We all know that Google has incredible computing power and some very complex algorithms, both of which it uses to compile its Search Engine Rankings.
Google Uses Human Checkers Too
What you may not know is that they also employ humans to check out if their automated procedures are working as well as they should do, these people having the ‘power’ to knock a site off or down the rankings if they are not delivering the goods or are using spamming tactics.
There was a huge furore last year when Google’s instructions on how to rate pages (their quality guidelines) were supposedly published on the web. It was soon being placed on websites around the globe, but eventually Google managed to persuade all to take it down, so you won’t find a copy about these days.
However, the good news is that a copy was located and analysed by some experts. The bad news is that it did not really contain anything new, just the sort of thing that any good SEO professional knew anyway.
The Key Points
Search Intent is a Critical Factor
It is interesting to note that Google emphasises that the intent of the search query is the most important thing to consider when rating results.
It breaks down these ‘intents’ into 4 groups
- Know queries – When the searcher is looking for information and data.
- Go queries – Here the search intent is to find a particular page on the Internet.
- Do queries – Where the searchee is looking for a site on the Internet where they can accomplish something, e.g playing a game, purchasing something, downloading a file etc.
- The Combination query – Any combination of more than one the above.
Pages meant merely to inform will not be rated any higher than ‘relevant’ (one of the ratings) when the searcher is clearly looking to complete an action, a ‘Do Query’.
Most product searches are to be considered both Know and Do queries.
If the search intent is unclear in that it can be interpreted in more than one way, raters will rate pages that satisfy the one that is thought to be the most common one higher than those that satisfy any of less common ones.
This seems to be rather hard on the Raters, but here Google allows them to draw conclusion as to the most common search intent by looking at the other results in the search results where they are unsure which interpretation is the most commonly used.
The Rating Categories
There are several categories that can be assigned to any URL (the one in the Search Results), reviewers being told that the page’s usefulness (or utility in Google speak) to the searcher for any particular query is the most important factor when rating a page.
They are 5 categories in all to choose from:-
This is rating is only ever given when some very special criteria are met, normally it being associated with a person, or a particular business etc.
The dominant search intent is a ‘Go Query’ one where a particular page is sought and the URL matches the target (ie a page all about The Rolling Stones.
This is the highest that most pages can expect to get and to do this they must be considered as very helpful.
To reach this high standard the URL should be of high quality and a good “fit” for the search query used. Other reasons for including them in this category are that they are highly satisfying, educational, authoritative, entertaining (items such as breaking news are also considered good).
These ‘Useful’ pages have to be well organized and ones that generate trust, in that they are sources that appear to be reliable.
Useful pages are never “spammy” either (see later in the text).
Such pages basically need to do a great job in answering / matching the searchers query, but, in today’s world where Social Media is so powerful, they should also invoke trust, so that they are more likely to be shared.
These ‘Relevant pages’ are the ones that are helpful for a high number of users, but are not as well targeted to the search query as the ‘Useful’ ones.
They are of course still on-topic, but are simply not as good. Some reasons why pages are rated this way are that they may not be as comprehensive, as up-to-date or not appear to be as trust-worthy.
Pages ranked as ‘Relevant’ may rank well for less competitive keyword searches or for less competitive market areas, but often find it hard to get onto the first page
The good news is that if you are on the second or third page (and may therefore be in this category) that the page can very possibly be improved to rate as Useful with some small changes. More on this in a later posting.
Further down the list are the pages ranked as ‘Slightly Relevant’. Theser are pages that do relate to the query but are not very helpful, at least for most users as they are of low quality and / or lack of helpful information / are out of date.
It should be noted that pages that would be relevant but for their low quality and/or are very shallow content get this rating, which is a shame.
As to the matter of what ‘low quality’ is in this instance, in Google’s eyes this includes pages that have copied content or original content which is just too general to be useful or is not authoritative enough.
Copied Content is Not Always a Bad Idea
One very interesting point here refers to the matter of copied content. Whilst duplicate content has always been considered a bad idea, here it is considered okay, IF it is considered as useful.
In order to be considered however, copied content must be well-organized and appear to be designed to be helpful for viewers. If the page information is being shown just as bait, to display ads for users to click on, then it will not be considered as OK.
As long as the latter is not the case, then copied content will be rated on how helpful the content is.
This means that just because a URL looks like an exact match for a query, that fact alone does not automatically make it useful or relevant.
The upshot of all this that Google simply doesn’t want Slightly Relevant pages to rank well, so there is little point in wasting your time creating them in the first place.
Off-Topic or Useless
These are those pages that are unrelated to the query or helpful to only a very few users or perhaps no users at all. In the majority of cases, such pages will also be classed as Spam (see later posting). Pages that only contain links and ads and no actual content fall into this category.
These are pages are really the pits, basically being so poor that they just cannot be categorised as they are in a language foreign to the search term or because they don’t load.
These types of pages may well be reviewed later on by unkinder eyes and thus dropped from the index or pushed back so far in the listings as to be invisible.
That about wraps it up, but now you know how Google rates pages (when using humans at least) and that could well explain why some pages have good rankings whilst not being especially well tuned SEO wise, it is after all the content that counts in the end..