Google algorythm ain’t what it used to be
Monday, September 15th, 2008Google has been discussed a lot this year based on the private document that has been readily available on the internet entitled “Google Guidelines for Quality Raters”.
Ok, to the general public, the Google Algorythm is computer-based, not human-based. Otherwise, how could it possibly be objective? Well, it can’t and so it was inevitable that Google would need to step in and ‘help’ the SERPs along. Does it bother designers and SEO-professionals to think that all of their hard work is open to the interpretation (assisted by these handy guidelines) of some underpaid, english-as-a-second-language workers outside of the US (no proof of this, however, how else could Google afford it?)…yes, I believe it does. And here is why:
After reading the document, it seems that it is assuming that the Quality Rater assume a lot - about intent, specifically. What is the user intending or hoping to get when they enter a particular search string/query? Humans are to a large degree the sum of their experiences - what I am hoping to get when I type something into the search engines is going to be a lot different from what my neighbor hopes to get. It’s insane to think that it could be otherwise.
Now, I understand the intent behind this - to reduce the spam and crappy results that were beginning to plague Google.com. However, is this the best way to do it? I don’t think so. Also, there has been quite a lot of talk of how Google is always protecting Google’s best interests. SEO Book puts it nicely here.
Ignoring the hub-bub, how does this apply to designers? I think there is a lot to learn based on the private document, first of which would be what Google has been saying all along:
- Provide excellent, relevant content and don’t try to spam the search engines
- Use the guidelines Google puts out to build your site to the best search-ability
If you want to read the entire document, it can be found in several places online. I’ve summed up the most relevant points here:
Query Types:
- Navigational: Looking for a site, such as a search for IBM.
- Informational: Looking for information
- Transactional: Looking to make a purchase
Quality Rating Scales: This seems straight forward when managing navigational queries but when faced with the other 2 (informational, transactional) it seems way too open to interpretation to me.
- Vital: The vital result comes from a query that is most likely navigational and the resulting page is the official web page of the query. Search for ‘google’ and the vital result would be google.com.
- Useful: A useful rating would be for results that “answer the query just right; they are neither too broad nor too specific.” One of the examples given for a useful rating would be a search on meningitis symptoms with a resulting web page of http://www.webmd.com/hw/infection/aa34586.asp
- Relevant: The guidelines say the result is often “less comprehensive, come from a less authoritative source, or cover only one important aspect of the query.” So, for example aAn example would be a review of laptop computers, but the review only takes five computers and not all computers within its class. Since it is not a fully comprehensive review, it would be rated as relevant and not useful.
- Not Relevant: This rating is used for pages that are not helpful to the query but are somewhat still connected to the original query. Classifications of a not relevant page would be “outdated, too narrowly regional, too specific, too broad” and so on. One of the examples give is a search for the ‘BBC’ that returns a specific article from BBS; it is too specific and is not relevant to the query at hand.
- Off-Topic: This is the lowest rating a page can receive for a query. If the returned page is completely not relevant to the query, it would be given a rating of “off topic.” An example given is a query on ‘hot dogs’ that returns a page about doghouses.
Spam Labels:
New to this guideline document, this is interesting and relevant:
- Not Spam: Given to a page that “has not been designed using deceitful web design techniques.”
- Maybe Spam: Page might be “spammy,” but you are not 100% convinced of that.
- Spam: Given to pages violating Google’s webmaster guidelines.