Metrics to Use for Calculating Price of Links

Meaningful Metrics for Determining the Appropriate Cost of a Link

[martinibuster]

This discussion is not asking if you use paid links. This discussion is not about whether paid links are ethical etc. It's just about what it says in the title, what Metrics should be used to judge the worth of a link?

Obviously, PageRank is a method many use, but I'm very skeptical of it. Having scanned websites for asking for free links, in certain niches I see the same sites plugging their anchor text with paid links. When I run their anchors on Google a lot of the time they are not ranking anywhere. My personal experience in obtaining free links from extremely low PR sites is that they do the job. Might take up to six months before they work, but they work.

At bottom, I think the same criteria for judging the worth of a paid link goes into the worth of a free link. What is it about a particular link that will make a site rank better? What are the criteria for judging the worth of a link?

Traffic as a metric
There are some webmasters who use PageRank as the basis for how much to charge. Some are asking from $700- $3,000 per year for a link off a PR 4 or 5 site. When you run their domain name through Quantcast and Google Trends for Websites it is very obvious their traffic levels are low. I know Quantcast is not accurate, but it gives an idea of traffic levels, especially when used with Google Trends.

So why is traffic level so important? I think it's important because it is an indicator, a metric, of how relevant the site is for certain keyword phrases. If relevance is the holy grail, then obtaining links from relevant sites is important. So I would suggest traffic levels as a metric for how important a site is if you are going to put a dollar sign on it.

Feel free to disagree. The above is only my opinion based on my observation. Your observations may be different and I would like to hear those. In fact, I will disagree with myself right now, I will offer a counterpoint. It is possible that what the site is relevant for might not be the big phrases, but if you look at the referral logs the core components of the keyword phrases are highly relevant to what is important to me, albeit in longtail form. So perhaps the traffic level doesn't tell the whole story? That maybe we should skip it and look at the keyword phrases used to find the site?

Or can the level of traffic still be used as a quality metric of how important (and effective) a link is?

Any other metrics to add to the list?

[FranticFish]

I don't see why traffic level wouldn't be important, even if Google can't always measure it effectively.

Related to that, how about how much traffic the link itself brings? A link that gets clicked a lot must surely be a useful / interesting / popular one.

I like the idea of both of them, but the problem to my mind is that Google can't extrapolate data from the sources they have to guess at what they don't know. However that might not necessarily bother them. If they can give certain links between sites an extra boost because they have all the data to do so, then they'd be boosting sites that they know people like.

There's certain links I buy for people from time to time, trade directories and the like. I still look at PR too (of the page, of the root), then I also look at

- the link profile; how many, what sort etc. Then also
- how many pages site has in Google, cache date, and
- rankings.

[buckworks]

how relevant the site is for certain keyword phrases

Up to a point I'd agree with that, but when I go link hunting my biggest focus is the relevance of the PAGE that my link would be on. That could easily have some nuances that are different from the site as a whole.

In my opinion, if you focus on obtaining links from places where the direct human traffic would be well targeted for what you're promoting, such links will send useful signals to the search engines. The strength might vary but they'd be the right signals.

[steve40]

I would think number of other outgoing links from the same page would be an extremely important factor, link on a page about widgets that gets reasonable traffic ( Implying reasonable ranking ( PR in calc but middling factor ) and just 2 external links to other widget based sites is more valuable than similar page with 10 other external links including some to widget sites. for both ranking and traffic purposes.

[FranticFish]

For PR certainly, and perhaps also for anchor text (i.e. more links = perhaps a wider spread of anchor text and subject matter, so less relevancy for any one link) but would any trust / localrank factors be affected by the presence of other links?

[CainIV]

I think traffic is one metric, and overall score of a link is another. Both are valid.

Traffic is fairly easy to measure, and ROI can be ascertained from that segment relatively quickly. It is link value that is more difficult to 'eyeball'

Personally, I am building my own tool for this which, will not take the work out of making final decisions, will be able to score links based on "assumptions" I have about link value (contextual, age of domain, number of inbound links on topic to that page, etc)

[martinibuster]

Some sites inflate their backlink counts with comment spam from a variety of blogs. The links show up in a backlink search but based on the lousy ranking/traffic of sites I have seen doing that, it's not going to be a useful link- regardless of what the PR meter shows.

[ken_b]

Is this discussion about ...

Metrics to Use for Calculating Price of Links

or

what Metrics should be used to judge the worth of a link

Cost and worth (or value) can be very different things.
.

[CainIV]

So why is traffic level so important? I think it's important because it is an indicator, a metric, of how relevant the site is for certain keyword phrases. If relevance is the holy grail, then obtaining links from relevant sites is important.

Martini, are you theorizing that the traffic level to a given page for keywords x,y and z might provide a SERP boost since Google assumes the page is attracting alot of traffic? (regardless of specific rankings)

[martinibuster]

Cost and worth can be very different things indeed. Worth is subjective. The cost of something is generally a price tag (unless it's at auction, where it has a starting price tag). The cost of a link is determined by the subjective worth. What (subjective) metrics are useful to calculate the worth to arrive at a cost?

Traffic level is a metric, one of several I use, in order to calculate what something is worth. If a site cannot attract enough traffic to cause a blip on Google Trends or Quantcast, what does that say about the power of that site's backlinks? What does that say about that site's importance within it's niche?

A metric like the quality of backlinks, to me, is more important than the traffic metric. A metric like the quality of the outbound links is more important to me than the traffic metric.

I am not saying that a site with miniscule traffic is worthless. I like low traffic sites with 30 honest links, I love those sites. I am using multiple criteria. I don't use the traffic criteria to decide if a site is worth less. I use it to decide if a site is worth more. Good traffic stats enhances the worth.
It's like, imagine you're putting together a team of fighters for an island death match. Do you pick the six foot weightlifter guy or the little guy who weighs 125 pounds?

How does the equation change if the little guy is Bruce Lee? There's nothing wrong with the six foot weightlifter, the guy with low traffic. Maybe his back isn't what it used to be and his asthma is acting up. How does that change the equation?

The backlink and outbound link metrics are metrics I use to determine if something is worthless, not worth bothering with. I don't use the traffic criteria to decide if a site is worthless. I use it to decide if a site is worth more.