Disagree @martinibuster!

Notice large number of niche associations, organizations, etc websites which are not at all optimized?

Completely ignorant about SEO to the point having “Welcome”, “Home”, etc as their title tags.

Have seen such sites have <1k , <2k low traffic, but TBPR 5,6, Mozrank north of 5, MozTrust north of 5,6

Despite the low traffic, you can bet the link would be passing high trust rank & consequently help in rankings too.

They may not send you targeted traffic, but the link still helps to improve rankability of the destination site by notching up its thematic relevancy, trust rank & real PR

Also, what if a site is getting major chunk of traffic from outside USA? In that case Sites like Compete/Quantcast will still reflect low traffic number whereas in reality the real traffic number could be substantially higher.


Conversely, have been sent 25k+ visitors traffic/day windfalls thanks to Digg FP links that did nothing to rankings nor few additional links I got due to it.

Bottom-line:

IMHO, we can’t pin down link quality to any single factor in isolation – neither TBPR; NOR traffic.

Personally, I always tend to evaluate based on several page level + domain level factors:

Mozrank, Moztrust, Domain Moztrust, TBPR & MR comparison for page/domain level, number of OBL’s, co-citation, page segmentation, any obvious signs of link selling, etc, etc

To sum it up: can that link pass human quality rater review as editorial citation? If yes, you bet it'd help despite Quantcast/Compete/G trends traffic scores

You raise some excellent points and I very much agree with much of what you state. The reason is because my point is not that link quality can be pinned down to traffic. That is not what I posted nor suggested. ;)

I am pointing out that it is an important one that can be used as a "quick" and "fairly accurate" measure to determine whether or not to dig any deeper. Traffic is generally the symptom of a healthy website. I agree there are more things to look at but if you need a quick snapshot to determine whether to dig any deeper, then traffic is a good starting point.

Despite the low traffic, you can bet the link would be passing high trust rank & consequently help in rankings too.

That's actually like .edu pages. There is an industry around high priced .edu links that is built around this idea of trust and that .edu pages have something extra to pass along. But I believe the advantage of .edu type links has been exagerrated.

The only advantage of a .edu link is the likelihood that there is an absence of negative signals. The trust test, in my opinion, is about whether or not your site gets dumped into the spam bucket or not. I think the issue of trust and relevance are distinct.

1. Big muscles are a sign that someone works out. Similarly, traffic is generally a sign that a site is relevant for it's niche. It's generally (but not always) a reflection of the quality of it's content and many other positive signals.

2. Traffic is a sign that a site is important. It is important because quite likely there are a lot of sites linking to it. I bumped into a blog with a small amount of PageRank, around three, maybe even less. Their advertising rates were pretty high and I thought maybe they were overreaching. So I checked it in QuantCast and the traffic numbers were high. I then did a backlink search and a search through Google News, Google Blogs, and Google Images- That site was being cited in a lot of places.

Clearly, traffic can be manipulated and it's important to know where the traffic is coming from. Not all traffic is equal. But in my experience the low quality traffic is an exception.


3.Traffic is not necessarily a sign that a site has been SEO'd. Aside from highly competitive affiliate niches, there are many successful sites in a wide range of niches rank well because their content is excellent, their authors are authoritative, and because their site looks good- despite a malformed title tag, despite poor optimization.

4. An association site is great and their links are useful. But an association site that has little traffic is itself diminished by this fact. If you have examined association sites then you will know what I'm referencing when I say that low traffic association sites are generally regional ones, and those sites tend to have relatively few link citations.

I certainly see your point & agree with overall gist of your post. I’m not disputing that usually traffic is indication of healthy site either .

I’m merely contesting that traffic can not be taken as a leading sign of quality of link – at least not yet.

Generally agree with point #1 & 2 mentioned in your post. Personally have a site of the type you mentioned in your point #3 ( not seo'ed at all, 200K social media views /pm )

Disagree with your point #4

I’m aware that WWW TOS do not allow external sites, but I hope I’ll be allowed to reference couple of not for profit international association sites to prove my point:

a ) hbscDOTorg/contactlinks.html ( Health Behaviour in School-aged Children Org )

Compete traffic uniques ZERO! , TBPR 6, MT 5.73, DMT 5.78, MR 4.49, PA 34

Another example:

b) ibfanDOTorg/useful-links.html ( International Baby Food Action Network Org )

Compete traffic uniques = just 421, Site TBPR5, page TBPR 0, Page MR 5.33, page MT 5.89, DMT 5.64

Look at the co-citation of above pages. Linking out to WHO, NHS, TheLancet, Unicef, etc , etc

Do you really think getting a link on those pages would not help just because the domains have ZERO traffic or <500 uniques?

I feel they would be excellent pages to get links from despite their abysmal traffic.

On the other hand of spectrum, we have lot of sites run by the list building , direct marketing guys syndicate who have long page sales letter type 1 page sites attracting 40-50K traffic .

Doesn’t mean search engines love them or they are of high quality? I don’t.

P.S: Most of what I know about SEO / Link building is thanks to You & other excellent members at WebmasterWorld, SEL & it’s fun to disagree with someone I’ve learned so much from :)

It's fun to disagree in a civil manner. It's good to work these things out. :)

Let's have a look at your first example, HBSC.ORG:

a ) hbscDOTorg/contactlinks.html ( Health Behaviour in School-aged Children Org )

Compete traffic uniques ZERO! , TBPR 6, MT 5.73, DMT 5.78, MR 4.49, PA 34

HBSC.ORG
Quantcast traffic report shows zero traffic. Let's hear how the Quantcast crap-o-meter stacks up against the other stats. ;)

Yahoo site explorer shows me 665 backlinks and Google reports 154. However I think that data is not correct so it's difficult to comment on whether that data undermines what's showing in the toolbar.

In terms of quality, the backlinks are coming from a wide variety of CCTLDs and from many languages. Hmmm... Subtract the non-English links and the number of relevant backlinks goes down even more.

The backlinks are all on topic, but it's not much of a topic, children's mental health. So it's to be expected that the traffic levels are going to be low. It's not really a useful link in terms of affiliate or AdSense earnings. Is traffic an indicator of that? Ok, hard to say definitively.

ibfan.org
The Quantcast crap-o-meter [quantcast.com] shows zero traffic. Let's see what the rest of the stats show.

Google reports 201 backlinks. Yahoo reports 28 inbound links. Again, I don't have confidence in those stats. I don't know how far off they are, but right now both Google and Yahoo are showing significantly less backlinks than I suspect these sites have. So I'm not sure if those stats undermines the PageRank 5 of that site.

It is clear that getting a link from this domain is useful. Every link is useful. But if a link buyer is going to put a price on this link based on Mozrank or PageRank then I believe they are going to be overpaying and would be better off bribing several dozen blogspot bloggers. I would go with the Quantcast stat.

On the other hand of spectrum, we have lot of sites run by the list building , direct marketing guys syndicate who have long page sales letter type 1 page sites attracting 40-50K traffic

You should run some of those URLs in the Quantcast traffic meter. You'll be surprised at how low the traffic is on some of those networks, despite how much traffic they get, how many appearances the blogger has made on their local morning show and how many journalism and marketing degrees the publishers might have. If they're blocking Quantcast then run. They're hiding the obvious, that their comment spamming and reciprocal link networks aren't working.

In shortlisting potential candidates for sending outreach emails, i typically am looking at a large pool of sites. Hence I tend to base my decisions using the SEObook/SEOmoz /QSS toolbar factors mentioned earlier than individually rechecking their backlinks via YSE.

IMHO, the problem with taking traffic as leading sign of a quality link is that a large number of legit, high quality sites still don't have high traffic because they simply do not know anything about website marketing; so their content basically remains unseeen by most - doesn't automatically mean it's low quality site/link, many of the times, it's merely case of lack of active promotion.

I'd any day try to get a link from kind of sites i mentioned than most of .blogspot blogs purely for the TRUST rank they would be passing - for me that's quality :)

try to get a link from kind of sites i mentioned

Are you talking free or paid?

Would you really pay high traffic prices for advertising from a low traffic site? Paying for that advertisement or link changes the equation. ;)

Now we are drifting from the thread topic ? ;) lol

I don't pitch to trusted sites with money, i pitch such trusted sites with my rich content & how it'd benefit their visitors ( nvm that their visitors are very less )

In my opinion, a site that does not show traffic on these services, regardless of PageRank, has something wrong with it.

I would disagree with the statement.

I have recently developed links from very authority websites who do not have a ton of traffic. Why?

They have no idea how to optimize a single title tag.

All of their internal links are about as far from optimized as possible, and their inbound links are generally the actual websitenameasonelongphrase.com

Not to say this is always the case, but certainly it does happen.

Mozrank, Moztrust, Domain Moztrust, TBPR & MR comparison for page/domain level

These are not the most stable metrics to use either. I can give you hundreds examples of webpages with high MR /MT and pagerank scores that currently are not cached in Google.

I think the best method of evaluating links is really a wide scope of metrics.

More than that - it is often about intuition and gut instinct about factors not adding up on a domain, even when many of the metrics have showed positive signals.

I like the traffic signal idea. Although it has its limitations (already noted on this thread), it has to be better than the prehistoric TBPR, for example.