• Why the Minisite Boom led to today's gloom

    About 10 years ago savvy people were forecasting that web search would have to move from being "free" to "paid". Nobody could see how the service could be provided for no money, and maintain the facilities needed to give a search that would satisfy most users.

    That view was also held by a lot of web search company executives, so when Gargyl came around waving cheque books, it took very little to persuade them to sell out their stuff to Gargyl. But Gargyl had a plan. If they could control websearch, they could also control advertising served up at the same time. That looked like it might pay well - and it did.

    Gargyl started to make profits, but there was a fly in their ointment. Their search results could be shown to favour established sites over new sites, and very few people thought that was fair. Somewhere around 2007 there was some sort of public outcry about it. An inquiry, a court case - I don't remember enough, and if you try a web search for facts about what happened the results are buried in noise - what else would you expect?

    Whatever. The result of the hoohah was that Gargyl got officially told to do something that would give more advantage to the "little guy" starting up from scratch, and as usual Gargyl didn't say anything about it publicly. But they did make changes, as ordered, and people noticed the changes.

    The result was the Minisite Boom. Put up a new site with five pages of info, use it to get an Adsense account. Put up lots more sites with five pages of info, all running Adsense. Gargyl would boost those new sites in their search results, but for how long?

    It was anything from six to twelve weeks, they didn't hit really new sites but they did hit all the small sites with every Gargyl Dance review and put them in the sandbox. That's how it was perceived. Truth was that the sandbox didn't exist. Every site had a rating, but the rating was augmented by the new site boost. Take that away, and most sites didn't have enough going for them to rate well in Gargyl. So if you put work into a site in the time when it was new, you could avoid the de-rating effect of losing the boost.

    But how much work was needed? Personally, I never found out, because I didn't do enough to keep any site away from the "sandbox". I was writing good, thoughtful, and useful content for my sites, and it seemed to me that "content couldn't be king", because Gargyl just dumped them so far down the SERPs that pumping air to them was doing no good.

    I saw other people getting around the problem using automated sites that pulled content from other sites all over the web, but I didn't think that was "fair tactics". Another thing people did to survive was get links into their sites from other sites, links from high PR sites were desirable, and subject relevance was not an issue at that time. Sites needed links, people got links, and for a period the SEO industry boomed.

    Some good sites were produced at that time, but there were a lot of rubbish sites, some completely unintelligible, filled with Ingrish retranslated from other languages to the point where it no longer made sense (to avoid duplicate content penalties).

    After that there was a trend towards PHP sites which fooled the spiders into thinking they had lots of pages, although in most cases those pages only had a title. If you opened such a page, you'd get a message like "We don't have any content for that subject. Would you like to write some in this box?".

    I could go on. The point I'm making is that most of the crud that the Panda and Penguin updates have just removed from the Gargyl SERPs sprang up in the first pace because of Gargyl's assessment methods. They like lots of pictures, and sites with lots of pages. And videos - they like videos. And they like to see sites regularly updated with new pages, even if the subject matter of a site was exhausted after three pages had been filled (they don't like "evergreen", it reminds them of funerals). Beyond that, the rating process doesn't seem very subtle.

    So, why am I griping?

    A large part of the pre-Panda/Penguin crud that was on the web until recently was built to deceive Gargyl in the first place, and in many cases rated higher than some really good sites that got "sandboxed" a long time back for being too small.

    We're back at the stage where Gargyl can be seen to be favouring the big and established sites over the new and fresh sites again, but with a big difference:
    It is a lot harder for a new site to get started than it was five or six years ago, because it is a lot harder to manipulate the search results than it was five or six years ago.

    But if Gargyl was hauled up and called to account for the situation again, they can say "Hey, we tried to give new sites a big chance, it doesn't work".

    As I see things, it didn't work because they didn't work at doing the job properly ...
    This article was originally published in blog: Why the Minisite Boom led to today's gloom started by crabfoot