Difference between using url-metrics ueid and the links API

[JohnR]

Hi,

Probably missing something here as I'm getting strange results when
trying to count the number of external links to a page (both juice
passing and non juice passing).

Since this can't be done using the free API, I tried getting the ueid
from the url-metrics API where I know I will only get the juice
passing external links.
Did that for a certain page and got 396.

Next I used the links API with this format: $linksService-
>getLinks($objectURL, LINKS_SCOPE_PAGE_TO_PAGE, LINKS_FILTER_EXTERNAL,
LINKS_SORT_PAGE_AUTHORITY, LINKS_COL_URL, 0 , -1); with the idea of
getting all external links to the page (juice passing and non juice
passing).
I counted the number of links it returned and got 126.

I know that the links API is limited to 25 links per source domain but
I don't think that is the cause for this huge difference, would expect
the links API to provide a bigger number as it's suppose to count non
juice passing links as well.

Is there anything I'm missing here that would explain this gap? If it
helps, I also checked yahoo site explorer and got the following
results:
126 incoming links from external pages to all pages of the domain
66 incoming links from external pages to the specific URL
which is also strange as my query using the links API should have
returned links to the specific URL and not the entire site.

Hope someone can sort things out for me here as it's driving me crazy.

thanks!

[Nick Stielau]

Hello,

Could you mention the URL in question?

The discrepancy may be caused by URL canonicalization, where the
url-metrics API may return metrics about links that are canonicalized
to the URL in question, whereas the links API might not. I'll look
into it. Thanks,

-nick

> --
> You received this message because you are subscribed to the Google Groups "SEOmoz API" group.
> To post to this group, send email to seomo...@googlegroups.com.
> To unsubscribe from this group, send email to seomoz-api+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/seomoz-api?hl=en.
>
>

--
Nick Stielau
206-245-7991

[JohnR]

I checked an adult oriented site, www.fetishxxx.com

thank you!

> > For more options, visit this group athttp://groups.google.com/group/seomoz-api?hl=en.
>
> --
> Nick Stielau
> 206-245-7991

[JohnR]

Sorry for the bump but I am still unable to figure out the
differences...

any help will be appreciated.

[Nick Stielau]

Hey,

I looked into this. The nature of these links is different than most
of the link reports I look at, for technical webpages, blogs,
companies etc.
For one thing, there are far more juice-passing links than
non-juice-passing links, which is usually not the case.

I believe that the discrepancy between the number of returned links
and the ueid is indeed due to the 25 link-per-domain restriction.

I get these numbers from the url-metrics calls
{
"ueid": 418,
"feid": 498,
"peid": 551
}

A links call with "Filter=external+follow" yields 213 links, with four
domains hitting the 25 link limit.

This would leave 205 juice passing links from those four domains.
That seems like a lot, but if you look at the incoming links, many are
clearly generated:

"...-tube.asp?Category_Criteria=1",
"...-tube.asp?Category_Criteria=2",
"...-tube.asp?Category_Criteria=2&Start_SortID=106",
"...-tube.asp?Category_Criteria=2&Start_SortID=16",
"/index.php?a=sendmessage",
"/index.php?a=stats",

I would guess the majority of links are coming from that domain.

Hope that helps.

-nick

> For more options, visit this group at http://groups.google.com/group/seomoz-api?hl=en.
>
>

--
Nick Stielau
206-245-7991