| Important Announcement: *Written SEO Website Reviews Available* |
![]() ![]() |
Dec 16 2009, 12:28 PM
Post
#1
|
|
![]() HR 5 ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 301 Joined: 5-June 04 User's local time: Jul 31 2010, 05:05 PM From: San Mateo, California Member No.: 3,834 |
Hi:
I have an international client that has websites in the various countries where their products are sold. Back in the summer we decided to develop a link building contest with all the country marketing departments to get them jazzed up about developing links to their country sites. Nothing like a good competition and an iPod for the winner! For each country I was tracking inbound link counts using Google Webmaster Tools and Yahoo data. About three months ago the data started showing massively different numbers for many of the countries I am tracking. For one country, I would see that one month Yahoo would show 10,000 inbound links and the next month only 300. We can't see anything that would make such an impact. I didn't want to rely solely on GWMTs because I have been seeing a downward trend in the number of reported links for all our clients this past year. Hence, I thought Yahoo might be more stable. So - the big question is; 1) If you were going to track links for 10 websites on a monthly basis what would you determine to be the most reliable tool? Is it even possible to run such a contest when search engines are getting so aggressive in discounting all sorts of links? Thanks. Rosemary |
|
|
|
Dec 16 2009, 02:30 PM
Post
#2
|
|
|
HR 9 ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 3,915 Joined: 5-April 05 User's local time: Jul 31 2010, 05:05 PM From: Seattle, WA Member No.: 7,091 |
In a situation like that I would have required the link builders to report their links in spreadsheets. That would have helped you implement some quality control over the process. Just sending people out to build links -- and incentivizing the process -- is a pretty risky behavior, in my opinion. Where are they getting the links? How do you know no one is cheating? What if someone does something to get one of the brand sites penalized?
I would never recommend using a third-party monitoring tool to judge such a contest. You need to make the link builders accountable for their work. |
|
|
|
Dec 16 2009, 06:41 PM
Post
#3
|
|
![]() HR 5 ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 301 Joined: 5-June 04 User's local time: Jul 31 2010, 05:05 PM From: San Mateo, California Member No.: 3,834 |
Michael: Thanks for your reply. How do you usually measure back links for your clients?
|
|
|
|
Dec 17 2009, 12:46 PM
Post
#4
|
|
|
HR 9 ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 3,915 Joined: 5-April 05 User's local time: Jul 31 2010, 05:05 PM From: Seattle, WA Member No.: 7,091 |
It depends on the client. Some don't want to know the details. For others we'll prepare either a summary (with appropriate backup available in case they want the detail) or we'll submit spreadsheets with detailed placements.
It also varies by contracted service, of course. Some of our link placements are not really intended to help with rankings but rather in just getting sites launched, indexed, etc. |
|
|
|
Dec 17 2009, 02:14 PM
Post
#5
|
|
![]() HR 5 ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 301 Joined: 5-June 04 User's local time: Jul 31 2010, 05:05 PM From: San Mateo, California Member No.: 3,834 |
Michael: Why do you not trust Google Webmaster Tools' information?
|
|
|
|
Dec 17 2009, 05:36 PM
Post
#6
|
|
|
HR 9 ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 3,915 Joined: 5-April 05 User's local time: Jul 31 2010, 05:05 PM From: Seattle, WA Member No.: 7,091 |
|
|
|
|
Dec 17 2009, 06:24 PM
Post
#7
|
|
![]() HR 4 ![]() ![]() ![]() ![]() Group: Active Members Posts: 132 Joined: 9-March 05 User's local time: Jul 31 2010, 05:05 PM Member No.: 6,852 |
What tool could accurately provide information about which links Google measures as "passing value"? Only Google can say with any accuracy which links it values and which it does not, and of course Google doesn't release that information. No link tool provides that information.
Of course GWMT is not 100% accurate. But no tool is. We have to work with what it available. Barry It's unreliable. Even if you could reasonably assume it was tracking all the links Google knows about (and they say it doesn't), it still won't report which links pass value and which links don't. |
|
|
|
Dec 17 2009, 07:16 PM
Post
#8
|
|
|
HR 9 ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 3,915 Joined: 5-April 05 User's local time: Jul 31 2010, 05:05 PM From: Seattle, WA Member No.: 7,091 |
Working with what is available is one aspect of what we do as search optimizers. But in the context provided above (judging a contest), there is simply no accountability if you are relying on tools to monitor link building. By extension (and I was inspired to write about this at length on Best SEO Blog today) we have no accountability to our clients and executives if they are just relying on tools or reports from tools to gauge link building.
Of course, the practice of selling link building services has been questioned a number of times. What is the return on investment if you just pay someone to build links for you? Links for the sake of links is not very helpful -- and some people may have been ripped off by link building companies that don't prove any value for their links. It seems to me you have to measure the work that is reported by the link builder. Why would they hold back some links? If you find links they don't claim, then you can attribute those links to other sources. In a contest, how would you know which links were obtained through the contestants' efforts and which links were naturally placed if you only looked at what link tools tell you? I just cannot see any value in using a link monitoring tool for that kind of purpose. You can certainly use these tools to look at trends and things like that, but they're not going to provide you with the granularity you need to hold people accountable. |
|
|
|
Dec 18 2009, 12:30 PM
Post
#9
|
|
![]() High Rankings Advisor Group: Admin Posts: 29,889 Joined: 21-July 03 User's local time: Jul 31 2010, 08:05 PM From: Ashland, MA Member No.: 2 |
QUOTE What tool could accurately provide information about which links Google measures as "passing value"? There are none. |
|
|
|
Dec 18 2009, 06:46 PM
Post
#10
|
|
![]() HR 5 ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 301 Joined: 5-June 04 User's local time: Jul 31 2010, 05:05 PM From: San Mateo, California Member No.: 3,834 |
|
|
|
|
Dec 18 2009, 10:31 PM
Post
#11
|
|
![]() Convert Me! Group: Admin Posts: 17,540 Joined: 17-August 03 User's local time: Jul 31 2010, 07:05 PM Member No.: 551 |
For a contest or if you're hiring someone to do link building Michael has the answer. Force them to give you a spreadsheet or even a simple list of where they've placed links (the actual page the links appear on, not just the domain) and have someone check to make sure the links are live.
As to the more general question, all I can tell ya is I gave up even attempting to track the number of links reported ages ago. It just seems like such a huge waste of time and doesn't really tell you anything. But if you absolutely have to do it, pick a tool and use it consistently. But don't freak out when it shows changes, especially when there are massive drops in reported links. And realize that even if a tool reported everything it knew about, it's still going to miss backlinks. And none of them report everything. On a slightly different subject when I want to gauge how widely an article, press release, tutorial, etc I write gets distributed I let good old Google Alert give me snapshots. I just load a unique phrase that appears somewhere in the document, typically from the title or headline if I can, into Google Alert and have it send me an email when it sees it pop up. Noting again that they're not going to necessarily report every instance they know about. But at least with this one you're asking them to report on content that matches instead of links. |
|
|
|
Dec 29 2009, 07:29 PM
Post
#12
|
|
![]() HR 5 ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 301 Joined: 5-June 04 User's local time: Jul 31 2010, 05:05 PM From: San Mateo, California Member No.: 3,834 |
Hi Randy: Thanks for your excellent response (as usual) (IMG:style_emoticons/default/smile.gif)
I don't spend a lot of time monitoring back links but I do use it as a tool to evaluate where a client 'fits' on the web - compared to competitors. Sometimes I have to explain to non-marketing executives why their site doesn't rank on page one for highly competitive single or two-word phrases and I have used Link Popularity (gauged by back links) as one piece of the discussion. If you do a competitive review, do you not look at link popularity? Rosemary |
|
|
|
Dec 30 2009, 11:31 AM
Post
#13
|
|
![]() Convert Me! Group: Admin Posts: 17,540 Joined: 17-August 03 User's local time: Jul 31 2010, 07:05 PM Member No.: 551 |
Yes I look at link popularity. But I always use the same tool and don't stress over the exact numbers.
Much like with keyword research I use these numbers as a comparative between 2 or more sites. The raw number of links probably have an effect, but doing it as a comparison tells me just as much. And you also have to keep in the back of your mind that just a few really good links with the right anchor text can cause one site to jump up in the rankings for the anchor text phrase, even if the raw number of links doesn't quite add up. Oh, and one thing to add to my Google Alerts comment from earlier. I would caution everyone to be a bit careful when clicking through to pages that show up in this report. For the last year or more I've been getting a fair number of instances where Google Alert will link me over to a page that contains the words and phrases I've entered into my account where the target page is actually trying to distribute a virus. So if you're going to click through on those reports make sure your PC security is top notch. And it's important to watch to see if the site is trying to access yet another site to download something to your computer. Luckily, my AV software catches 99.9% of those attempts. Plus I use Chrome as my default browser, which is a bit better than IE in fending off these attempts. |
|
|
|
Jan 6 2010, 12:27 AM
Post
#14
|
|
|
HR 1 ![]() Group: Members Posts: 1 Joined: 6-January 10 User's local time: Aug 1 2010, 09:05 AM From: Osaka Member No.: 27,739 |
Hi,
I am an in-house SEO working on sites in about 40 countries. We regularly use short-term staff or interns to do link development work in various markets, but also use our local staff in various sales offices. I understand how difficult it is to monitor and evaluate what those people are doing, particularly when you don't speak the languages of the sites they are contacting. The main way we keep track of what they are doing is by manual reporting, as others have recommended. Each link builder keeps a detailed spreadsheet with pre-defined fields for each site. The spreadsheets are stored in a place we can get them even after short-term staff leave or move on to other positions. In addition to the spreadsheets, we have a phone meeting or face to face meeting every 40 hours of work or so (some people work part time so we meet once a month or every two weeks). The link builder reports orally on what he's done that works well, what he's proud of, what he's working on now, etc. The meeting really adds a lot in my experience on top of the spreadsheet since the link builder can pick out things of importance to report. It also gives us a chance to hear them talk about their work and catch any mistaken strategies or misconceptions they voice during that reporting. These meetings can include several link builders at once to save time. We do monitor backlinks automatically as well though for a couple of reasons. - First, we like to keep an eye on the links we consider to be most important to us. This is a value judgement and we could be wrong about their importance of course. But when we get a really valuable link, we want a piece of software to alert us if that link disappears. Of course sometimes the software is wrong and it's a false alert, but on more than one occasion this monitoring has allowed us to recover an important link that was lost because of site restructuring or a misunderstanding. - Second, as others have said, it is nice to have a general idea of the number of links being reported by various engines and how those evolve over time. Yes, sometimes they jump around and we ignore that. But in my experience when we're actively building links in a market we see a general positive evolution whereas when nobody is doing anything we see a pretty flat line. In several instances I've found out about a successful campaign or a new link building intern in a local office after having noticed an increase in backlinks from the software. - Third, because we work on lots of markets and some of those markets share languages, I like to keep an eye on the domain extensions that are linking to our various sites. Our rule is that sites in one market link to their own version of our site, so .de sites link to our .de, .at sites link to our .at etc. It happens regularly that someone over-zealous in Germany for example will go around contacting Austrian sites and getting links to their .de site. We don't want to get into internal fights about this. A quick glance through the automated new links reports will show me if this is happening on any scale so I can stop it before it becomes an issue. The software we use is Advanced Link Manager, which just compiles backlink lists from the engines you designate and then goes and visits the pages to check if the links are still there or not. It's the same data you'd get from checking the engines manually, but keeps you from having to do it by hand. It cannot get the Google Webmaster tool data, but it does compile data from the public interfaces of several engines. The support on these products has been stellar in my experience, which can be really important when working with lots of little local engines in various markets and various languages/encodings. Hope that helps. |
|
|
|
Apr 13 2010, 09:56 AM
Post
#15
|
|
![]() HR 5 ![]() ![]() ![]() ![]() ![]() Group: Active Members Posts: 301 Joined: 5-June 04 User's local time: Jul 31 2010, 05:05 PM From: San Mateo, California Member No.: 3,834 |
Does a link that was created from link shortening services like bit.ly have the same value as any other regular link?
Thanks Rosemary |
|
|
|
![]() ![]() ![]() |
|
Lo-Fi Version | Time is now: 31st July 2010 - 07:05 PM |