Deconstructing Google’s Trustrank Algorithm
29. Jan 2007 | 2 Comments
It has long been thought that Google has a Trust mechanism for each & every site it indexes. Just like there has been evidence of an Aging Delay on new sites (Sandbox).
I’ve seen it happen first hand with many clients & personal websites that have been starting from scratch. Yet with others, like Earners Blog I’ve not noticed any signs that could pertain to a “lack of trust.”
Firstly we need to know why Google is starting to play a lot of importance on Domain Trust:
- To reduce spam from making it into the main index
- To prevent webmasters leaking Pagerank between their sites
- To push less relevant pages into the Supplemental Index
- To reward Webmasters for much more relevant & topic specific links
Lets face it, Google is going to have a hard time trusting a brand new domain. No matter who owns it. From my observations there seems to be a pattern of events that may reduce your sites ability to gather trust quickly (hence the aging delay or sandbox type effect):
- Too many links too quickly (i.e. 1000 links in the first day can look *suspicious*)
- Too many links with the same Anchor Text (Sets off instant spam red flags)
- Too many pages within the site initially (I prefer to grow sites rather than place everything down at once)
- Not enough links (Not promoting the site at all)
- No Topic Specific links (I.e. links coming from off topic websites)
- Too many thin links (Links from Directories, Blog Comments or Social Networking Sites)
- Links from the Same C Class IP (Has been disputed but I’m more than sure G keeps track of who owns which site via whois info)
- Duplicate Content (Copying Articles from Article Directories etc)
- Too much Recropical Linking (Especially if Off Topic)
- Outbound links to Unrelated Sites
These are all classic Reg Flag scenarios for Google. Not always, but lets not be complacent enough to think you’re immune or they “won’t detect anything.”
For example I read a forum post today about someone who had created the ultimate spam promotion tool. It does all of these methods automatically:
trackback spam
comment spam (filters out blogs from msn, blogger, and free wordpress blogs)
guestbook spam
forum spam
blog and ping (uses blogger blogs and free wordpress blogs)
referrer spam
rss injection (inserting html link into rss feeds so any feed searches or directories that don’t properly sanitize feeds will give me a backlink)
It’ll generate 10,000 links to one site within 24 hours…..
I closely followed the thread & within 6 days a brand new domain with 8000 pages had been fully indexed by Google. However he’s still only getting a mere 25 unique visitors per day (he didn’t mention what niche he was targeting).
Whilst these sort of tactics may be good for the short “hit & run” type website on a .info domain, taking a new website from scratch & making it into an authoratitive domain will take a lot more than a load of link spamming with the same Anchor Text to multiple sources.
So lets look at some factors that Google might look favourably upon:
- Growing your site gradually (Growing slowly at the start then picking up the pace)
- Growing your links gradually (2 per day for the first week, then 3 per day & so on)
- Attracting Topic Specific Links (Blogs are perfect for this)
- Outbound Linking to Authority sources within pages (Lets G know you’re talking about a specific topic that’s relevant to the on page copy)
- No Affiliate Links (I’ve had a sneaking feeling that Google will check new sites & devalue them if they are full of Affiliate Links)
- Domain Age (This plays a HUGE importance in Trust Factor, any domain that’s pre 2005 has a huge advantage)
Now, how do you know whether your current site is well trusted or not already? With the latest Link Buying fiasco Google is now (according to Mr Cutts) starting to devalue a sites ability to pass Page Rank & Trust if it is brokering text links openly via services like Textlinkads. Whether this is true or not is hard to tell but lets look at example of a site that was smacked for doing this not so long ago, W3C Supporters Page Anyone? But don’t you find it Ironic that Google puts a rel=nofollow on all Google Video links except those from its sponsors?
(I guess that’s for another post).
If your site is well trusted it’s very likely that you will notice things like:
- You’ll keep a steady amount pages in the main index
- New Pages will get indexed quickly (within 24 hours)
- New Pages will rank quickly
- Decreased number of pages in the Supplemental Index
- Spiders will follow & trust your external links
Whilst the word of the year is “links” the thing to remember is that a Page Rank 6 link may be worth less than a very targeted on topic Page Rank 3 link. This is I guess another awesome argument for creating very targeted Linkbait.
So remember, not all links are created equal. Nor are all websites. You have the ability to make or break your site in Google’s eyes. Whichever path you choose ![]()
Enjoy this post? Get the RSS FeedTags: Search Engine Optimisation or Share This
|
| ![]() |
2 Comments on "Deconstructing Google’s Trustrank Algorithm"
From my experience I think you are pretty right on. I was able to get a lot of IBLs fairly fast on Myspace and StumbleUpon but it took a couple of months to get quality links from sites with the same content. Though, I saw that Google responded much better to the quality links, than those from Myspace and StumbleUpon.
Though, after only 6 months of launching the site, I have a PR of 3 and I’m on the first SR page for many of my keywords.
Add a Comment
Categories
- Affiliate Marketing
- Affiliate Marketing News
- Affiliate Networks
- Copywriting
- Earners Blog News
- Earnosphere
- Email Marketing
- General Musings
- Guides
- Keyword Research
- Link Building
- Marketing Tools
- Monetization
- Outside the Box
- PPC
- Productivity
- Reviews
- Search Engine Optimisation
- Social Media Optimisation
- Tips & Tricks
- Top Posts
- Traffic Generation
- Wealth Creation
- Web Development
- Wordpress





while I generally agree here, the point you made in #3 - Too many pages within the site initially, I think is overrated.
The number of pages in a site is not as important as the number of pages spidered. If I build a 1000 page site intially and google only spiders 23 pages, then effectively it’s a 23 page site.
Now, if you have a single sitemap with 1000 links on it and G spiders the sitemap it is certainly possible that with the page it “sees” 1000 links and can add a negative trust factor.
But if you have a blog cms for instance and publish without pinging 1000 pages into multiple categories, etc. it is likely that gbot will only spider one or more categories initially, again effectively reducing the size of the site it sees.
Excessive pinging showing rapid content creation seems to be much more of an easily detectible trust issue in our experience.