their websites' internal links are displaying much higher numbers than ours (27 vs ~300) even though our websites are of similar size. Could this possibly highlight an issue with our website structure (i.e. crawlers are having an issue finding ways back to the homepage via inner pages, even though we have three on each page around the website), or is this a case of MOZ crawling perhaps showing incorrect statistics?
I'm trying to eliminate the possibility of OSE's statistics being incorrect before we attempt to change the structure of the website and begin testing etc.
Thanks!
-
Hi Jordan!
There are a lot of reasons you could be seeing these data discrepancies. If you're seeing more GWT errors, it could be because they have 'discovered links' through different means. Moz starts with your home page and uses recursive crawling to find pages on your site. We keep crawling until we stop finding unique links, or until we hit your page crawl limit. Bottom line, when we crawl your site your homepage is the only seed. We know that Google uses multiple seeds, so it could be possible Google is indexing more pages that way. In sum Google will never match up with OSE but it's good to use both along with other backlink tools to cover all your bases.
Hope this helps!
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Moz's new Link Explorer, including our revamped index and DA/PA scores is now open to everyone!
Hey Moz Community, Link Explorer is now open to the public! Everyone can access it via a subscription or a free Moz ‘Community’ account. As you may know by now, the brand-new Link Explorer tool is primed to replace Open Site Explorer as Moz’s link building and analysis tool. The Link Explorer project is the result of an incredible amount of perseverance and hard work by the team, and we’re proud to be able to finally share it with you — we know it’s going to revolutionize how you approach link building and make your job easier. You can read more about the tool here in Sarah Bird’s announcement post. Because Link Explorer improves on almost every aspect of Open Site Explorer, the metrics have improved, too. That means you’re likely going to see some Domain Authority and Page Authority discrepancies between OSE’s index and Link Explorer’s index. We definitely suggest you use the new DA/PA from Link Explorer, as they’re more accurate and refresh daily rather than monthly, as was the case with OSE’s index. However, we also realize that many of you use these metrics to report to your clients and colleagues, and a sudden change or fluctuation could potentially make your job harder. Which DA is the real DA? The new DA is based on a much larger index that has many improvements, several of which are designed to make the index more like Google’s than ever before. You should consider moving towards the new DA (and the old DA won’t be updated after April 26th 2018, so the sooner the better). While there will be fluctuations as we improve the model and add features to the index, we expect it to remain largely stable and to be a far more accurate picture of a site’s authority according to how it’s seen by Google. Why is Link Explorer’s DA/PA considered better than OSE’s, and which should I trust? The larger link index with improved crawl selection allows us to produce a stronger model that includes a much larger proportion of the web. That being said, DA and PA should always be considered in the context of your competitors. A drop in PA or DA relative to the old OSE is of little concern if your competitors saw similar movement. Is Domain Authority/Page Authority an absolute score or a relative one? Both DA and PA are relative to the Internet as a whole. If Facebook acquired a billion new links, everyone’s PA and DA would drop relative to Facebook. Because of this, it’s always best to look at PA and DA in comparison to your competitors. What does a drop/raise in DA mean in Link Explorer vs OSE? How can I explain this to my clients when I’m reporting it? DA and PA should always be considered in the context of your competitors. A drop or raise in PA or DA relative to the old OSE is of little concern if your competitors saw similar movement. Reporting that your site has moved from a DA of 45 to a DA of 42 doesn’t tell the whole story, but reporting that your site has a DA of 42 while your main competitor moved from a 43 to a 37 shows that, relative to the sites you’re competing against in the SERPs, your site has significantly more authority and ranking power. What’s happening to MozTrust and MozRank and why, and what should I replace those with? The improvements to our DA/PA and Spam Score metrics now now account for more important nuances in helping you determine one site’s ability to rank higher than another. Because they no longer correlate with Google’s ranking model as well as they used to, MozRank and MozTrust are being deprecated for better metrics. Users should rely on Page Authority, Domain Authority, and Spam Score to determine the importance and quality of pages, domains, and links. I have historical data I use to help my clients benchmark their progress. What do I do now that DA is calculated differently? You should annotate any KPI changes referencing the change in DA and PA. However, most importantly, you should compare those changes to your competitors, as this will best show how strong your site’s authority is relative to the sites you’re competing against in the SERPs. We take updating our metrics very seriously, and our last major update to the model was 7 years ago. Users of Domain Authority and Page Authority can expect us to continue to produce steady, reliable metrics for the long haul, and only make changes to these metrics when we believe the benefits dramatically outweigh the stability of the metric. Do you have any questions about the new metrics? Anticipating a tough time reporting changes to clients or bosses? Metrics, features or functionality missing that you would want to see? Let us know in the thread, and we’ll work to find a good answer for you. Hope you enjoy the new Link Explorer product and the amazing new link index powering it. We are very excited to provide this valuable data to our community and customers.
Link Explorer | | IanWatson9 -
Why does OSE show different link equity for homepage with and without trailing slash?
Hi all, I am working with a developer who believes OSE (Open Site Explorer) is flawed. His reasoning is that OSE shows completely different link equity and domain authority for these two URLs [company name is purposely withheld]: http://www.mydomain.com/us/ and http://www.mydomain.com/us NOTE: http://www.mydomain.com/us has been properly 301-redirecting to http://www.mydomain.com/us/ for years. The developer's argument is that the URL that is being 301-redirected shouldn't have any link equity at all, therefore OSE is flawed. Here is the response I would like to give him. Please feel free to poke holes, provide feedback or flat out correct me if I am wrong: Dear inquisitive developer, Any URL that is being linked to by another site that has link equity will get link equity passed to it. This is true even if the URL is a page that doesn't exist. What it can't do (if it doesn't exist) is then pass that link equity on to another page. Moz's OSE is showing separate link equity for the two versions of this URL because other sites have linked to both versions. The fact that one 301-redirects to the other is good and ensures that link equity from one is passing to the preferred URL. The 301-redirect does not, however, remove link equity from the first URL, but rather passes the existing link equity from one to the other. Consequently, Moz's OSE is not flawed, but rather is displaying the link equity of one URL that is benefiting the other via a 301-redirect. Sincerely, Dana
Link Explorer | | danatanseo3 -
See Lost Links?
My Moz analytics campaign says i have lost some links since the last crawl on one of my sites, Is there an easy way in moz to see which links these are i lost and where from? James
Link Explorer | | Antony_Towle0 -
Why does OSE redirect to another url when the url you're testing doesn't redirect in your browser?
Run an OSE report on www.schoolmatters.com/ca/san-diego/personal-injury-lawyer/. OSE redirects to brandingwithseo.com. Why? And if you go to brandingwithseo.com in your browser, it redirects to seocompany.org. Why is OSE redirecting? It appears the links that come up are a combination of links for the seo company and for schoolmatters.com.
Link Explorer | | katandmouse0 -
Facebook likes in OSE when there is no Facebook Like button on the web page
In OSE how are page specific facebook likes calculated when the page does not have a Facebook like/share button on it? I have seen many pages with multiple Facebook likes counted against it in OSE but cannot see how people could be liking the page.
Link Explorer | | TheHutGroup0 -
OSE Says I have 3,115 total links but I can only export 660 via advanced reports
Hi Everyone! My site (http://carwow.co.uk) has 3,115 total links according to OSE, I need to export them all but advanced reports export will only give me 660. Please advise! Thanks, James
Link Explorer | | JamesPursey0 -
Updating OSE data
Hello Moz, I was wondering if anyone knows how Moz goes about updating their OSE data? I've been doing some link building that past couple of weeks, and I notice that the number of "just discovered" links will update every time I gain a new link, however, the "Linking domains" number doesn't (or maybe, rarely) update. I've checked through the "Linking domains" list to see if the link is there, but none of the links I've just recently obtained have shown up under that metric. Does anyone have any insight on this or is anyone experiencing the same issue? Thanks for reading!
Link Explorer | | GalcoIndustrial1 -
Unable to download OSE Backlink attachment. Showing error
Hi While downloading attachment of backlinks from open site explorer - facing error that something went wrong and reported to moz tech team can you resolve it pl
Link Explorer | | Modi0