Forum Moderators: Robert Charlton & goodroi
Details
After the Bourbon update, the websites that were doing well, maintained their positions and the new, sandboxed sites came from oblivion to first page ranks. To my delight, no website was adversly affected and this is the case with about 25 websites from different domains that I handle.
Diagnosis
Well, this is what I/we do for any website and there is nothing extraordinary and complex tricks involved in implementation.
1. Content
Contrary to many who suggest to keep writing new content each day, most of these websites have almost the same content as was a year ago. Few websites are under 10 pages with mostly the details about the company, than the subject. But the content that is there is unique and complete with word stemming, semantics. KW density? didn't bother much.
2. Design
Just a search engine friendly static websites, with judicious use of CSS and sparing use of ALT text.
3. META Tags
Title attribute gets the maximum importance and description with words from Title in stemmed form and some semantics. Keyword tag: don't have all the words from Title and some new words that are not there in Title. Have an intution, if not for giving weigtage to Keyword tag, to look for over optimization, SEs may use keyword tags. But largely feel it doesn't matter if you use Keyword tag or not.
4. Internal Navigation
Pretty innocent navigation, like Home, About Us, Contact types. Regular use of interlinking between pages from contextual words in content.
Overall, On-Site factors get about 10-20% of importance in our SEO plan.
5. Inbound Links (IBLs)
The major part of the whole plan. Part of the links achieved to websites is done as a 3-way. Part of the links achieved are paid for directory links and the last part is from direct link exchanges.
Never used just keywords as anchor text, but along with the unique company name and randomized a bit. PR is used only to check if Google has indexed the page but relevency of linking page is very important. Regularly try to get links from pages that themselves rank for the target search terms (Easy way of finding relevant, authority pages).
No automation done in linking part, but hard, manual work. Not more than one link per domain in most case.
No. of links isn't a big factor, since for many keywods where we are ranked, our websites have the least no. of IBLs.
Few sites have outbound links to relevant authority sites, few haven't.
6. Other webmaster issues
Employed as per need. Where Google had indexed both canonical versions of websites, employed 301.
Employed 301 from any duplicate sites. Most websites are not registered for more than one year for their domain names. All are on different servers and few are on shared IP and few are on dedicated IP. None on dedicated server. No client sites interlink, so no issues of C-class etc.
[b]Further info on sites[b/]
No 302 hijacks.
Industry varies from Mortgage, to Technology to Shopping, to Travel.
Competition: Ranges from 200,000 to 350 million in Google results.
Keyword popularity: Ranges from 100 to 30,000 as in Overture.
I have tried to compile the featurs which I thought will play a role in success. The inferences are upto you to make, such as LSI, Trustrank, Hilltop and more.
I am sure the analysis will never be comprehensive but hope this will provide some fodder for thought.
Best Wishes
so better keep your greetings and cheers for your self
Oh, I should have kept mum, than try and share some info. Wouldn't that be rather selfish? I am not saying that I am great or trying to promote myself (I don't even have my website in my profile). Its upto you to take it or leave it.
Is this the general feeling here guys? Then please say so, so I won't waste my time in posting such threads.
First, thanks for starting this thread.
News.... Just as much it is important to know what caused failure, I guess it is also important to know how your success came about, so you can try and repeat it.
I agree. Those of us that have somehow made it through update after update might be able to share something that others find interesting.
Details
While I have a single site, it has somehow managed to survive every update since I joined WW. And it continues to gain in serps and traffic from Google. Something must be working.
1. Content
... almost the same content as was a year ago. .... But the content that is there is unique and complete with word stemming, semantics. KW density? didn't bother much.
I haven't bothered with semantics. Stemming has played a role on some pages, not most, and I don't give it a lot of thought. I don't keep tack of keyword density, but on some of my pages it's probably 25% or more, on others it' probably very low. Just depends on the page.
2. DesignJust a search engine friendly static websites, with judicious use of CSS and sparing use of ALT text.
No CSS on my site. Just plain old poorly written html code, all pages are static.
3. META Tags
I do have my pages titled individually, use meta keywords and discriptions, and Hx on page titles. My page title, description and H1 are often identical, or at least very close.
4. Internal Navigation
Pretty innocent navigation, like Home, About Us, Contact types. Regular use of interlinking between pages from contextual words in content.
Same here
5. Inbound Links (IBLs)
With 2 or three exceptions for reciprocal links, I've pretty much ignored getting links.
I link out a lot (I think) if the other site wants to link back that's fine, but I don't ask.
6. Other webmaster issuesEmployed as per need. ...employed 301.
I did a 301 back in March, and also converted all my internal links from relative to absolute.
The only effect I've noticed was a very fast switch in the google serps so they no longer show the non-www version.
[b]Further info on sites[b/]
I'm probably playing with fire here, I do have some pages that are 302 hijacks (as I understand the issue) but haven't bothered trying to correct that.
I'm not in what I'd think of as a very competitive niche, maybe a sub, sub niche of automotive. Think photo gallery for the most part. I'm on the first page for terms showing anywhere from less than 50 results to more than 10,000,000.
I do have 1,400+ pages, all added together they are attracting 1,000,000+ pageviews a month for the site total. Most of my traffic enters through lower level pages. I'm pretty sure less than half my traffic comes from search engines, but that goes up just after an update.
I do add pages, usually in clusters of 20 - 50 or so at a time. I update several time sensitive pages as needed, sometimes 5 a week, sometimes 25 a day.
Is this the general feeling here guys?
NO! I read everything you wrote with great interest, as I have ken_b's follow up. It is always useful and valuable to learn what is working for people, as it is to learn what didn't work for people.
There will always be those who prefer just moan about their 'bad luck', and be personally affronted by someone else's good fortune and successful effort. It doesn't seem to me to be a very constructive or positive way to approach the world but it's their choice.
FWIW, I haven't seemed to have any serious impact up or down from any of the updates, ever, I'm always just floating at the surface. Now that I've said that I'm sure to be doomed and will sink deeper than the Titanic never to be seen again! :)
After that, I decided to play it cautious and not change anything on the site. I've only added a couple of links, and have added a new page of content every week or so.
The Bourbon update seems to have bumped all my pages by a couple of spots, i.e. from #9 to #5 or #6.
I swear there's no rhyme or reason.
I'm also wondering if judicious on page, in text, internal linking plays a part as opposed to just a navigation bar
It'd seem to suggest that the author's considered their documents more carefully, rather than just pumping text into a template, so it should be a sign of quality.
vigo, It is best to learn wisdom by the experience of others.
My site also improved a little during the update. Nothing special. Simply earlier, searching for my main specific keywords I saw first download sites that are linked to me or other sites. Now I see me first.
I am not a webmaster (I am software author), so it is not self advertising. Just to confirm that the advices of McMohan and ken_b are good. Thanks.
Vadim.
McMohan, good post, ignore the sour grapes.
Yes, I agree. The positve sharing of info is what this site is about.
I can't help thinking that some of the negative critics in this thread, who do not share info, are not doing very well because of their overall negative and sour attitude.
I link out a lot (I think) if the other site wants to link back that's fine, but I don't ask.
PS - This has been my experience too.
Do not underestimate on topic outgoing links. Now that PR is not as important, these guys are very much underrated!
I am not resourceful enough to objectively analyse and come to a conclusion - what is working and whats not. So, I had stated "Its upto you to make inferences" in my post, and have only given what we did aross all the website for aspects that we thought have a role in SEO. Just the observations. There is no conclusion and summary such as - "so, to be able to succesfully rank your websites, you should..." no.
No blackhat techinques, all perfectly honest stuff and this is in one of the most competitive industries.
there's been a lot of discussion by people who didn't do so well, but it's great to see something from those who survived (thrived on?) bourban.
every time i come into ww i find something that wouldn't have occurred to me, this thread is certainly helping me to take a fresh look.
eg. i never thought that linking within my text would be useful for seo, i just did it for my viewers, but now it seems fairly obvious. ;)
tallis
Your posts reflect the true spirit of WebmasterWorld as it was originally intended. Thanks.
We had some sites removed from the serps in that update and others which stayed. There were dozens of theories on WebmasterWorld about what the common theme could be, but each theme could not explain at least why one of our sites was either in or out of the serps.
The fact is that google has so many factors at play, and they vary them so often, and also throw in a random element to derail anyone smart enough to run linear regression to try and work out what counts.
That was a year and a half ago. I stopped trying to second guess google at that point. Now it can only be harder.
I admire you guys for trying to work out what's what, but your valiant effort will only result in the conclusion that you just need to build a site for the user with minimal SEO which is deemed white hat, and perhaps go out and get some backlinks to your landing pages.
There is no holy grail - the more black hat you go the more you risk a penalty. It's just a tradeoff now, and you need to pick your spot on the risk curve that suits your stomach.
In other words it is still useful to elaborate the basic principles that works in the long run.
We should study Google algorithm not like an algorithm but rather like a person. What are the interesst of this persons? What are the goals? How fast the person to react? How often the person makes mistakes. How soon the person correct the mistakes etc.
Vadim
KW - tried to check density but lost enthusiasm real fast - some pages almost spammy 30-40% some very little density. no KW meta
lots of text- at least 100 words per page - 150? pages.
heavy into the "theme-based" site structure. Where there are several 'main' pages. (homepage is regional, each municipality within the region has it own little 'homepage'. there is a photo gallery and other similar sections which are 'regional' but are heavily crosslinked with the 'municipal' areas AND with each other (just from the natural order that the content provides)
This site has been doubling in traffic every month since last OCT. There was a bit of a lull in April (MAY only increased about 60%) but this month it is on the road to 120% increase over May
Google accounts for approx 60% of traffic.
I have made several changes - restructuring globally several times but have not done much at all since the middle of last month. No matter what I do doesn't seem to affect the rate of growth. I'm tempted to wait to see when it stops so I have a benchmark to see the results of my changes.
Today my site is back with homepage indexed and showing title and description. SERPs showing my site to few of my kw. Got listed in Google Directory.
Anyone has noticed the same? Hoping my site will do a good SERP in couple of days.
I survived and thrived for:
<specific widget>
<specific widget> bats
history of <specific widget>
<specific widget> statistics
But got nailed for:
<general category>
<general category> equipment
history of <general category>
<general category> statistics
--even though the generalist terms are often less competitive than the specific ones.
Most of my backlinage is intentionally highly focused, carefully gathered & relevant to my trade. Despite this I still managed to gather a ton of quality links. I'm thinking maybe I need to now go after link exchanges with really generalist/irrelevant sites.
I haven't added a lot of fresh content or new links in a long while. I got a little burnt out, and I had been doing so very well in the serps that I had become a bit complacent.
I'm kind of cynical too: if not for Google, I think we'd all have, at best, only about 5% of the links we now create. The contest isn't for relevancy, it's just for who can exchange the most links. But I'm just griping, I know.
Despite doing a search from USA for a wholesale product supplier, the serps in my trade are dominated post-bourbon by sites overseas in Indonesia, Thailand & India. Essentially, these sites are there because they assigned employees to gather zillions and zillions of links, most without the slightest relevancy to their keywords/content. Google doesn't seem to care about the quality/relevancy of the links. It just wants more, more, more.
[edited by: ciml at 10:53 am (utc) on June 28, 2005]
[edit reason] Widgetised [/edit]
