By Jim Hedger – October 18, 2006
The mysterious “Google Sandbox” has been a hot discussion topic for search engine optimizers since the phenomena was first noted and named in late April 2004. Since then, ideas on the function, scope and even existence of the sandbox have been a mainstay in SEO and Google related forums, chats and articles.
The term, “sandbox” describes a process of Google’s ranking formulas that appear to slow the debut of new sites in the Top10 listings. Whether or not the sandbox exists tends to depend on the side of the black hat / white hat debate one comes from. SEOs dedicated to “pure”, non-spammy SEO tend to downplay the effects of the sandbox while those who use dark-art tactics know from personal experience that the effects are very real.
In a November 2005 interview with WebmasterWorld’s Brett Tabke, Google chief engineer Matt Cutts appeared to acknowledge it saying, “How many feel there is no such thing as a sandbox? SEOs normally split down the line. There are some things in the algorithm that may be perceived as a sandbox that doesn’t apply to all industries.”
While Matt’s comments did not exactly confirm or deny the existence of a sandbox, they did leave the barn door open wide enough for a great deal of discussion. Between November 2005 and September 2006, well over 300 credible articles, forum threads and blog posts have addressed the topic.
For some, such as Search Engine Guide editor Jennifer Laycock, the Google sandbox is an expression of importance and credibility imbued upon sites that have established themselves over time in Google’s index. In a June 2006 article, Jennifer wrote, “… there is no Google sandbox! It simply doesn’t exist.”
When new sites are spidered and brought into the index, Google compares them with similar sites already in its index and makes a judgment on the relevance of that site against the others. “After all,” she wrote, “how many mortgage application sites does Google really need to list? Why should they think that your brand new mortgage site is any more worthy of a ranking than the 1.5 million (yes, MILLION) sites that are already indexed for the phrase ‘mortgage application.'”
Jennifer wrote her piece in agreement with a post Sheri Thurow from Grantastic Design made to LED Digest (issue 2177) where Sheri proclaimed, “There is no such thing as a Google Sandbox. It’s one of those terms that self-proclaimed search engine “experts” came up with to explain why their methodologies don’t work.”
Sheri’s post generated a great deal of controversy, partially because it was perceived by readers as a rant. She explained her point by writing, “A search-friendly Web site is built on a solid foundation of keyword-focused text and giving spiders a means of accessing that text. Then, objective 3rd parties should basically confirm what you say about your own content. It has been this way for years. For that reason, no Google Sandbox. If your site doesn’t have the foundation, its pages won’t appear in a search engine results.”
What Jennifer and Sheri are saying is Google is not going to automatically give a first page listing to a new site, no matter how well optimized that site might be. There might be thousands or even millions of competing websites being compared against a site that has not had time to establish itself as a peer-referenced site.
Peer referencing is an important cornerstone of Google’s overall ranking formula. Originally, Google built its search results based on the number of links directed to a specific document from other web documents. As technology progressed, the creation and maintenance of web documents became infinitely easier, especially with the deployment of blogs. Driving the creation of millions of fresh documents, sites and blogs was the popularization of Google’s paid search advertising distribution program AdSense. Suddenly, a direct profit motive existed for some in the SEO sector to use their knowledge and immense talents to game Google’s SERPs six ways to Sunday.
Links, which have long been like gold for search savvy webmasters, became increasingly important following the series of algorithm upgrades that started with Hilltop in the summer of 2004. Since that time, Google has raced to keep up with a myriad of methods devised to game its ranking methods. The implementation of an critical measure of how a site, or a document originating at a specific domain, has established itself along side similar sites and/or documents, is accepted by most SEOs as a long leg in Google’s race against manipulation of their search results.
Google has consistently moved to limit link-based exploits of its link-based organic ranking system. Starting this time last year, it implemented a series of measurements examining a wide assortment of data about documents in its index, along with data derived from linking documents known as the Jagger update.
It is safe to say that for the past two years, the majority of research and advancement in the field of search engine optimization comes from the study and analysis of link structures. When it comes to getting a strong ranking on Google and to a lesser degree, on Yahoo, link partners are as well scrutinized as the specific page or domain being indexed.
One of the newer clichés being thrown around SEO circles is the phrase “Link Baiting”. Links from well-established, relevant sites is good at Google. Getting good quality links is a getting harder every day. Link baiting describes a tactic to entice others to link to your site by presenting enticing content.
One of the best-known and well-respected link analysts, Andy Hagans wrote a strong piece earlier this week titled, “Secrets to Beating the Sandbox 2.0 REVEALED: The Ultimate Guide“. Though the title is in itself a fine example of link-bait (as noted by RustyBrick), the content of the piece is extremely thorough and well written.
In his opinion, the Google Sandbox definitely exists. “The sandbox/Trustbox is a set of filters in Google’s search algorithm that together prevent new sites from ranking well until they gain trust.” Andy redefines the debate by suggesting “Trustbox” as a far better name than “sandbox”. It is certainly more descriptive.
Andy’s “Ultimate Guide” opens with a concise tutorial on what the “sandbox” is, how it effects sites, and methods he uses to dig through it. As with most pieces by Andy, he clearly delineates tactics that might be considered black-hat from those that are simply smart work on the part of webmasters.
Though I earlier noted that those who believe in the sandbox tend to fall on the dark-arts side of the SEO sector, I don’t intend to suggest all discussion about the phenomenon comes from black-hat SEOs or to imply that all those who find themselves playing in the sandbox are spammers. Due to their need for rapid placement and their propensity of burning domains, the function of a Google Sandbox affects black-hat (or, more appropriately, self-servicing) practitioners far more directly than it does in-house or agency focused SEOs.
I don’t mean to disagree with Sheri or Jennifer either, not exactly anyway. I suspect that for them the existence of a sandbox is irrelevant and therefore far off their radar screens. It is not in Jennifer or Sheri’s interest to spend time delving into the dark-arts of SEO as neither practice black-hat techniques.
Something resembling the Google Sandbox, or as Andy calls it, Trustbox does exists for new websites. Jennifer Laycock actually summed it up best herself, speaking in a session about small business at SES San Jose, in which she used the opening of a new restaurant to describe the non-existence of “sandbox effect”.
In the analogy, Jennifer explained that she was a fan of Chinese and Ethiopian cuisine. In her mid-west city, there are hundreds of Chinese restaurants, several of which she has eaten at or regularly frequents. The opening of a new Chinese restaurant might register on her radar screen but would not likely cause her to race out to try it. If a number of friends recommended the new restaurant, she would be more likely to try it out. Chinese restaurants are ubiquitous in North American cities. If a new Ethiopian eatery opened in that same large mid-west city, she would be very likely to go out of her way to eat there, with or without recommendations from trusted friends, simply because there are very few restaurants specializing in Ethiopian food.
In other words, Jennifer’s observation brings us back full circle to Matt Cutts’ original affirmation, “There are some things in the algorithm that may be perceived as a sandbox that doesn’t apply to all industries.”