Google Sandbox: solved?

The Google Sandbox is one of those contentious issues in SEO - some people deny it’s existence, while even those who advocate its existence don’t understand what is going on.

I set up a public record of the Google Sandbox announcements, and how the story broke into the wider SEO community here: The Google Sandbox guide. I also posted up an example of ranking statistics for a link building client for whom I was building a few thousand links pre day for here: What the Google Sandbox looks like.

Google Sandbox: the mystery

One thing had really puzzled me about the Google Sandbox. I was building extensive networks of links for clients, but seeing the effect of anchor text impact delayed quite significantly. However, people using the free Digital Point Co-op Advertising Network - effectively a massive-scale dynamic link exchange program - were reporting quick ranking results. I joined the network myself, and my test keyphrases climbed within days.

How on earth could large scale themed-link building be failing, but widespread off-topic reciprocal linking be working?

That question has bugged me since I started. It didn’t make sense. Unless…

Google Sandbox: revealed?

There is one variable in the entire link building equation that I now suspect is key to the entire matter: PageRank.

It occurred to me this morning that PageRank was the only variable that could possible explain why Shawn’s free Co-op Advertising Network could provide results within days, while my paying link-building clients were squarely Sandboxed.

Why?

Google Sandbox: PageRank still applied?

We all know that looking at our Google Toolbar PageRank indicator is pretty worthless these days. This has lead a lot of the SEO community to pretty dismiss PageRank entirely.

However, logically speaking, if the Toolbar is showing unreliable values then that does not invalidate the very PageRank supposedly being measured - merely the way in which PageRank values are publicly released and displayed to the public.

As reported, though, in Google PageRank just for fun a Google AdWords rep is alleged to have stated that “Google updates the PageRank data very infrequently…the PR that is displayed in the Google Toolbar is several months old.”

No great revelation there - excepting if you read beneath the lines of the entire quote in the link, you have a tacit suggestion that Google is still applying PageRank, but is trying to avoid displaying any kind of useful PageRank data for webmasters to use.

So how might PageRank be the culprit for the Google Sandbox?

Google Sandbox: How it might work

The clearest different between the Digital Co-op Network and my own is one of PageRank. The Co-op network relies on people offering reasonably high PageRank pages on a sitewide basis, whilst my own link building offers clients reasonably low PageRank pages on a sitewide basis.

Now, John Scott stated that he had inside information on a new system being applied by Google - that of “grandfathering of links”.

The question is, how were links “grandfathered”?

My speculative answer would now be that PageRank is one of the key engines behind grandfathering itself.

How?

Simple - those links with highest PageRank are “grandfathered” quickest - whilst those links with lowest PageRank are “grandfathered” slowest.

In practical terms - high PR links count quickly; low PR links take a long time to impact - and all based on their respective PageRank values.

This key hypothesis would explain precisely why Shawn’s Co-op network can significantly impact the SERPs within days - there are an untold number of higher Pagerank pages - anything up through the 5s, 6s, and even a few 7s - which are all stamping their own “authority” on the new links.

And this precisely addresses the issue of why sites that might be regarded as having “authority” on the internet are able to have their own links indexed and ranking very quickly. Major news sites like the BBC, CNN, and CNet, for example, can all - through natural structuring - point a lot of pages with very high PageRank at new URLs.

Of course, PageRank is almost certainly not acting on its own - as John Scott originally pointed out, other factors come into play, such as IP range, and age of a page and the age of the links themselves may well all play some role in this - the big question of course being to what degree.

After all, one of the major advantages of the Digital Point Co-op Network is that it covers a very extensive range of IPs. So do I - it has to be standard link building practice. However, the IP spread would not explain the issue of fast-tracked rankings alone if all links were in themselves sandboxed - unless a further qualifier was introduced to deal with this - such as PageRank for the purposes of “grandfathering” the links. More specifically - PageRank as determining the authority of the links themselves, and thus how quickly those links should impact ranking results fom the Google index.

Google Sandbox: Suggestions to avoid it

PageRank is built on the concept of “recommendation”. This means that to help avert the Google Sandbox, link builders would have to take this even more in mind.

For example, get as many higher PageRank pages to link to your target pages as possible. And don’t stick to one domain or IP range - ensure that the links are from as many different sites as you can budget for, to make your recommendation stronger.

This is basic link-building practice, but the point to be made here is that you would need to emphasise building on high PR pages, rather than going for links simply in number - if you wanted to see results avoiding sandboxing delays.

That means that single page text-link advertising, across multiple sites would, help serve to deliver better short-term results - whereas sitewide links resulting in large numbers of additional low PageRank pages would simply serve as an investment for the longer term (though bearing in mind potential devaluation of links in high numbers due to Hilltop and LocalRank).

Another way to help with “recommending” new links would be to use dynamic feeds. By inserting XML feeds, such as RSS, for dynamic internal content - forums and blogs, for example - across the major pages of a site, you strengthen the “recommendation” for your particular target pages linked to. What is even better is if you can get those feeds syndicated from other sites, because then you have multiple sites making the same quality recommendations.

Ultimately, the issue becomes one of considered linking, and attempting to get links on pages according to PageRank value, rather than sheer numbers - or even topic - first.

This, of course, is precisely what Google wishes to frustrate - so until we can reliably gauge PageRank values of specified pages, then it’s going to have to be for webmasters to gauge the value of pages for linking purposes based on a combined judgement of old PageRank data - along with some common sense and creative thinking.

Is the Google Sandbox solved by this hypothesis? I’m not convined it would be wise to claim so - but I do suggest the idea to the wider SEO community as some way to explaining what is actually going on, in a way that makes sense to all SEOs when we use the term “Google Sandbox” and “sandboxing”.

It especially helps explain to extensive link-builders why their low-PR links do not impact for a long time, and to content-focused people what on earth the link builders may be harping on about. It also helps explain a curious anomaly - someone I know who is a reciprocal link exchange specialist, but also saw no effect, which would be completely explained if part of his remit involved getting reciprocal links from sites with higher PageRank.

Regardless of the speculations and interpretations doing the rounds on the SEO forums, for the time being, I’m pointing my finger squarely at PageRank as the cause of Google Sandboxing. PageRank, that was so cleverly hidden from us until we thought it had gone away and died, only to find that it is still very much at the helm of Google - and pulling it’s weight in new and novel ways to vex link building webmasters and frustrate their efforts and patience.

Discussion on this in the Platinax forums here.